Digital vs. Film Quality Now A Non-Issue? - L
Lewis Lang
Some questions...
Only had a chance to skim through the new Pop Photo articles that compared 35mm
film against digital at different MP from 1-9MP and the other article on the
1Ds and its (apparently) clearer images (though minisculey less resoltuion) due
to no film grain clouding up the images. If we concede that 35mm film
resolution and digital resolution of 9-12 MP cameras are about equivalent in
terms of final output at 15x20" on ink jet (which I believe was the output
device), then what advantage is left to 35mm film other than its greater
longevity and future readability and possibly greater future hardware support
down the line/through the years over CDs/other digital storage media/etc.? Does
film still have the edge over non-Foveon (and perhaps Foveon sensors too?) for
color depth/hue discrimination? Should film only be used now for its longer
life and cheaper costs to get into? Is 35mm film vs digital quality now a
non-issue? - should 35mm film be used just for its "look" (ie. you prefer its
rendition over digital for suitable for certain types of subject matter) if
both present more or less equivalent resolution (and possibly color depth)?
Will future DSLRs have better hue discrimination and not just higher MP? This
is not a troll (though there are a lot of speculative questions in this post) -
I'm interested in thoughtful/well reasoned comparisons, suggestions, answers,
more questions, etc.
Any thoughts?
P.S. - Anybody else find it odd that they used possibly grainier 35mm Fujicolor
100 film over less grainy 100 Provia F or even Sensia/Astia (and possibly
Velvia) too?
Regards,
Lewis
Check out my photos at "LEWISVISION":
http://members.aol.com/Lewisvisn/home.htm
Due to spam, I no longer accept emails at this address. Ping me on the ng or go
to my website & click on "LEWISCONTACT" to email. Sorry for the inconvenience
:-).
Alan Browne
post, I am in a vile and lazy mood today.
On storage I confess to confusion. Consumer written CD-ROM's are stated
to be good for about 100 years. I suspect that the real "life" is
probably closer to 200 years. Will the readers be around? When I made
a living writing software I recovered data off of a 9 track tape that
was about 12 years old. With parity and checksum there were 0 errors on
roughly half a meg of data (a huge amount in those days). 9 track tape
was "rated" for 5 years storage. While checksums are not the best way
to verify data, it is reasonable and the data (navigation system flight
recordings) had no blatent errors in it.
Film stored in the right conditions of temperature, humidity and
darkness can beat 100 years easily, but at storage expense. I suppose
if you're rich enough you can setup a trust fund to pay to keep your
images cold, dry and dark until the money runs out. If film could be
stored at very low temperatures and recovered I would intuit 1000's of
years. Try it and tell me how it works out when we meet at Times Square
on 23 Dec 3543.
OTOH if you keep it in sleeves in a sealed metal can at -20C, then I
would bet on a couple hundred years, at least.
Digital renditions of some subjects has an artifical quality to it...the
lack of grain noise in some cases makes things look to smooth. This is
analogous to quantization noise in digital audio sampling, where the
digitized input is run through a filter to "smooth" it out. In some
cases this is not perceptible, in some cases (blue skies being one) it
is flagrant.
In digital there is a discontinuity between adjacent pixels. That is to
say a physical gap in the recorded data (again similar to digital
audio). The chemical "grains" in film that record light also have a
peculiar discontinuity, but it is very random and analog.
If the film "look" is important to you, then it answers itself.
Digital will only get better. There is competition and competition
breeds winners. Winners strive to improve. The most important sense is
vision. Imaging will be important regardless of the media as long as
humans exist. So, Yes, DSLR's will continue to improve, and I suspect
that the day will come when it will be pointles (gasp) to use Velvia.
I do find it unusual that they didn't use a "pro" level film to evaluate
pro level DSLR's.
MO, rambling as usual.
Cheers,
Alan
Tony Spadaro
Sorrry Lewis, but your questions are pretty close to 'how many angels can
dance on the head of a pin?' at this point -- Answers are going to end up in
the philosophy bin.
--
http://chapelhillnoir.com
and partial home of
The Camera-ist's Manifesto
The Links are at
http://home.nc.rr.com/tspadaro/links.html
"Lewis Lang" <cont...@aol.comnospam> wrote in message
news:20030202100004...@mb-fn.aol.com...
Quietlightphoto
Silver based film will always be superior to digital for the following
reason. "Because it is film" Please excuse my nostalgic wanderings........
Phil Stripling
> Only had a chance to skim through the new Pop Photo articles that
> compared 35mm
>SNIP<
> its (apparently) clearer images
>SNIP<
> If we concede that
>SNIP<
> in terms of final output at 15x20" on ink jet
>SNIP<
> what advantage is left to 35mm film other than
>SNIP<
> Will future DSLRs have better hue discrimination and not just higher MP?
>SNIP<
> there are a lot of speculative questions in this post) -
> I'm interested in thoughtful/well reasoned comparisons, suggestions, answers,
> more questions, etc.
Unfortunately, with all your caveats, failures to read, concessions, and so
on, 'thoughtful, well-reasoned answers' can be all over the issues without
ever arriving at any 'real' answers. Fortunately, someone will give you the
answer you're looking for, and you will find your opinion confirmed. :->
--
Philip Stripling | email to the replyto address is presumed
Legal Assistance on the Web | spam and read later. email to philip@
http://www.PhilipStripling.com/ | civex.com is read daily.
Gordon Moat
issue. An RGB sensor is a poor approximation for the CMYK colour space. Also, it
is not a good sensor set-up for true black and white photography.
The MP issue is a quantifiable tool useful for marketing purposes. It is the
easiest point of comparison, and is often stated by the great unknowing masses as
the main comparison. Most people do not lock their cameras in tripods, shoot test
charts, or even keep their lenses clean, making resolution comparison not a good
correlation for average photographic usage.
If all you wanted was to print to a high dot gain inkjet, then it should not make
much difference. With higher quality printing systems, there are more noticeable
differences. Also, while archive quality inks and papers are available, they are
expensive to use for all outputs. The other factor is how much is your time on the
computer worth to get the colour output the way you want it.
I hope that a different chip idea will help the colour issues. I am not convinced
that the Foveon idea is the way forward, but this is only the first example of
that. Perhaps with more research and refinement, the colour issues will be solved.
Drum scanners, and film scanners, are still capable of better colour capture range
that direct digital cameras. Digital scanning backs are getting much closer, but
at great expense. Go to a larger film format, and drum or film scanning, and the
tonality of colours improves more.
Okay, last thoughts. If all someone wanted was to print photos off their under
$300 inkjet, does it really matter what quality the digital image?
Ciao!
Gordon Moat
Alliance Graphique Studio
<http://www.allgstudio.com>
Lewis Lang
>From: Alan Browne alan....@videotron.ca
>Date: Sun, Feb 2, 2003 5:55 PM
>Message-id: <3E3D5B95...@videotron.ca>
Hope your mood gets better Alan. No problem about the top posting. What does MO
stand for, "mouthing off"? I couldn't say anything about CD hardware
reliability (assuming you could still find a computer w/ the right level
operating system in 100 years time to run such a device, plus computers 100
years from now might not even resemble what we have today in the way they
work/function/record and retrieve data!)
You said:
>Digital renditions of some subjects has an artifical quality to it...the
>
>lack of grain noise in some cases makes things look to smooth. This is
>
>analogous to quantization noise in digital audio sampling, where the
>digitized input is run through a filter to "smooth" it out. In some
>cases this is not perceptible, in some cases (blue skies being one) it
>is flagrant.
I am not understanding the point you are making about blue skies being a
flagrant example.. of what, oversmoothing, a lack of grain, something else?
Anywat, here's hoping the rest of your Sunday/mood improves. :-)
Lewis Lang
>Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
>From: "Tony Spadaro" tspa...@ncmaps.rr.com
>Date: Sun, Feb 2, 2003 6:27 PM
>Message-id: <Jqd%9.159287$qq2.9...@twister.southeast.rr.com>
>
>Maybe.
>
>Sorrry Lewis, but your questions are pretty close to 'how many angels can
>dance on the head of a pin?' at this point
Rock, disco or fox trot? (sorry, had to get that in ;-)).
-- Answers are going to end up
>in
>the philosophy bin.
>
True, but I was/am still interested in what others
think/feel/experience/hypothesize about the subject...
>--
>http://chapelhillnoir.com
>and partial home of
>The Camera-ist's Manifesto
>The Links are at
>http://home.nc.rr.com/tspadaro/links.html
Regards,
Lewis Lang
>Date: Sun, Feb 2, 2003 7:07 PM
>Message-id: <20030202140720...@mb-bj.aol.com>
>
>Lewis,
> Silver based film will always be superior to digital for the following
>reason. "Because it is film" Please excuse my nostalgic wanderings........
No problem :-)
I am secretly hoping for a medium that incorporates the best of both media -
the grainlessness (when desired) of digital, the high hue discrimination and
resolution of film, the longevity of film, the non-quality loss of digital and
the price of a Big Mac ;-)
I know, I know, shoot film and buy a scanner (when I'm flush) still one can
dream about what has not yet/might/might not happen - so long as they figure
out a way to lower the choloesterol of this new medium too ;-).
Lewis Lang
>Date: Sun, Feb 2, 2003 7:38 PM
>Message-id: <3q3cn6s...@shell4.tdl.com>
>
>cont...@aol.comnospam (Lewis Lang) writes:
>
>> Only had a chance to skim through the new Pop Photo articles that
>> compared 35mm
>>SNIP<
>> its (apparently) clearer images
>>SNIP<
>Unfortunately, with all your caveats, failures to read, concessions, and
>so
>on, 'thoughtful, well-reasoned answers' can be all over the issues without
>ever arriving at any 'real' answers. Fortunately, someone will give you
>the
>answer you're looking for, and you will find your opinion confirmed. :->
>
>--
>Philip Stripling | email to the replyto address is presumed
>Legal Assistance on the Web | spam and read later. email to philip@
>http://www.PhilipStripling.com/ | civex.com is read daily.
I don't have an opinion formed yet, just questions, Phillip, have read more in
depth than you might suspect w/ regard to the two Pop Photo articles, which is
why I started this thread in the first place, was hoping someone could chime in
w/ some real life experiences/examples as well as hypothesize, but thanks for
your SNIP "contribution".
Tony Spadaro
because I always actually picture them dancing on the point of the pin
instead of the head -- prolly the reason I've never been good at debating
theosophy.
--
http://chapelhillnoir.com
and partial home of
The Camera-ist's Manifesto
The Links are at
http://home.nc.rr.com/tspadaro/links.html
"Lewis Lang" <cont...@aol.comnospam> wrote in message
news:20030202150433...@mb-fn.aol.com...
Alan Browne
Lewis Lang wrote:
>
>
> Hope your mood gets better Alan. No problem about the top posting. What does MO
> stand for, "mouthing off"? I couldn't say anything about CD hardware
> reliability (assuming you could still find a computer w/ the right level
> operating system in 100 years time to run such a device, plus computers 100
> years from now might not even resemble what we have today in the way they
> work/function/record and retrieve data!)
MO: My Opinion, but Mouthing Off fits most of the time.
Yes, we have no bananas: a CD-ROM might be considered an antique
earring 100 years from now.
>
> You said:
>
>
>>Digital renditions of some subjects has an artifical quality to it...the
>>
>>lack of grain noise in some cases makes things look to smooth. This is
>>
>>analogous to quantization noise in digital audio sampling, where the
>>digitized input is run through a filter to "smooth" it out. In some
>>cases this is not perceptible, in some cases (blue skies being one) it
>>is flagrant.
>
>
> I am not understanding the point you are making about blue skies being a
> flagrant example.. of what, oversmoothing, a lack of grain, something else?
Many shots of the sky seem to be a flat mono blue on DSLR. It's
artificial, like it had been colored in by an illustrator.
>
> Anywat, here's hoping the rest of your Sunday/mood improves. :-)
Already better.
Cheers,
Alan
Lewis Lang
>Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
>From: Gordon Moat mo...@attglobal.net
>Date: Sun, Feb 2, 2003 7:53 PM
>Message-id: <3E3D773E...@attglobal.net>
>
>You almost answered your questions on your own. The colour quality is the
>biggest
>issue. An RGB sensor is a poor approximation for the CMYK colour space.
>Also, it
>is not a good sensor set-up for true black and white photography.
>
Or even fine hue discrimination in RGB space photography since w/ multiple
pixel arays (actually multiple sets of different RGB bayer patterned pixels)
covering each point instead of one pixel per color a la Foveon a lot of color
descriptive information which could be there - isn't.
>The MP issue is a quantifiable tool useful for marketing purposes. It is
>the
>easiest point of comparison, and is often stated by the great unknowing
>masses as
>the main comparison. Most people do not lock their cameras in tripods, shoot
>test
>charts, or even keep their lenses clean, making resolution comparison not
>a good
>correlation for average photographic usage.
>
True, but those who do (or use flash will benefit from more quality MP, so MP
is important but its just one (albeit) overhyped aspect of the final digital
image quality. By the time someone gets a sophisticated DSLR they are probably
more mindful of quality concerns anyway, though I do agree that at times even
the most careful user may be forced to use techniques/lenses that will
compromise his/her final technical digital image quality. Since I am more on
the fine art end I am at least always interested in better technical quality
(as well as artistic quality) when/if conditions permit...
>If all you wanted was to print to a high dot gain inkjet, then it should
>not make
>much difference. With higher quality printing systems, there are more
noticeable
>differences. Also, while archive quality inks and papers are available,
>they are
>expensive to use for all outputs. The other factor is how much is your time
>on the
>computer worth to get the colour output the way you want it.
>
Even though I believe the Pop articles mentioned ink jets I am more concerned
w/ chemical prints from digital files. I mentioned the ink jets because that's
how I believe final resulats were compared in the article(s), not because I
wanted to go that route myself for ultimate quality regardless of shooting film
or digital as a capture medium - for me, at least, chemical prints (regardless
of capture source) are where its at.
>I hope that a different chip idea will help the colour issues. I am not
>convinced
>that the Foveon idea is the way forward, but this is only the first example
>of
>that. Perhaps with more research and refinement, the colour issues will
>be solved.
>
A different chip idea? 3 separate B&W chips w/ a beam splitter, perhaps a
different non-Bayer RGB or CMYK (if this is even possible) array pattern?
>Drum scanners, and film scanners, are still capable of better colour capture
>range
>that direct digital cameras. Digital scanning backs are getting much closer,
>but
>at great expense. Go to a larger film format, and drum or film scanning,
>and the
>tonality of colours improves more.
>
Actually they seem about equal to me in different ways between 6x7cm film drum
scanned and the EOS 1Ds's output. Here's a link (if you haven't been there
already):
<A
HREF="http://www.luminous-landscape.com/reviews/shootout.shtml">http://www
.luminous-landscape.com/reviews/shootout.shtml</A>
>Okay, last thoughts. If all someone wanted was to print photos off their
>under
>$300 inkjet, does it really matter what quality the digital image?
>
Yes, if they ever have higher aspirations for their image's quality in addition
to the convenience of a large ink jet output. I have ink jets taped up all over
my office (from when I had a working printer, long story I won't go into) and
find it a convenient jogger for creativity as well as fun to look at, but
again, when it comes down to high quality I prefer the depth and color
discrimination and seemless tonality of chemical prints whether my original
source be 35mm, medium/larger film formats or digtal. When the original source
is of high quality you have the option of making the prints as well done or
crappy (for memory /personal display purposes for me as mentioned above) as you
like. Its nice to have options. When I first shot with a (friend's) DSLR (the
EOS D30) I got a few "keeper" shots that would have been nicer to keep as
higher 5-12 MP instead of 3 MP for when I want larger prints. Genuine
Fractals/Live Picture are/were great tools but you can't add detail that wasn't
there to begin w/ when you res up no matter how much "sharpening" (which I've
heard actually destroys finer detail) you do. Of course I am only me and
everyone's quality acceptability needs and preferences will differ...
>Ciao!
>
>Gordon Moat
>Alliance Graphique Studio
><http://www.allgstudio.com>
Regards,
Annika1980
>Many shots of the sky seem to be a flat mono blue on DSLR. It's
>artificial, like it had been colored in by an illustrator.
Have you looked at a clear blue sky lately?
That's what many of them look like.
Why do people insist on "texture" in snowscapes or blue skies when there isn't
any there to the naked eye?
Lewis Lang
>Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
>From: Alan Browne alan....@videotron.ca
>Date: Mon, Feb 3, 2003 2:29 AM
>Message-id: <3E3DD424...@videotron.ca>
>
>
>
>Lewis Lang wrote:
>>
>>
>> Hope your mood gets better Alan. No problem about the top posting. What
>does MO
>> stand for, "mouthing off"? I couldn't say anything about CD hardware
>> reliability (assuming you could still find a computer w/ the right level
>> operating system in 100 years time to run such a device, plus computers
>100
>> years from now might not even resemble what we have today in the way they
>> work/function/record and retrieve data!)
>
>MO: My Opinion, but Mouthing Off fits most of the time.
>
:-)
>Yes, we have no bananas: a CD-ROM might be considered an antique
>earring 100 years from now.
>
And true dye readability on those discs may last only about 10 years or so, so
I guess in a hundred years time there will be alot of worthless antique
coasters that used to contain readable images for anyone who wants a special
promotion "Earth classic coffee" (like "Coca Cola Classic" TM) on Starbuck's on
Titan (jupiter or Saturn's moon, I forget which right now...).
>>
>> You said:
>>
>>
>>>Digital renditions of some subjects has an artifical quality to it...the
>>>
>>>lack of grain noise in some cases makes things look to smooth. This is
>>>
>>>analogous to quantization noise in digital audio sampling, where the
>>>digitized input is run through a filter to "smooth" it out. In some
>>>cases this is not perceptible, in some cases (blue skies being one) it
>
>>>is flagrant.
>>
>>
>> I am not understanding the point you are making about blue skies being
>a
>> flagrant example.. of what, oversmoothing, a lack of grain, something
>else?
>
>Many shots of the sky seem to be a flat mono blue on DSLR. It's
>artificial, like it had been colored in by an illustrator.
>
Ah, like the old Agfa Ultra 50 (and 100 too?) technique/rendition w/o the hyper
saturation aspect - "you can have any color/hue of red you want so long as you
don't mind it's coming out in only one color, our Ultra saturated red paint
red".
>>
>> Anywat, here's hoping the rest of your Sunday/mood improves. :-)
>
>Already better.
>Cheers,
>Alan
>
:-) :-)
Lewis Lang
>From: "Tony Spadaro" tspa...@ncmaps.rr.com
>Message-id: <vDg%9.19286$hG6.8...@twister.southeast.rr.com>
>
>I've always assumed they would be doing a one footed hop - but that's
>because I always actually picture them dancing on the point of the pin
>instead of the head -- prolly the reason I've never been good at debating
>theosophy.
>
>--
>http://chapelhillnoir.com
>and partial home of
>The Camera-ist's Manifesto
>The Links are at
>http://home.nc.rr.com/tspadaro/links.html
Hi Tony:
I wonder if the "how many angels can dance on the head of a pin quote" comes
from a reference to miniaturist painting somehow? A wonderful surreal metaphor
though as doubtful as a description as the "Its A Wonderful Life" Hollywood
version of angels "earning" their wings every time a bell rings - though what
are called "angels" in Hollywood (producers and/or backers of productions? but
nowadays probably the Hollywood studios themselves) probably do earn enough to
go buy a set of wings everytime the kerching of a cash register goes off at the
movie theater that's showing their movie. Angels themselves, at least when they
appeared on Earth (in disquise or not as men) seemed by inference to probably
be about the same size as men instead of microscopic sparkling "Wonderful World
of Disney" titles sparkling Tinkerbell sized ;-) To get back on topic before
the net police "Smokies" come down on me/us, we'd probably need a 35mm or 35mm
lens oriented DSLR attached to an electron? microscope to be able to image all
those disco dancing/fox troting angels (not the Hollywood kind which might need
a wide angle lens to get their bellies in ;-)) on the head of a pin ;-).
Alan Browne
Really? The sky is usually graduated from lighter to darker tones both
up and down. Film captures this.
Snow has incredible variety, texture and detail.
Next time you look at either, or anything for that matter, turn off the
part of your brain that says "See Dick, See Jane, See Dick Run." and
start looking at detail and the way light interacts with those details.
Cheers,
Alan
Annika1980
>> Why do people insist on "texture" in snowscapes or blue skies when there
>isn't
>> any there to the naked eye?
>
>
>Really? The sky is usually graduated from lighter to darker tones both
>up and down. Film captures this.
As does digital. What digital doesn't do is introduce artificial grain or noise
which may give the illusion of texture.
>Snow has incredible variety, texture and detail.
And digital can capture that too.
My point was that a digital picture of a snowscape like the pic I posted below
can capture everything that the eye can see, or at least everything that film
can capture.
Kinon
you be sure?
"Annika1980" <annik...@aol.com> wrote in message
news:20030203111101...@mb-md.aol.com...
Lewis Lang
>Date: Mon, Feb 3, 2003 2:40 PM
>Message-id: <3E3E7F44...@videotron.ca>
He can't Alan, if he turned off that part of his brain he'd have no part of his
brain left on! ;-)
Annika1980
>> My point was that a digital picture of a snowscape like the pic I posted
>below
>> can capture everything that the eye can see, or at least everything that
>film
>> can capture.
>Unless you took a pic with film of the same scene at the same time, how can
>you be sure?
I have.
http://www.pbase.com/image/11766189
You'll notice in the above example that the film shot (on the left) does appear
to show more texture in the snow. Upon closer examination, however, what looks
like texture is really just film grain or noise.
I plan on having that slide scanned at 4000dpi to see what differences are
there.
Gordon Moat
> Or even fine hue discrimination in RGB space photography since w/ multiple
> pixel arays (actually multiple sets of different RGB bayer patterned pixels)
> covering each point instead of one pixel per color a la Foveon a lot of color
> descriptive information which could be there - isn't.
You just gave me an idea on this. With the RGB colour space, the relationship is
to CMY. With printing, the K (black ink) is added because CMY does not give enough
shadow detail (and other reasons, but I won't go into that now). Correct the
limitations of the RGB (often technically sRGB, or Adobe RGB 1998) colour space,
and printing quality of colour can be improved.
Anyway, you can check this by taking an RGB image, converting to CMYK, then
shutting off the K channel. The results should show anyone a great deal about
colour conversion issues for printing. If you would like, I could post some
samples on a web page. I was recently scanning an old Ektachrome that my Dad
photographed in 1966 in Vietnam. The image was a boy standing in some construction
pipes. The RGB image did not allow for any facial details. Going to LaB mode
showed the L channel very similar to the G channel, but still no detail
improvement. Converting to CMYK, then viewing the K channel only, showed enough
detail that the boys face could be recognized.
So my point is that having a K (black or grey level) correction to add to an RGB
chip might get the proper colour representation. Unfortunately, that is only one
problem. A pure Yellow is very tough to photograph with a digital camera, and any
true Cyan rarely records properly. To make things even worse, all computer
monitors do not display colours accurately when they are close to Yellow, or Cyan.
That limitation puts a wide range of colours out of the accuracy ability of
current chip capture devices. Films with extra emulsion layers do better in this
regard, though scanning can sometimes still be an issue.
> >If all you wanted was to print to a high dot gain inkjet, then it should
> >not make
> >much difference. With higher quality printing systems, there are more
> noticeable
> >differences. Also, while archive quality inks and papers are available,
> >they are
> >expensive to use for all outputs. The other factor is how much is your time
> >on the
> >computer worth to get the colour output the way you want it.
> >
>
> Even though I believe the Pop articles mentioned ink jets I am more concerned
> w/ chemical prints from digital files. I mentioned the ink jets because that's
> how I believe final resulats were compared in the article(s), not because I
> wanted to go that route myself for ultimate quality regardless of shooting film
> or digital as a capture medium - for me, at least, chemical prints (regardless
> of capture source) are where its at.
Perhaps a better test would be Fuji Frontier prints, since that is more common.
Many of the current labs can use these for 35 mm films, 120 films, and digital
files. I know that many consumers have inkjet printers, but publications and fine
art are rarely ever printed that way. To a consumer audience that Pop Photo
addresses, the ink jet comparison may be more relevant, though I think they would
be better off reading some computer oriented magazines instead.
>
>
> >I hope that a different chip idea will help the colour issues. I am not
> >convinced
> >that the Foveon idea is the way forward, but this is only the first example
> >of
> >that. Perhaps with more research and refinement, the colour issues will
> >be solved.
> >
>
> A different chip idea? 3 separate B&W chips w/ a beam splitter, perhaps a
> different non-Bayer RGB or CMYK (if this is even possible) array pattern?
Like 3 chip video cameras? Actually, I wonder why that has not been done yet for
still cameras, since there is a noticeable improvement in colour, contrast,
tonality, and dynamic range in video when compared with 1 chip video cameras.
Some of the new miniDV video cameras make great low cost digital stills cameras,
though not for enlarging images.
Better to base a future camera on a wavelength meter that knew what colour
reflection corresponded to which colour. I think chips attempt to work that way,
but the accuracy could be improved. Some of this might be software, but I think
Bayer pattern limitations are somewhat responsible. Why is it that film scanners
can do so well with colour quality, and direct digital SLRs have so much trouble .
. . ?
>
>
> >Drum scanners, and film scanners, are still capable of better colour capture
> >range
> >that direct digital cameras. Digital scanning backs are getting much closer,
> >but
> >at great expense. Go to a larger film format, and drum or film scanning,
> >and the
> >tonality of colours improves more.
> >
>
> Actually they seem about equal to me in different ways between 6x7cm film drum
> scanned and the EOS 1Ds's output. Here's a link (if you haven't been there
> already):
>
> <A
> HREF="http://www.luminous-landscape.com/reviews/shootout.shtml">http://www
> .luminous-landscape.com/reviews/shootout.shtml</A>
I have downloaded his images, and checked the Histograms in the past. Some of them
show JPEG compression artefacts, though almost all of them show oversharpening. If
this guy knew anything about publication printing, or colour correction, then his
comments would be better.
His articles about digital are relevant for those seeking to get the best outputs
on inkjet printers. Modern printing image setters, and plateless presses, used for
publication, and commercial printed materials, are infinitely better than any
inkjet printer. The dot gain of all inkjet systems limits the quality of the final
images to lower than what is possible on press systems (not newsprint).
The internet is a tough place to find useful examples of valid comparison. Few
people will want to get involved enough in printing technology to understand what
I am stating. I urge you, or anyone else interested in this, to attend a free
Print Week, and get some paper samples with printed images. The difference between
those and inkjets will be easily viewable. Even the exact same digital file will
turn out better from a press output than an inkjet.
The Printing Industry of America runs many Print Weeks in larger cities across the
US. More information about this is at
<http://www.gain.net/PIA_GATF/non_index.html>
>
>
> >Okay, last thoughts. If all someone wanted was to print photos off their
> >under
> >$300 inkjet, does it really matter what quality the digital image?
> >
>
> Yes, if they ever have higher aspirations for their image's quality in addition
> to the convenience of a large ink jet output. I have ink jets taped up all over
> my office (from when I had a working printer, long story I won't go into) and
> find it a convenient jogger for creativity as well as fun to look at, but
> again, when it comes down to high quality I prefer the depth and color
> discrimination and seemless tonality of chemical prints whether my original
> source be 35mm, medium/larger film formats or digtal. When the original source
> is of high quality you have the option of making the prints as well done or
> crappy (for memory /personal display purposes for me as mentioned above) as you
> like. Its nice to have options. When I first shot with a (friend's) DSLR (the
> EOS D30) I got a few "keeper" shots that would have been nicer to keep as
> higher 5-12 MP instead of 3 MP for when I want larger prints. Genuine
> Fractals/Live Picture are/were great tools but you can't add detail that wasn't
> there to begin w/ when you res up no matter how much "sharpening" (which I've
> heard actually destroys finer detail) you do. Of course I am only me and
> everyone's quality acceptability needs and preferences will differ...
Interesting that you have LivePicture. I was thinking of getting a used version of
that, since it works well for large image files. However, you are correct that
upscaling an image cannot add information that was originally captured. To some
people, that extra information can just be a distraction. To many use to Frontier
prints, they may never notice the difference. I notice that when I have scanned
negatives, there is often more information than appeared on a Frontier print . . .
though it is subjective to what level the lack of information degrades an image,
if at all.
Polaroid transfers, and IRIS (or Giclée) prints can be compelling art images in
their final forms. Oil painting is not high resolution, yet that does not degrade
from the appeal of some works. There is no reason why someone cannot make
compelling images using a digital camera, or a film camera. To address your
Subject question, this is no longer an issue in regards to creative limitations.
I take exception to the display of gallery images as "digital photography" . . .
what is really on display is a photographic print (Fuji Crystal Raster Archive
Print, etc.), or an inkjet output (IRIS, Rainbow, Giclée, etc.). If I display a
painting, the description would be oil on canvas, not Windsor & Newton Brushes, W
& N oils, and charcoal . . . . Calling something on display a "digital photo" is
at best a gimmick, and at worst a downright lie . . . the printing method is on
display, not the device!
Sorry for the rant, but just because something is new, does not eliminate all that
came before. The quality is a non-issue for digital now, because a handful of
cameras are not so incredibly bad that you know from the moment you see an image,
that it was done digitally. Those of us in the pre-press, publication, and design
community can tell you what software was used to layout each and every magazine on
the market, and spot a scanned image from a digital only one . . . this is
experience, and should be completely irrelevant for 98 % of the population,
Each of us needs to decide our own quality levels. If selling quality, then every
aspect from image capture to output should be immaculate. Leave "good enough" and
"acceptable" to the consumers, and "selling digital" to the marketing people.
Lewis Lang
>Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
>From: Gordon Moat mo...@attglobal.net
>Date: Tue, Feb 4, 2003 1:34 AM
>Message-id: <3E3F188E...@attglobal.net>
>
>Lewis Lang wrote:
>
>> Or even fine hue discrimination in RGB space photography since w/ multiple
>> pixel arays (actually multiple sets of different RGB bayer patterned pixels)
>> covering each point instead of one pixel per color a la Foveon a lot of
>color
>> descriptive information which could be there - isn't.
>
>You just gave me an idea on this. With the RGB colour space, the relationship
>is
>to CMY. With printing, the K (black ink) is added because CMY does not give
>enough
>shadow detail (and other reasons, but I won't go into that now). Correct
>the
>limitations of the RGB (often technically sRGB, or Adobe RGB 1998) colour
>space,
>and printing quality of colour can be improved.
>
Interesting, how do you propose to correct these limitations. Actually, I just
had an idea myself which might be too retro (and it may not work, but I'll say
it anyways)... I wonder if (control your laughing) digi camera went back to 3
tubes instead of CCD's as sensors to have a more even/finer representation of
colors. Another tactic might be to (w/ either chip or tube) to have a beam
splitter which works w/ more than 3 chips/tubes much in the same way film has
extra color layers to deal w/ hard to render colors. There is no reason, other
than cost tothe manufacturer and what the market place will bare, why digital
cameras should be limited to no more thanone chip or tube, or no more than 3
chips/tubes. I realize that older tube video cameras had "smear"/"cometing of
bright light sources and possibly other problems. But I think that there needs
to be more research (and products) done to expand color space and, more
importantly, hue discrimination. Where digital still photography is sorely
lacking (compared to film) is/seems to be in its subtlety and depth, not just
saturation - film images seem to have both a tonal and color depth that even
some of the best digital lacks. Digital looks flat/plasticky to me (despite its
"clarity"/lack of grain and higher and higher resolution) that reminds me of
its "surfaceness"/superficiality. This problemis just aggravated by large blow
ups, which, though lacking grain, also lack detail, and tonality, beyond any
reaching/seeing the limits of resolution/pixelization. A similar thing happens
when you blow up film too much, but at least you have the "balm" of a nice
grain pattern to hold the eye/mind's interest. And this grain pattern is
inherent/integral to the structure of the image itself, not added on
(intentionally added on or not) digital chip/CCD noise, which also looks like
superficial colorized video snow to me. Recently I had the chance to see a
former United Nations children photographer who now (I'm assuming) does mostly
if not all his freelance shooting w/ an Olympus E-20 (and possibly the E-10).
He was at the jacob Javitz center Photo Expo show signing his posters made w/
the Olympus digi camera. From a distance both the poster and huger photographic
prints (20x24" or 24x30" or 30x40"?) look great, up close you can see what
amounts to either digital artefacting and noise where smooth tonality should be
(like standing way too close to a tv when you're watching it). This "noise" due
to lack of info/artefacts/etc. completely destroyed the tonality of the skin
(shots of kids/people deserve to have wonderful skin tones) and I thought
detracted from the image. I guess its a matter of how much quality you are
willing to let go of for the sake of an all in one (one lens) digital ZLR
solution convenience. When blowing up any image film or digital, but especially
w/ digital, the more you blow an image up, its like adding more and more water
to lemonade until eventually you lose the "lemonadeness" of the drink and it
tastes like water w/ a touch of lemon in it. In output, this is especially
evident in ink jet outputs,as the more you look at the lighter areas and the
closer you get the more the image depth begins to break down as your eye
latches onto the dot spray pattern and the spaces between (no matter how fine
in pico liters or stochastic printing this dot pattern is). Chemical
photographic prints and some dye sub prints (at least to my eye) appear more
continuous toned.
>Anyway, you can check this by taking an RGB image, converting to CMYK, then
>shutting off the K channel. The results should show anyone a great deal
>about
>colour conversion issues for printing. If you would like, I could post some
>samples on a web page.
Thanks, please do and let me know the URL(s). :-)
I was recently scanning an old Ektachrome that my
>Dad
>photographed in 1966 in Vietnam. The image was a boy standing in some
construction
>pipes. The RGB image did not allow for any facial details. Going to LaB
>mode
>showed the L channel very similar to the G channel, but still no detail
>improvement. Converting to CMYK, then viewing the K channel only, showed
>enough
>detail that the boys face could be recognized.
>
Like squeezing the lemon harder to get more lemon into your lemonade ;-).
>So my point is that having a K (black or grey level) correction to add to
>an RGB
>chip might get the proper colour representation.
Why not a 4 chipped (or tubed) digital still camera w/ one chip for C, one for
M, one for Y and one for K - this to me would make more sense for those already
shooting for output in paper publication (magazines, newspapers, etc.)
Unfortunately, that is
>only one
>problem. A pure Yellow is very tough to photograph with a digital camera,
>and any
>true Cyan rarely records properly. To make things even worse, all computer
>monitors do not display colours accurately when they are close to Yellow,
>or Cyan.
Pardon my ignorance but why not also have special monitors (if possible) that
have C, M, Y, and K (spearate luminance for black and white gray levels) pixels
to alleviate this problem.
>That limitation puts a wide range of colours out of the accuracy ability
>of
>current chip capture devices. Films with extra emulsion layers do better
>in this
>regard, though scanning can sometimes still be an issue.
>
Home scanners density ranges as well as hue discrimination abilities could also
stand some improving. I recently had an URL, now I can't find it for the life
of me, that compared scans of various formats from 35mm up to 4x5" scanned on
both desk top and drum? scans (but not drum scans on all formats). Anyway it
showed that 35mm (besides how fuzzy/lack of detail it is/scans really requires
6-8000 ppi to get all of the info out of it, 4000 (mpost home scanner's upper
limits) doesn't really cut the mustard. Wish I still had/could find the URL...
You wouldn't happen to know which web page I'm talking about and could post the
URL? Thanks.
>> >If all you wanted was to print to a high dot gain inkjet, then it should
>> >not make
>> >much difference. With higher quality printing systems, there are more
>> noticeable
>> >differences. Also, while archive quality inks and papers are available,
>> >they are
>> >expensive to use for all outputs. The other factor is how much is your
>time
>> >on the
>> >computer worth to get the colour output the way you want it.
>> >
>>
>> Even though I believe the Pop articles mentioned ink jets I am more
concerned
>> w/ chemical prints from digital files. I mentioned the ink jets because
>that's
>> how I believe final resulats were compared in the article(s), not because
>I
>> wanted to go that route myself for ultimate quality regardless of shooting
>film
>> or digital as a capture medium - for me, at least, chemical prints
(regardless
>> of capture source) are where its at.
>
>Perhaps a better test would be Fuji Frontier prints, since that is more
>common.
Indeed (and better quality of at least the illusion of continuous tonality).
>Many of the current labs can use these for 35 mm films, 120 films, and digital
>files. I know that many consumers have inkjet printers, but publications
>and fine
>art are rarely ever printed that way. To a consumer audience that Pop Photo
>addresses, the ink jet comparison may be more relevant, though I think they
>would
>be better off reading some computer oriented magazines instead.
>
>>
>>
>> >I hope that a different chip idea will help the colour issues. I am not
>> >convinced
>> >that the Foveon idea is the way forward, but this is only the first example
>> >of
>> >that. Perhaps with more research and refinement, the colour issues will
>> >be solved.
>> >
>>
>> A different chip idea? 3 separate B&W chips w/ a beam splitter, perhaps
>a
>> different non-Bayer RGB or CMYK (if this is even possible) array pattern?
>
>Like 3 chip video cameras? Actually, I wonder why that has not been done
>yet for
>still cameras, since there is a noticeable improvement in colour, contrast,
>tonality, and dynamic range in video when compared with 1 chip video cameras.
>Some of the new miniDV video cameras make great low cost digital stills
>cameras,
>though not for enlarging images.
>
Market demand. The market demands and is willing to pay for (up to a point)
more resolution. Color depth and hue discrimination may be
"backburner"/additional selling points but not a priority. I hope Foveon/etc.
changes this market need/demand so more life-like, dare I say, "film-like"
color depth could be achieved. The major P&S digital market is in love w/ MP
and not bit depth and other important factors - so long as Auntie Matilda can
e-mail images w/ her 3MP P&S she's happy (until her files become
unreadable/unusable 10 years or less down the road, but that's another issue
which is not getting enough attention either in the film world or the digital
media world...). :-(
>Better to base a future camera on a wavelength meter that knew what colour
>reflection corresponded to which colour.
Tubes! ...or their equivalents (redesigned for the 2000's w/o the smear/other
degradation factors). :-)
I think chips attempt to work that
>way,
>but the accuracy could be improved. Some of this might be software, but
>I think
>Bayer pattern limitations are somewhat responsible. Why is it that film
>scanners
>can do so well with colour quality, and direct digital SLRs have so much
>trouble .
>. . ?
Too bad digi camera chips can't be more like film scanners but recorded in one
pass. 3 or 4 chip/tube digi still cameras (and other pattern arrays and
processing algorithms) may help bridge the gap between "film quality" and "digi
quality" - too bad there isn't a medium that incorporates the best of both
worlds/mediums, the quality (high resolution and color depth and long life) of
film and the reusability and ease of reproducibility w/o degradation through
many copies/generations of digital...
>
>>
>>
>> >Drum scanners, and film scanners, are still capable of better colour
>capture
>> >range
>> >that direct digital cameras. Digital scanning backs are getting much
>closer,
>> >but
>> >at great expense. Go to a larger film format, and drum or film scanning,
>> >and the
>> >tonality of colours improves more.
>> >
>>
Of course. Too bad I can't put my people subjectsdirectly on a drum scanner
like some people do w/ xerography though... ;-)
>> Actually they seem about equal to me in different ways between 6x7cm film
>drum
>> scanned and the EOS 1Ds's output. Here's a link (if you haven't been there
>> already):
>>
>> <A
>> HREF="http://www.luminous-landscape.com/reviews/shootout.shtml">http://www
>> .luminous-landscape.com/reviews/shootout.shtml
>
>I have downloaded his images, and checked the Histograms in the past. Some
>of them
>show JPEG compression artefacts, though almost all of them show
oversharpening.
No surprise there. Sharpening is like a chef knowing how much salt (or other
spice) to add, if any to a meal/dish. It takes sense and sensitivity...
>If
>this guy knew anything about publication printing, or colour correction,
>then his
>comments would be better.
>
I just take it worth a grain or pixel of salt and just smile to myself- it will
be along time before I can afford even a second or thirdhand 1Ds. Film still
makes better economic sense to me in the short and long run for my needs,
though I think that the digital darkroom is where its at for color and final
output at least so long as I'm outputting onto a photo chemical print like
Crystal Archive/Frontier prints.
>His articles about digital are relevant for those seeking to get the best
>outputs
>on inkjet printers. Modern printing image setters, and plateless presses,
>used for
>publication, and commercial printed materials, are infinitely better than
>any
>inkjet printer. The dot gain of all inkjet systems limits the quality of
>the final
>images to lower than what is possible on press systems (not newsprint).
>
Perhaps the reason why he oversharpens his images (consciously or
unconsciously) to make up for this ink jet dot gain - or perhaps he's just
addicted to oversharpening period? ;-)
>The internet is a tough place to find useful examples of valid comparison.
>Few
>people will want to get involved enough in printing technology to understand
>what
>I am stating. I urge you, or anyone else interested in this, to attend a
>free
>Print Week, and get some paper samples with printed images. The difference
>between
>those and inkjets will be easily viewable. Even the exact same digital file
>will
>turn out better from a press output than an inkjet.
>
That makes sense.
>The Printing Industry of America runs many Print Weeks in larger cities
>across the
>US. More information about this is at
><http://www.gain.net/PIA_GATF/non_index.html>
>
Thanks for the link.
This is its forte. Cloning/rubber stamping and other functions may be done
better in Photoshop, though (I'm guessing here since I don't own a current
version of Photoshop and have just used PS on some one else's computers in the
distant past...
However, you are correct
>that
>upscaling an image cannot add information that was originally captured.
>To some
>people, that extra information can just be a distraction.
Depends upon both the type of image it is and what your preferences are for.
More abstract and/or more soft focus images may get along fine w/o the extra
detail, though some landscape photographers may cry if they lose one iota or
one twig unresolved from their images ;-).
To many use to
>Frontier
>prints, they may never notice the difference. I notice that when I have
>scanned
>negatives, there is often more information than appeared on a Frontier print
>. . .
>though it is subjective to what level the lack of information degrades an
>image,
>if at all.
>
I'm surprised, I thought Frontier scans in images at 303.5? ppi which should be
more than enough for good detail as 300 ppi is considered normal "good" and 240
ppi acceptable for most ink jet printers and 200 ppi (according to their most
recent articles) acceptable for Pop Photo's minimum acceptable amature
standards. If I had enough money/storage space I would scan at 400 ppi just for
that extra bit of detail, but that's just me...
>Polaroid transfers, and IRIS (or Giclée) prints can be compelling art images
>in
>their final forms. Oil painting is not high resolution, yet that does not
>degrade
>from the appeal of some works.
I know, but in at least some oils you want to be able to see and get lost in
the individual brush strokes, w/ most Giclee/etc. done from high res digi or
film capture usually (unless a special effect/(s) filtering is done/desired)
you want to focus your intention on the images themselves and the
"brushstrokes" (pixelization, grain, ink spatter pattern) can just get in the
way and make you too conscious you are just looking at an image instead of
being drawn into the image. I know in some images, the process doesn't get in
the way but is the point of the picture (like in the abstract expressionist
drip paintings of Pollack, the simulated Ben-day dots of Lichtenstein, and even
some of Rembrandt's later styled portraits when seen up close, etc.), but this
is a different aim/effect.
There is no reason why someone cannot make
>compelling images using a digital camera, or a film camera.
Right, whichever medium works best for your imagery...
To address your
>Subject question, this is no longer an issue in regards to creative
limitations.
>
>I take exception to the display of gallery images as "digital photography"
>. . .
>what is really on display is a photographic print (Fuji Crystal Raster Archive
>Print, etc.), or an inkjet output (IRIS, Rainbow, Giclée, etc.). If I display
>a
>painting, the description would be oil on canvas, not Windsor & Newton
Brushes,
>W
>& N oils, and charcoal . . . . Calling something on display a "digital photo"
>is
>at best a gimmick, and at worst a downright lie . . . the printing method
>is on
>display, not the device!
>
People have a (marketing driven) hyped up love affair w/ the device. Art is
art.
>Sorry for the rant, but just because something is new, does not eliminate
>all that
>came before.
"New and _Improved_!" :-)
The quality is a non-issue for digital now, because a handful
>of
>cameras are not so incredibly bad that you know from the moment you see
>an image,
>that it was done digitally. Those of us in the pre-press, publication, and
>design
>community can tell you what software was used to layout each and every
magazine
>on
>the market, and spot a scanned image from a digital only one . . . this
>is
>experience, and should be completely irrelevant for 98 % of the population,
>
>Each of us needs to decide our own quality levels. If selling quality, then
>every
>aspect from image capture to output should be immaculate. Leave "good enough"
>and
>"acceptable" to the consumers, and "selling digital" to the marketing people.
>
>Ciao!
>
>Gordon Moat
>Alliance Graphique Studio
><http://www.allgstudio.com>
>
Amen :-)
© 2003 Lewis Lang
All (New and Improved!) Rights (and non-Bayer patterned pixels and film grains)
Reserved
Void where prohibited, not connected w/ any other advertising offers, see store
flyer for details. Keep America green, hug an unflushed toilet today ;-) :-)
Kinon
behind the pin, and you'll see.
"Annika1980" <annik...@aol.com> wrote in message
news:20030203195053...@mb-ba.aol.com...
Simon Stanmore
Simon
Gordon Moat
> Hi Gordon:
Guten morgen Lewis,
>
>
> >With the RGB colour space, the relationship
> >is
> >to CMY. With printing, the K (black ink) is added because CMY does not give
> >enough
> >shadow detail (and other reasons, but I won't go into that now). Correct
> >the
> >limitations of the RGB (often technically sRGB, or Adobe RGB 1998) colour
> >space,
> >and printing quality of colour can be improved.
> >
>
> Interesting, how do you propose to correct these limitations. Actually, I just
> had an idea myself which might be too retro (and it may not work, but I'll say
> it anyways)... I wonder if (control your laughing) digi camera went back to 3
> tubes instead of CCD's as sensors to have a more even/finer representation of
> colors. Another tactic might be to (w/ either chip or tube) to have a beam
> splitter which works w/ more than 3 chips/tubes much in the same way film has
> extra color layers to deal w/ hard to render colors. There is no reason, other
> than cost tothe manufacturer and what the market place will bare, why digital
> cameras should be limited to no more thanone chip or tube, or no more than 3
> chips/tubes.
I think the 3 CCD (or CMOS, or whatever) idea would help a bit, but still not the
tonal range of film. There are quite small 3 chip video cameras already, so I do
not see why some company has not yet packaged a 3 chip stills camera, other than
perhaps too high costs. The market may be quite limited for high priced digital
SLRs, so perhaps little incentive to make something better. The other quality
limit is that most of the pro use of these is for newspapers, who have some of the
worst printing, and lowest requirements of all professional publishing, so again,
not much reason to improve things.
> I realize that older tube video cameras had "smear"/"cometing of
> bright light sources and possibly other problems. But I think that there needs
> to be more research (and products) done to expand color space and, more
> importantly, hue discrimination. Where digital still photography is sorely
> lacking (compared to film) is/seems to be in its subtlety and depth, not just
> saturation - film images seem to have both a tonal and color depth that even
> some of the best digital lacks. Digital looks flat/plasticky to me (despite its
> "clarity"/lack of grain and higher and higher resolution) that reminds me of
> its "surfaceness"/superficiality.
The flatness reminds me of one of our earlier discussions. I think you remember
when I mentioned setting up lighting for video. The same considerations of
avoiding high lighting contrast ranges applies to digital stills. One needs to
avoid blowing out highlights just to get shadow detail, or loosing shadow detail
to get good highlight area definition. Basically, the even lighting can look flat,
but trying to push farther can loose detail aspects of shots that may otherwise be
interesting.
> This problemis just aggravated by large blow
> ups, which, though lacking grain, also lack detail, and tonality, beyond any
> reaching/seeing the limits of resolution/pixelization. A similar thing happens
> when you blow up film too much, but at least you have the "balm" of a nice
> grain pattern to hold the eye/mind's interest. And this grain pattern is
> inherent/integral to the structure of the image itself, not added on
> (intentionally added on or not) digital chip/CCD noise, which also looks like
> superficial colorized video snow to me.
I have seen quite a bit about grain, though usually the pundits complain about it
with scanned images. Almost all film scanners have noise in one or more colour
channels. There are ways around this in post processing, but nothing that is
written in any books or manuals. It took me a few years of working in PhotoShop to
realize what was happening, and how to improve the scanned images.
True grain is often triangle or hexagonal microscopic crystals. These are somewhat
random patterns, and there is overlap of many of these for each colour sensitive
layer. They also can appear to clump, though these become much more visible in
scans than on prints. I have also seen microscope images of film base that showed
imperfections in the materials. These small imperfections looked like small
bubbles under the microscope, yet would show up quite well in clear areas, or
areas of large colour. This is another source for the "grain" look in some scanned
photos.
So this brings me to another observation. Chips capture to square or rectangular
pixels. The computer monitors also display in regular square (rectangular) pixels.
Anytime there are edges that do not fit into these shapes, there will be some
error in recording. Many digital cameras already correct for this in built-in
software, though not all scanners do this. Currently, since many digital camera
users really do not understand PhotoShop that well, the corrections made within a
digital camera can give better images for some users than they could do with film.
This is not a quality issue, but one of allowing more consistent results. I have
seen quite a few people improve their photography using a digital camera, then go
back to shooting a film camera because they wanted better colour quality prints
(or B/W in a few cases). In this way, digital photography experience can help
improve film camera photography.
> Recently I had the chance to see a
> former United Nations children photographer who now (I'm assuming) does mostly
> if not all his freelance shooting w/ an Olympus E-20 (and possibly the E-10).
> He was at the jacob Javitz center Photo Expo show signing his posters made w/
> the Olympus digi camera. From a distance both the poster and huger photographic
> prints (20x24" or 24x30" or 30x40"?) look great, up close you can see what
> amounts to either digital artefacting and noise where smooth tonality should be
> (like standing way too close to a tv when you're watching it). This "noise" due
> to lack of info/artefacts/etc. completely destroyed the tonality of the skin
> (shots of kids/people deserve to have wonderful skin tones) and I thought
> detracted from the image. I guess its a matter of how much quality you are
> willing to let go of for the sake of an all in one (one lens) digital ZLR
> solution convenience. When blowing up any image film or digital, but especially
> w/ digital, the more you blow an image up, its like adding more and more water
> to lemonade until eventually you lose the "lemonadeness" of the drink and it
> tastes like water w/ a touch of lemon in it. In output, this is especially
> evident in ink jet outputs,as the more you look at the lighter areas and the
> closer you get the more the image depth begins to break down as your eye
> latches onto the dot spray pattern and the spaces between (no matter how fine
> in pico liters or stochastic printing this dot pattern is). Chemical
> photographic prints and some dye sub prints (at least to my eye) appear more
> continuous toned.
Interesting, though I think the images may be better judged in their own context.
With painting, the detail information and resolution are not that high, yet the
minds eye fills in the details. Why does it seem that with photography, we want
more information, and we notice lack of information more?
>
>
> >Anyway, you can check this by taking an RGB image, converting to CMYK, then
> >shutting off the K channel. The results should show anyone a great deal
> >about
> >colour conversion issues for printing. If you would like, I could post some
> >samples on a web page.
>
> Thanks, please do and let me know the URL(s). :-)
I will start working on that one this afternoon. I just figured out a way to put
together the channel images. It will be posted on a new thread when I have it
uploaded to my site, since I think some other people may want to view it as well.
>
> >So my point is that having a K (black or grey level) correction to add to
> >an RGB
> >chip might get the proper colour representation.
>
> Why not a 4 chipped (or tubed) digital still camera w/ one chip for C, one for
> M, one for Y and one for K - this to me would make more sense for those already
> shooting for output in paper publication (magazines, newspapers, etc.)
One problem is that a 4 channel image takes up more storage space. It also takes
longer to store each image, making consecutive shots much slower. Printing has
already moved towards HiFi and Hexachrome, which are six colour processing. Even
though most colour printing is still CMYK, the other options are becoming more
affordable and more common.
All these are still compromises. Some films add extra colour layers to compensate
towards some areas of the visual spectrum, so they are already more complex than
RGB. To do this with digital SLRs would be costly. Also, remember that film is
well developed evolutionary product. The advent of T Grain, and similar chemical
changes was a big step, and may have been though by many to be the last great film
innovation. However, now we are getting newer films that have even greater colour
range capabilities, and even lower grain indexes.
Meanwhile, the historically newer, and less well developed digital chip ideas have
experienced few changes. We are still stuck with the Bayer pattern idea, though
the Foveon at least attempts to point another direction. Most consumers are buying
the cheap digitals, and less than 11% are printing anything, so again, where is
the incentive to improve anything?
>
>
> Unfortunately, that is
> >only one
> >problem. A pure Yellow is very tough to photograph with a digital camera,
> >and any
> >true Cyan rarely records properly. To make things even worse, all computer
> >monitors do not display colours accurately when they are close to Yellow,
> >or Cyan.
>
> Pardon my ignorance but why not also have special monitors (if possible) that
> have C, M, Y, and K (spearate luminance for black and white gray levels) pixels
> to alleviate this problem.
Monitors function in the RGB colour space. Some display a wider range than others,
but true Cyan and true Yellow are just beyond the colour space capability of any
monitor. The way to get around this while working in PhotoShop is to sample with
the colour picker to ensure that those areas not properly displayed should print
correctly. You can also look at the channels individually, if you know what to
look at, and how they should appear. The only real way to do that is to get a
proof. After doing enough of these, one gets a feel for how something will turn
out by looking at the image on a monitor, though a printed proof is still a good
idea. You want real complexity, there are differences in proofing systems as well,
with some better than others, though only a press proof is truly accurate.
Literally every step is a best guess, with the skill of the preparer of the image
often being the limiting factor.
People who sit at home and have finally figured out how to get good inkjet prints
often get baffled why something press printed looks different. A really good
service bureau may adjust the digital file to match the inkjet, but usually at
great expense. When the minilabs start taking more digital files, I would hate to
be that person behind the counter trying to explain to someone why the printed
images do not look like what the customer expected.
>
>
> >That limitation puts a wide range of colours out of the accuracy ability
> >of
> >current chip capture devices. Films with extra emulsion layers do better
> >in this
> >regard, though scanning can sometimes still be an issue.
> >
>
> Home scanners density ranges as well as hue discrimination abilities could also
> stand some improving. I recently had an URL, now I can't find it for the life
> of me, that compared scans of various formats from 35mm up to 4x5" scanned on
> both desk top and drum? scans (but not drum scans on all formats). Anyway it
> showed that 35mm (besides how fuzzy/lack of detail it is/scans really requires
> 6-8000 ppi to get all of the info out of it, 4000 (mpost home scanner's upper
> limits) doesn't really cut the mustard. Wish I still had/could find the URL...
> You wouldn't happen to know which web page I'm talking about and could post the
> URL? Thanks.
I think I remember that one, though there are several sites that address this same
issue. Try this one, and see if that is what you meant (large download):
<http://www.users.qwest.net/~rnclark/scandetail.htm>
> Market demand. The market demands and is willing to pay for (up to a point)
> more resolution. Color depth and hue discrimination may be
> "backburner"/additional selling points but not a priority. I hope Foveon/etc.
> changes this market need/demand so more life-like, dare I say, "film-like"
> color depth could be achieved. The major P&S digital market is in love w/ MP
> and not bit depth and other important factors - so long as Auntie Matilda can
> e-mail images w/ her 3MP P&S she's happy (until her files become
> unreadable/unusable 10 years or less down the road, but that's another issue
> which is not getting enough attention either in the film world or the digital
> media world...). :-(
I think it may. The video camcorder market has progressed quickly to better colour
quality, and even a "film look" attempt. After the video market addressed the
"broadcast quality" (don't you enjoy these meaningless marketing terms ;-0 ),
and met the resolution of television, it only took a few years to improve the
colour quality. I think we are just two or three years away from better colour
quality in digital stills cameras . . . I hope.
>
>
> >Better to base a future camera on a wavelength meter that knew what colour
> >reflection corresponded to which colour.
>
> Tubes! ...or their equivalents (redesigned for the 2000's w/o the smear/other
> degradation factors). :-)
Yeah . . . it may take a new spin on an old idea. Okay, how about this for weird,
a moving strip of sensitized material that then goes through a line scanner,
within the camera body. The sensitized material holds the colour information only
for a very short period of time, during which is passes a recording areas that
discharges the photons, and reads the emitted wavelengths. Then that information
is stored as a digital file. The moving strip could be a continues loop within the
camera body that returns to a neutral state upon discharge. This could be similar
to the way some radiation monitoring equipment works, though perhaps some of the
patents that Polaroid has may be more applicable for further development.
It constantly amazes me how many people always sharpen every image file, and often
by the exact same amount. If Adobe had intended it that way, there would be no
adjustments to that filter, or there would be no filter, it would be automatic.
The reality is that selective sharpening can help with some printing, though it
can create real problems with other printing. When I mentioned the Carbro process,
this is one type of printing that is such high resolution that it would be very
difficult to use any sharpening, and still avoid artefact flaws. Most people
sharpen until they can see it on a monitor screen, but often that is way too much.
The only proper way to do this is by comparing printed samples, though that takes
time, experience, and expense.
>
>
> >The internet is a tough place to find useful examples of valid comparison.
> >Few
> >people will want to get involved enough in printing technology to understand
> >what
> >I am stating. I urge you, or anyone else interested in this, to attend a
> >free
> >Print Week, and get some paper samples with printed images. The difference
> >between
> >those and inkjets will be easily viewable. Even the exact same digital file
> >will
> >turn out better from a press output than an inkjet.
> >
>
> That makes sense.
>
> >The Printing Industry of America runs many Print Weeks in larger cities
> >across the
> >US. More information about this is at
> ><http://www.gain.net/PIA_GATF/non_index.html>
> >
>
> Thanks for the link.
>
> I'm surprised, I thought Frontier scans in images at 303.5? ppi which should be
> more than enough for good detail as 300 ppi is considered normal "good" and 240
> ppi acceptable for most ink jet printers and 200 ppi (according to their most
> recent articles) acceptable for Pop Photo's minimum acceptable amature
> standards. If I had enough money/storage space I would scan at 400 ppi just for
> that extra bit of detail, but that's just me...
The Frontier, and other systems like that, match their printed output. Compare
this to an image setter that outputs 2400 (or 2540) ppi separations, then can
overlap those individual outputs. An example is a 2400 Cyan output, overlaid with
a 2400 Magenta output, a 2400 Yellow output, than finally a 2400 Black (K) output.
Since it is also possible with current papers to get 300 % ink coverage (or
better), the theoretical resolution is 2400 x 3, or 7200, though dot gain still
brings that down a bit. Anyway, I am oversimplifying something that is actually
much more complex in operation. Frontier type systems only need to match their
output, so over sampling would just slow them down. They are better than many
inkjets, and their colour quality and accuracy fit a fairly large printed colour
gamut. Pop Photo would do well to talk to their print shop, or have their lead
graphic designer explain things better. I think they do know these issues, but why
confuse the average reader, and average inkjet owner . . . .
>
>
> >Polaroid transfers, and IRIS (or Giclée) prints can be compelling art images
> >in
> >their final forms. Oil painting is not high resolution, yet that does not
> >degrade
> >from the appeal of some works.
>
> I know, but in at least some oils you want to be able to see and get lost in
> the individual brush strokes, w/ most Giclee/etc. done from high res digi or
> film capture usually (unless a special effect/(s) filtering is done/desired)
> you want to focus your intention on the images themselves and the
> "brushstrokes" (pixelization, grain, ink spatter pattern) can just get in the
> way and make you too conscious you are just looking at an image instead of
> being drawn into the image. I know in some images, the process doesn't get in
> the way but is the point of the picture (like in the abstract expressionist
> drip paintings of Pollack, the simulated Ben-day dots of Lichtenstein, and even
> some of Rembrandt's later styled portraits when seen up close, etc.), but this
> is a different aim/effect.
Yeah . . . too many digital images are just like a bad Seurat painting ;-)
Bling, bling . . . fo dat pinging thang!
Patrick L.
"Lewis Lang" <cont...@aol.comnospam> wrote in message
Dear LewisLewis,
Enjoyed your photos on your web site, but I couldn't help noticing
your infatuation with your own name. :)
PatickPatrick.
Lewis Lang
>Date: Tue, Feb 4, 2003 10:51 PM
>Message-id: <PtX%9.7792$ek4.7...@newsread2.prod.itd.earthlink.net>
Thanks Patrick, whose name should I be "infatuated" with? Its just a way fo
advertising the fact that the real subject are not landscapes or somesuch but
my "subject" is my way of seeing. There's always room on the internet for a
"PATRICKVISION" ;-) website should you want to advertise your own way of seeing
;-).
Regards,
Lewis
Check out my photos at "LEWISVISION":
http://members.aol.com/Lewisvisn/home.htm
Lewis Lang
>Date: Tue, Feb 4, 2003 7:43 PM
>Message-id: <3E4017C5...@attglobal.net>
>
>Lewis Lang wrote:
>
>> Hi Gordon:
>
>Guten morgen Lewis,
>
Sodium glutamate me fine feathered droog ;-),
Makes sense, not that I like it/the lack of a market for such a device, but
makes economic sense anyways until digi sensor chips become as cheap as potato
chips ;-).
Not only the clarity but the flatness of the lighting range and less
tonality/hue discrimination of digital reminds me of high resolution video - a
very "plasticky" look - "plasticky" not in the sense of its old meaning of
three-dimensionality but in the opposite meaning of like clear shiny plastic,
all surface, no depth...
>> This problemis just aggravated by large blow
>> ups, which, though lacking grain, also lack detail, and tonality, beyond
>any
>> reaching/seeing the limits of resolution/pixelization. A similar thing
>happens
>> when you blow up film too much, but at least you have the "balm" of a
>nice
>> grain pattern to hold the eye/mind's interest. And this grain pattern
>is
>> inherent/integral to the structure of the image itself, not added on
>> (intentionally added on or not) digital chip/CCD noise, which also looks
>like
>> superficial colorized video snow to me.
>
>I have seen quite a bit about grain, though usually the pundits complain
>about it
>with scanned images. Almost all film scanners have noise in one or more
>colour
>channels. There are ways around this in post processing, but nothing that
>is
>written in any books or manuals. It took me a few years of working in
PhotoShop
>to
>realize what was happening, and how to improve the scanned images.
>
Any noise reducing techniques for scanned color film images you'd care to
share? :-)
>True grain is often triangle or hexagonal microscopic crystals. These are
>somewhat
>random patterns, and there is overlap of many of these for each colour
sensitive
>layer. They also can appear to clump, though these become much more visible
>in
>scans than on prints. I have also seen microscope images of film base that
>showed
>imperfections in the materials. These small imperfections looked like small
>bubbles under the microscope, yet would show up quite well in clear areas,
>or
>areas of large colour. This is another source for the "grain" look in some
>scanned
>photos.
>
I believe I may have seen these "bubbles" before though I am surprised that
scanning does actually pick up these imperfections...
>So this brings me to another observation. Chips capture to square or
rectangular
>pixels. The computer monitors also display in regular square (rectangular)
>pixels.
>Anytime there are edges that do not fit into these shapes, there will be
>some
>error in recording. Many digital cameras already correct for this in built-in
>software, though not all scanners do this. Currently, since many digital
>camera
>users really do not understand PhotoShop that well, the corrections made
>within a
>digital camera can give better images for some users than they could do
>with film.
Better in what/which sense(s)?
>This is not a quality issue, but one of allowing more consistent results.
>I have
>seen quite a few people improve their photography using a digital camera,
>then go
>back to shooting a film camera because they wanted better colour quality
>prints
>(or B/W in a few cases). In this way, digital photography experience can
>help
>improve film camera photography.
>
Improve their photography in what way(s), composition and/or exposure due to
the availability of instant playback on the cameras' LCD screen on back,
improvement insome other aspect(s) of their photography - which aspects of
their photography get improved?
Think back to the past, way in the past w/ Daugerrotypes and other 19th and
20th century processes that usually used large format and/or the vfinal viewed
image was recorded directly onto the bitumen? or other surface. Weston/etc. w/
their 8x10" contact prints, etc. Even go all the way back to the camera obscura
several hundred years before there was any medium to record light's images.
We've come to expect from the word/process of "photographic" that means highest
quality which equals "holding up a mirror to nature" which means the highest
detail possible recorded. Other effects are considered either "painterly" (soft
focus) or just plain bad technique (to the sharpness obsessed). Among other
19th century and 20th century art/photography movements (which are like bowel
movements but not as smelly ;-)) there has always been a battle between photo
realism of "the clear ones" (f/64 group, etc.) where heaven is infinite
resolution and hell is poor resolution and the more painterly "pictorialists"
(Julia Margaret Cameron, Clarence White, and into the 20th century w/ Steiglitz
and Mortensen, etc.) which sacrificed detail for overall mood/emotional effect.
>>
>>
>> >Anyway, you can check this by taking an RGB image, converting to CMYK,
>then
>> >shutting off the K channel. The results should show anyone a great deal
>> >about
>> >colour conversion issues for printing. If you would like, I could post
>some
>> >samples on a web page.
>>
>> Thanks, please do and let me know the URL(s). :-)
>
>I will start working on that one this afternoon. I just figured out a way
>to put
>together the channel images. It will be posted on a new thread when I have
>it
>uploaded to my site, since I think some other people may want to view it
>as well.
>
Great and thanks - could you explain more explicitly what you mean by "putting
together the channel images" and what that would accomplish/show? Thanks.
>>
>> >So my point is that having a K (black or grey level) correction to add
>to
>> >an RGB
>> >chip might get the proper colour representation.
>>
>> Why not a 4 chipped (or tubed) digital still camera w/ one chip for C,
>one for
>> M, one for Y and one for K - this to me would make more sense for those
>already
>> shooting for output in paper publication (magazines, newspapers, etc.)
>
>One problem is that a 4 channel image takes up more storage space. It also
>takes
>longer to store each image, making consecutive shots much slower.
This will change as chips get faster and storage capacity improves...
Printing
>has
>already moved towards HiFi and Hexachrome, which are six colour processing.
>Even
>though most colour printing is still CMYK, the other options are becoming
>more
>affordable and more common.
>
HiFi and Hexachrome sound great, but even w/ 6 inks, the more/more accurate
CMYK file that a printer has to work w/, chances are the more
accurate/"life-like" the final result would be. So a multi-chipped/sensors CMYK
camera still seems a better quality, if not cheaper in price, solution to more
accurate/better hue discrimination w/ perhaps a larger gamut/color space due to
having more accurate info available than any Foveon or Bayer patterned RGB
sensor array might provide.
>All these are still compromises. Some films add extra colour layers to
compensate
>towards some areas of the visual spectrum, so they are already more complex
>than
>RGB. To do this with digital SLRs would be costly.
Costly now, but if this expanded color technology is persued now, years down
the road, costs will most likely plummet as both the technology improves and
becomes cheaper and finds its market. The same technology might be used in some
future digi video format which would itself help offset its costs for usage in
digital still photography. RGB and even CMYK multi chipped/sensored cameras are
not the limits of color technology - only the beginning.
Also, remember that film
>is
>well developed evolutionary product. The advent of T Grain, and similar
>chemical
>changes was a big step, and may have been though by many to be the last
>great film
>innovation. However, now we are getting newer films that have even greater
>colour
>range capabilities, and even lower grain indexes.
>
Hot diggity dog! (Translation: that makes me very happy - I am _not_saying this
w/ sarcasm but w/ joy! - I like! film, and hope its technology will still
continue to develop regardless of digital and other future light/image
recording mediums).
>Meanwhile, the historically newer, and less well developed digital chip
>ideas have
>experienced few changes. We are still stuck with the Bayer pattern idea,
>though
>the Foveon at least attempts to point another direction. Most consumers
>are buying
>the cheap digitals, and less than 11% are printing anything, so again, where
>is
>the incentive to improve anything?
>
I think even Auntie Matilda would like a quantum leap in what could be
advertised as a selling point such as "more life-like colors" just as Sony
Trinitron tubes televisions have been sold because of their exceptional image
quality. It wouldn't surprise me if tv (in whatever form) becomes the 21st
century equivalent of the old slide projector and screen for showing images,
though I don't think that anything will replace the tactility of a photographic
album regardless of what types of prints those albums have - people enjoy
holding/touching things. Also, as you have mentioned before in some form, some
digi video cameras can make low quality stills too, a combined digi video/still
multi-chipped camera might be the perfect way to explore and financially
supplement both sides (still and video) of newer more accurate color
technology(ies).
Yes, but I still don't see why a CMYK monitor w/ Cyan Yellow Magenta and Gray
level pixels couldn't be designed. I guess the world is stuck w/ the
limitations of RGB monitors because that's only what the manufacturers make
currently, or am I wrong about this and someone actually has come out w/ a CMYK
CRT/LCD computer and/or tv display?
Thanks Gordon, I'll check out that URL. :-)
Hopefully - as my photographer friend in Scotland likes to say "we shall
see..."
>>
>>
>> >Better to base a future camera on a wavelength meter that knew what colour
>> >reflection corresponded to which colour.
>>
>> Tubes! ...or their equivalents (redesigned for the 2000's w/o the
smear/other
>> degradation factors). :-)
>
>Yeah . . . it may take a new spin on an old idea. Okay, how about this for
>weird,
>a moving strip of sensitized material that then goes through a line scanner,
>within the camera body. The sensitized material holds the colour information
>only
>for a very short period of time, during which is passes a recording areas
>that
>discharges the photons, and reads the emitted wavelengths. Then that
information
>is stored as a digital file. The moving strip could be a continues loop
>within the
>camera body that returns to a neutral state upon discharge. This could be
>similar
>to the way some radiation monitoring equipment works, though perhaps some
>of the
>patents that Polaroid has may be more applicable for further development.
>
I'll go you one better Gordon - didn't the first/original tv cameras have some
kind of moving shutter (my memory is hazy so perhaps I am imagining this) - why
not make a video or a still camera or a combined video/still camera that uses
its imaging sensors (RGB or CMYK, one snesor for each color in either case) as
shutterblades that spin, but not to block light but to record an image w/ each
of the color sensors as they take their position behind the lens. A synched
real shutter or LCD shutter would be used to let light onto each chip as it
moves into position (and the chip remains stationary long enough for a
still/clear image to be recorded in each of the RGB or CMY and K "colors").
Alternatively you could have a moving beam splitter-like/fan-like shutter
(almost like the type used on some movie cameras) made up out of _solid_ (_not_
semi-transparent) mirrors to reflect light to each chip as it/the shutter
rotated upon its central axis and stopped temporarily to record each static
image so there would be no light loss due to semi-silvered beam splitters
sucking up the light. If you wanted to get even wilder you could have each 3, 4
or more sided pixel element suspended in liquid and/or a magnetic field
spinning or possibly have their chemical nature/orientation of crystals change
w/ whatever wavelength of light hits them so both amplitude of light (gray
scale/"K") and hue could be recorded _continuously_ in analog form and then
possibly (or possibly not) broken down into digital data to be recorded. Wild
ideas, eh? (I'm using that last almost word to practice my Canadian "eh" ;-)).
The old "no one's gone broke underestimating the intelligence of their
audience" scenario/mind think ;-).
I can see Seurat (or "sewer rat" as I like to call him (if you think this is
bad I believe Gauguin called him "The Mad Dotter" ;-), yes I love his work, not
just for his technique but for his vision/treatment of his subject matter and
their mood/emotion, especially the circus stuff though Le grand Jette is both
beautiful and a bit "stiff"/cold in its lack of emotion, almost as if he was
taking future lessons from Escher ;-)) rolling over in his grave or in his
impressionist dots.
>>
>> The quality is a non-issue for digital now, because a handful
>> >of
>> >cameras are not so incredibly bad that you know from the moment you see
>> >an image,
>> >that it was done digitally.
Which of the few portable DSLR still digi cameras, if any, look
indistinguishable to you in color depth and contrast range (and not just
resolution) from film? What qualitiy(s) do they have that are they
indistinguishable from film in image quality?
Those of us in the pre-press, publication,
>and
>> >design
>> >community can tell you what software was used to layout each and every
>> magazine
>> >on
>> >the market,
How do you tell Quark Xpress from another layout program, what visually gives
away the software used?
and spot a scanned image from a digital only one . . . this
>> >is
>> >experience, and should be completely irrelevant for 98 % of the population,
>> >
Make that 99.999999998% of the population - yes, I'm being over generous here
;-).
I thought bling bling was money, do you know what you're even trying to say
here to me, if so, I don't, shake it don't break it, uh huh uh huh, uh huh,
wazzzzzzzzzzzzzzzzzzzzuuuuuuuuuuuuuuuuuupppppppppppp????????? :-)
>
>Ciao!
>
>Gordon Moat
>Alliance Graphique Studio
><http://www.allgstudio.com>
>
Pip, pip, CheerioTM(s) (the breakfast cereal) and I'll see you on the flip flop
magop!
This post is...
© 2003 Lewis Lang
All (breakfast cereals and) Rights (and non-Bayer patterned aspirin and film
grains) Reserved
Kinon
detail. Look closley, and the slide is better. Scanner noise and loss of
sharpness are evident, though.
"Simon Stanmore" <nom...@thanks.com> wrote in message
news:104438685...@damia.uk.clara.net...
Candice H. Brown Elliott
I hope y'all don't mind my jumping into the middle here...
cont...@aol.comnospam (Lewis Lang) wrote in message news:<20030205055136...@mb-mc.aol.com>...
> >Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
> >From: Gordon Moat mo...@attglobal.net
> >Date: Tue, Feb 4, 2003 7:43 PM
> >Message-id: <3E4017C5...@attglobal.net>
> >
Most consumers
> >are buying
> >the cheap digitals, and less than 11% are printing anything, so again, where
> >is
> >the incentive to improve anything?
> >
The incentive is in marketing hype... bigger MegaPixel numbers are the
present fight... but color will be the next. Also, dynamic range.
The CMOS imagers are being improved using some of the very same
techniques found in the mamalian retina, processing at the pixel level
that allows for wider dynamic range so that shadows and highlights are
fully imaged, compressed, without saturation (solarization) into the
white or black.
Ya'll also touched on Foveon and the dominance of the Bayer pattern.
Fuji-film devised a better color filter array (CFA)... that better
captures the high spatial frequencies, with less artifacts, than the
Bayer pattern... %$#@! even Bayer knew of a better pattern... in the
very same patent in which he disclosed the classic Bayer pattern, he
showed an improved one that better matched the chromatic spatial
frequency sensitivity of the human eye. But it is / was a larger
repeat pattern, requiring more processing (which in the ol'days was
more expensive than today). And even we at ClairVoyante Labs have
been developing better CFAs, mostly for the display side, but also for
cameras.
> >>
> >> Unfortunately, that is
> >> >only one
> >> >problem. A pure Yellow is very tough to photograph with a digital camera,
> >> >and any
> >> >true Cyan rarely records properly. To make things even worse, all computer
> >> >monitors do not display colours accurately when they are close to Yellow,
> >> >or Cyan.
> >>
I must interject here... While it is true that cyan cannot be
reproduced well on an RGB display, yellow is usually very well
reproduced on RGB LCDs. The real problem area is deep greens through
cyans. This is because, in the interest of bright white, the green
primary of the displays are very "yellow" to start with. The whole
upper left hand side of the 1931 CIE color coordinate chart is lopped
off!
> >> Pardon my ignorance but why not also have special monitors (if possible)
> that
> >> have C, M, Y, and K (spearate luminance for black and white gray levels)
> pixels
> >> to alleviate this problem.
>
> Yes, but I still don't see why a CMYK monitor w/ Cyan Yellow Magenta and Gray
> level pixels couldn't be designed. I guess the world is stuck w/ the
> limitations of RGB monitors because that's only what the manufacturers make
> currently, or am I wrong about this and someone actually has come out w/ a CMYK
> CRT/LCD computer and/or tv display?
Such a monitor would have to be a subtractive system, with four
separate flat panel LCD displays stacked ontop of each other. They
have been built in R&D labs, but they are far too expensive to
actually market !!! They also require very bright backlights that
gulp power. Perhap in the future, such systems will be reach the
market, but I doubt it. Actually, the displays may go the other
way... build emissive panels, such as Organic Light Emitting Diode
(OLED) systems with more that three primaries, as you guys already
touched on.
I have read a number of papers on projector systems with more than
three primaries, to better cover the full natural color gamut of the
human eye. One such uses two conventional-three-color projectors
aligned to project on the same screen, the primaries of each of the
projects is shifted to create a six primary system, with a color gamut
mapping algorithm processor splitting the image between the six
primaries.
Candice H. Brown Elliott
Chief Technology Officer
ClairVoyante Laboratories
www.clairvoyante.com
Paul Repacholi
> I think the 3 CCD (or CMOS, or whatever) idea would help a bit, but still not the
> tonal range of film. There are quite small 3 chip video cameras already, so I do
> not see why some company has not yet packaged a 3 chip stills camera, other than
> perhaps too high costs.
There used to be 3 CCD ENG TV cameras, but the small cost increase
forced them out. :(
Small is also a problem. Remember, you are dealing with a device that
works by storing charge, and if the chearge get too high, it spills
over into adjoining pixels and you get flair. It also often saturates
things and can take a LONG time to recover. Smaller CCD equals smaller
volume pixel, equals smaller well capacity. (And more surface effects
and other agrivation)
With room temperature readout noise of probably 30 e- of so (Does
anyone have any solid numbers on this? The saleks go all gooie and the
best I have been able to get is `better'...) from a total well
capacity of 100K to 180K e-, there are not top line astrominy grade
chips remember!, then we are very limited it what can be done.
Interestingly, `blad has just released a 645 and has a digital back
for it.
I will be having a look at a Sigma this afternoon. Be interesting to
see how it goes. If they let me take photos in the darkroom that is!
--
Paul Repacholi 1 Crescent Rd.,
+61 (08) 9257-1001 Kalamunda.
West Australia 6076
comp.os.vms,- The Older, Grumpier Slashdot
Raw, Cooked or Well-done, it's all half baked.
EPIC, The Architecture of the future, always has been, always will be.
Paul Repacholi
> On storage I confess to confusion. Consumer written CD-ROM's are
> stated to be good for about 100 years. I suspect that the real "life"
> is probably closer to 200 years.
Try 0-10
> Will the readers be around? When I made a living writing software
> I recovered data off of a 9 track tape that was about 12 years old.
> With parity and checksum there were 0 errors on roughly half a meg
> of data (a huge amount in those days). 9 track tape was "rated" for
> 5 years storage. While checksums are not the best way to verify
> data, it is reasonable and the data (navigation system flight
> recordings) had no blatent errors in it.
Tell me :) Or better, find me a 7 track head!
You need the media to still hold the data, you need the do-hickey to
read out the bits, and you need to know what the bits mean! Drop one,
and it is all just noise...
Simon Stanmore
there is clearly image information on the digi' capture that just isn't on
the scan - mainly in that foilage area. Of course that may be an issue with
the taking and/or the scan of the slide though . . .
Simon
ThomasH
>
> >Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
> >Date: Sun, Feb 2, 2003 7:07 PM
> >Message-id: <20030202140720...@mb-bj.aol.com>
> >
> >Lewis,
> >reason. "Because it is film" Please excuse my nostalgic wanderings........
>
> No problem :-)
>
> I am secretly hoping for a medium that incorporates the best of both media -
> the grainlessness (when desired) of digital, the high hue discrimination and
> resolution of film, the longevity of film, the non-quality loss of digital and
> the price of a Big Mac ;-)
I agree with one exception: You mention the longevity of film
as superior, but we know that the film demands expensive archiving
to preserve its information, maybe with an exception of Kodachrome.
If stored in regular room temperature conditions, E6 film can not
stay in good shape for over a decade or two. Not to talk about
the possible impact of fungi, bacteria and dust. All my slides
from the 80ties look a bit faded out, especially highlights
are almost without detail.
For me the digital data storage was always synonymous with
"forever unchanged."
If you mean that a media might deteriorate, sure, but you can
copy your data routinely to a next generation storage and the
charm of it is that every copy is precisely exactly like the
original.
For example I found my old 1st generation DAT tape with a copy
of my Phd theses. My DAT DSS3 Seagate Scorpio (24gig capacity)
could read that tape recorded in mid 1980ties without a problem.
The tape was almost 18 years old. Of course a bit luck helped.
I used back than a PC-DOS program called Arcada Backup to store
my data. Arcada was than purchased by Seagate and Seagate Software
made a contract with Microsoft to deliver this very backup as the
W95 backup, what is still to this day a valid Microsoft format.
Otherwise I would have to boot my computer in PC mode and try
my best with decoding the information! Here we come to another
not yet mentioned problem with digital storage: The choice of
both media and format. I always recommend to go with the major
standard, regardless the price and capacity. All these QIC and
Travan tapes may become worthless tomorrow. DAT somehow
perseveres with Sony and HP behind and is in mass usage for
storage and studio audio technology. Meanwhile exist DSS4 DAT
data storage standard.
Another good storage seem to be the very regular hard drives,
which now may be used externally via Firewire or USB2. This is
how I duplicate all my data for safety purposes aside of the
DAT tape backup.
Cost? Lets just look:
One single respectable lens like 70-200 f2.8 costs $1500 or so.
A poor mans quick-done storage cost just a fraction of one
single lens:
120Gig ATA drive cost approx. $140.
See storagereview.com for details of the good, bad and ugly.
I recommend Seagate Barracuda V or IBM deskstar
Firewire/USB2 card cost $30-$80
See Adaptec, Orange etc. double standard cards.
Drive kit housing for Firewire, like the ADS Pyro cost $50-$100.
ADStech.com as one example. I use it without a problem.
Lovely, faster than one old internal drive.
Thus you invest some $300 for 120Gig of external storage, now.
'Tomorrow' this standard will be outgunned by a larger solution.
If you are a profi, triplicate, quadruple your stored images, keep
them apart. One of these copies will survive than no matter what.
Copy them every 5 years from one media to another.
Can you do that with film? No, not a few hundreds buckazoids anyway.
Thomas.
>
> I know, I know, shoot film and buy a scanner (when I'm flush) still one can
> dream about what has not yet/might/might not happen - so long as they figure
> out a way to lower the choloesterol of this new medium too ;-).
>
Alan Browne
I need Aspirin.
Candice H. Brown Elliott wrote:
>
> Ya'll also touched on Foveon and the dominance of the Bayer pattern.
> Fuji-film devised a better color filter array (CFA)... that better
> captures the high spatial frequencies, with less artifacts, than the
> Bayer pattern... %$#@! even Bayer knew of a better pattern... in the
> very same patent in which he disclosed the classic Bayer pattern, he
> showed an improved one that better matched the chromatic spatial
> frequency sensitivity of the human eye. But it is / was a larger
> repeat pattern, requiring more processing (which in the ol'days was
> more expensive than today). And even we at ClairVoyante Labs have
> been developing better CFAs, mostly for the display side, but also for
> cameras.
>
>
Lewis Lang
>Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
>From: ThomasH tho...@coco.net
>Date: Thu, Feb 6, 2003 12:07 AM
>Message-id: <3E41A757...@oracle.com>
>
>Lewis Lang wrote:
>>
>> >Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
>> >From: quietli...@aol.com (Quietlightphoto)
>> >Date: Sun, Feb 2, 2003 7:07 PM
>> >Message-id: <20030202140720...@mb-bj.aol.com>
>> >
>> >Lewis,
>> > Silver based film will always be superior to digital for the following
>> >reason. "Because it is film" Please excuse my nostalgic wanderings........
>>
>> No problem :-)
>>
>> I am secretly hoping for a medium that incorporates the best of both media
>-
>> the grainlessness (when desired) of digital, the high hue discrimination
>and
>> resolution of film, the longevity of film, the non-quality loss of digital
>and
>> the price of a Big Mac ;-)
>
>I agree with one exception: You mention the longevity of film
>as superior, but we know that the film demands expensive archiving
>to preserve its information, maybe with an exception of Kodachrome.
>
Slide/neg pages aren't all that expensive and I got a rather (can't remember
the price offhand) inexpensive metal filing cabinet from KMart for storage that
should do me for quite a while.
>If stored in regular room temperature conditions, E6 film can not
>stay in good shape for over a decade or two. Not to talk about
>the possible impact of fungi, bacteria and dust. All my slides
>from the 80ties look a bit faded out, especially highlights
>are almost without detail.
>
Newest data from Henry Wilhelm suggests dark storage life of 40 years for
Fujichromes and 120+ years from E-6 Ektachromes. I haven't had problems of
either fading on me yet and I 've been using both since about 1983...
>
>For me the digital data storage was always synonymous with
>"forever unchanged."
>
True, if recopied regularly every 0-5 years (I'm only being slightly
conservative here) the data should be fine. However the dyes in th CDs are very
unstable in comparison to film dyes, so readability/access/compatability to
that "forever unchanged" data, not to mention the fact that the hardware to
support reading that data may not be available in 20 years time (digital ='s
planned obsolescence) means that even Fujicolro negs, rated by Wilhelm at 20
years dark storage life before significant change, seems to look like a
Methsalean medium compared w/ digital storage. At least film doesn't need to be
recopied every 5-10 years if its stored well and freezing film may help it last
centuries. I am not against digital, but I see it as a short term back up
solution in case of disaster, not along term archival storage medium as it
needs to be constantly recopied every so often both while you are alive and by
your heirs/whomever once you pass on. Digital is not an archival solution just
an archival problem delayed, despite its benefits. Others on here have posted
to the ng mentioning deterioration of CDs in as little as 2-3 years.
>
>If you mean that a media might deteriorate, sure, but you can
>copy your data routinely to a next generation storage and the
>charm of it is that every copy is precisely exactly like the
>original.
>
And you are tied to a life time of copying and recopying huge amounts of data
on a non-archival unstable storage medium that may or may not exist a short
time down the road of life...
>For example I found my old 1st generation DAT tape with a copy
>of my Phd theses. My DAT DSS3 Seagate Scorpio (24gig capacity)
>could read that tape recorded in mid 1980ties without a problem.
>The tape was almost 18 years old. Of course a bit luck helped.
>I used back than a PC-DOS program called Arcada Backup to store
>my data. Arcada was than purchased by Seagate and Seagate Software
>made a contract with Microsoft to deliver this very backup as the
>W95 backup, what is still to this day a valid Microsoft format.
>
Had you waited too long, perhaps into the mid 90's you might not have been able
to read that file. Also, relying on the "generosity" of Microsoft or any other
company to keep alive a way(s) of reading your data is a non-prudent gamble at
best.
>Otherwise I would have to boot my computer in PC mode and try
>my best with decoding the information! Here we come to another
>not yet mentioned problem with digital storage: The choice of
>both media and format. I always recommend to go with the major
>standard, regardless the price and capacity. All these QIC and
>Travan tapes may become worthless tomorrow. DAT somehow
>perseveres with Sony and HP behind and is in mass usage for
>storage and studio audio technology. Meanwhile exist DSS4 DAT
>data storage standard.
>
Not yet mentioned problem? All we fever do here is talk about the obsolecsnece
of digital storage drives/media when the subject of archival longevity of film
vs. digital media comes up on this ng. The Fujitsu magnetic optical
drives/media? are rated at about 10 years on their website, these seem to be
some of the longer lasting media/drives and have been around for years though
they are don't have "Zip disc" and "CD/DVD" popularity. I aggre in theory w/
what you say about always choosing the most popuar media/drives however I am
leaning towards less popular and longer lasting (both archivally/longevity and
in terms of how long the company/format has been in business selling like
products to the public) digital media such as magnetic optical. Frankly I feel
that all digi solutions for both archival longevity and
brand/company/drive/media longevity in terms of sales availability of the
product (or similar products years down the line to read the older products)
are piss poor to poor at best at this time. I am looking at external hard
drives also as a _longer_ but not long term solution not only for its cheapness
but for its longevity. My own computer's hard drive is going on 9 years, though
I have no info on how long other external hard drives as well as removable
media last, I wish Henry Wilhelm would do something more on digital sotrage
media archival longevity as well as an update to his film/paper longevity
info...
>Another good storage seem to be the very regular hard drives,
>which now may be used externally via Firewire or USB2. This is
>how I duplicate all my data for safety purposes aside of the
>DAT tape backup.
>
I should have read this first instead of doing all that typing above! :-)
>
>Cost? Lets just look:
>
>One single respectable lens like 70-200 f2.8 costs $1500 or so.
>A poor mans quick-done storage cost just a fraction of one
>single lens:
>
My lens was in the $100-150 range and I can't afford a pro level fast constant
aperture lens (nor is it necessary for my work, a constant high quality f/4
does me fine) but anyways...
> 120Gig ATA drive cost approx. $140.
> See storagereview.com for details of the good, bad and ugly.
> I recommend Seagate Barracuda V or IBM deskstar
>
> Firewire/USB2 card cost $30-$80
> See Adaptec, Orange etc. double standard cards.
>
Don't think I need a card for MAcs though...
> Drive kit housing for Firewire, like the ADS Pyro cost $50-$100.
> ADStech.com as one example. I use it without a problem.
> Lovely, faster than one old internal drive.
>
>Thus you invest some $300 for 120Gig of external storage, now.
>'Tomorrow' this standard will be outgunned by a larger solution.
>
>If you are a profi, triplicate, quadruple your stored images, keep
>them apart. One of these copies will survive than no matter what.
>Copy them every 5 years from one media to another.
>
_Exactly_ my point, even if one doesn't mind spending for 2 or 3 additional
hard drives every 5 years this is not really a long term solution just a
problem delayed... Manufacturers need to concentrate on longer lasting digital
media, not just hyped up claims of such...
>Can you do that with film? No, not a few hundreds buckazoids anyway.
>
>Thomas.
>
Just back up the film to digital for copies as needed (_not every image needs
to be scanned_, just the best/most useful) or have optical or digital dupes
made of only the very best of theimages onto film again.
Lewis Lang
>Date: Wed, Feb 5, 2003 7:51 PM
>Message-id: <1e9ffba8.03020...@posting.google.com>
Thanks for "chipping" in w/ your input Candice. I can't help but wonder in the
rush to digital and more MP that not only color, up till recently
(Foveon/etc.), has been largely ignored/unmarketed as a major selling point but
whether there are analog and other yet to be disclosed or even tested
technologies that could record more life-like info than the digital process of
chopping up everything into bits and bytes. Chemical (film) keeps on improving
(as does digital) in their ability to record info but digital seems (for the
present anyways) locked into a one technology (CCD/CMOS Bayer patterned)
mindset. Nikon's D1x and the Fuji S2's octaganol/hexagonal technologies seem
more like kludges to get more data/detail out of the current chip resolutions.
Foveon at least takes things in adifferent direction as far as color
depth/accuracy are concerned. But I am left with the feeling that digital
records, no matter how good, only a portion of the information that it could
and that something analogue (whether film or some other medium) either instead
of or added to digital capture might be able to overcome the limitations on
color and resolution quality...
Lewis Lang
>Date: Thu, Feb 6, 2003 12:39 AM
>Message-id: <3E41AEB5...@videotron.ca>
>
>
>
>I need Aspirin.
>
Bayer patterned, Foveon or Tylenol? ;-)
Regards,
Alan Browne
Lewis Lang wrote:
>>Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
>>From: Alan Browne alan....@videotron.ca
>>
>>I need Aspirin.
>>
>
>
> Bayer patterned, Foveon or Tylenol? ;-)
>
;-)
Kinon
the scanner. But the tones on the digital shot are flat, while on the film
shot, they are more realistic. Look at the base of the tree, and some of the
patterns on the dig shot are merely fields of a like color, while on the
film shot, you can see gradations. I'd like to see a better scan, as well.
"Simon Stanmore" <nom...@thanks.com> wrote in message
news:104448638...@doris.uk.clara.net...
Simon Stanmore
graduation on the tree base and the shadows. Still, there is detail in the
foilage on the D60 capture that just isn't on the scan. It's swings and
roundabouts to me on detail... but... The digi's lack of noise/grain and
it's far more accurate colour rendition are both very obvious and
significant advantages IMO. I would take those qualities over the oh-so
subtle benefits of the slide scan every time (caveat being THIS slide scan)
Simon
Gordon Moat
> Any noise reducing techniques for scanned color film images you'd care to
> share? :-)
I will give you a simple idea, though it takes some experience to properly
implement this. The main idea is that digital noise shows more in one colour
channel than another. Just like sharpening the entire image can make that channel
noise worse, trying to smooth digital noise should only be done on one channel at
a time. Try using very small radius Gaussian Blur on just one or two noisy
channels. You may need to look at the image in both RGB and CMYK, to find which
channels contain the most noise. You could also do this in limited areas by using
the Blur Tool, though I recommend creating a selection, or custom channel mask
first, to limit your corrections area.
There is no such thing as a technique that works every time, nor with every image.
The final output must also be considered, since you do not want your corrections
to be visible either. It would be bad to get rid of noise, and introduce flaws.
>
> >So this brings me to another observation. Chips capture to square or
> rectangular
> >pixels. The computer monitors also display in regular square (rectangular)
> >pixels.
> >Anytime there are edges that do not fit into these shapes, there will be
> >some
> >error in recording. Many digital cameras already correct for this in built-in
> >software, though not all scanners do this. Currently, since many digital
> >camera
> >users really do not understand PhotoShop that well, the corrections made
> >within a
> >digital camera can give better images for some users than they could do
> >with film.
>
> Better in what/which sense(s)?
Since many consumer digital cameras have automatic white balance, this eliminates
some colour inaccuracy that would otherwise appear with film shots. The yellow
house interior lighting effect on some prints would be an obvious example.
However, these digital cameras often end up not teaching the user anything about
exposure.
>
>
> >This is not a quality issue, but one of allowing more consistent results.
> >I have
> >seen quite a few people improve their photography using a digital camera,
> >then go
> >back to shooting a film camera because they wanted better colour quality
> >prints
> >(or B/W in a few cases). In this way, digital photography experience can
> >help
> >improve film camera photography.
> >
>
> Improve their photography in what way(s), composition and/or exposure due to
> the availability of instant playback on the cameras' LCD screen on back,
> improvement insome other aspect(s) of their photography - which aspects of
> their photography get improved?
Having less exposure related, or colours being way off, leaves room to concentrate
on other issues. Most people are terrible at composition, or even getting the idea
of focal points. However, the LCD on the camera is a joke, and tells almost
nothing except facial expressions, and perhaps a tiny bit about composition. A
larger viewing area is needed, meaning playback on a television, or computer
screen. There may be more hand held (like Palm, HandSpring, etc.) that allow
slightly larger image preview in the near future, if there is market demand for
such products. Leaf makes something like that for their digital back for medium
format (expensive).
>
> Think back to the past, way in the past w/ Daugerrotypes and other 19th and
> 20th century processes that usually used large format and/or the vfinal viewed
> image was recorded directly onto the bitumen? or other surface. Weston/etc. w/
> their 8x10" contact prints, etc. Even go all the way back to the camera obscura
> several hundred years before there was any medium to record light's images.
> We've come to expect from the word/process of "photographic" that means highest
> quality which equals "holding up a mirror to nature" which means the highest
> detail possible recorded. Other effects are considered either "painterly" (soft
> focus) or just plain bad technique (to the sharpness obsessed). Among other
> 19th century and 20th century art/photography movements (which are like bowel
> movements but not as smelly ;-)) there has always been a battle between photo
> realism of "the clear ones" (f/64 group, etc.) where heaven is infinite
> resolution and hell is poor resolution and the more painterly "pictorialists"
> (Julia Margaret Cameron, Clarence White, and into the 20th century w/ Steiglitz
> and Mortensen, etc.) which sacrificed detail for overall mood/emotional effect.
Yeah, though it seems that many consumers fed on a diet of disposable cameras,
point and shoot cameras, APS, and cheap digital cameras, have come to expect
everything in focus. Even the kit zoom lenses usually sold with SLRs result in far
more large DOF shoots. Familiarity can get a first reaction, like fast food, but
to really continue that interest something difficult is needed. I like to shoot
many of my shots with lots of defocus area, but not everyone likes that . . . oh
well . . . .
Okay, time for a quote that relates to this:
"Photography is the only genre that can achieve a popular effect on the immediate,
visible level, and an elitist one after its initial impact on a deeper, more
subtle level." - Horst Wackerbarth
>
>
> >>
> >>
> >> >Anyway, you can check this by taking an RGB image, converting to CMYK,
> >then
> >> >shutting off the K channel. The results should show anyone a great deal
> >> >about
> >> >colour conversion issues for printing. If you would like, I could post
> >some
> >> >samples on a web page.
> >>
> >> Thanks, please do and let me know the URL(s). :-)
> >
> >I will start working on that one this afternoon. I just figured out a way
> >to put
> >together the channel images. It will be posted on a new thread when I have
> >it
> >uploaded to my site, since I think some other people may want to view it
> >as well.
> >
>
> Great and thanks - could you explain more explicitly what you mean by "putting
> together the channel images" and what that would accomplish/show? Thanks.
Not yet ready. When I post it, then ask more questions. Each channel that I post
will be a separate B/W JPEG, so three images for RGB, and four images for CMYK, so
you can look at each channel separately, with the full colour image in the centre
for reference only. I may do a second page that shows the same image after some
corrections, though it is tough to show CMYK issues in an RGB (internet) world.
> >All these are still compromises. Some films add extra colour layers to
> compensate
> >towards some areas of the visual spectrum, so they are already more complex
> >than
> >RGB. To do this with digital SLRs would be costly.
>
> Costly now, but if this expanded color technology is persued now, years down
> the road, costs will most likely plummet as both the technology improves and
> becomes cheaper and finds its market. The same technology might be used in some
> future digi video format which would itself help offset its costs for usage in
> digital still photography. RGB and even CMYK multi chipped/sensored cameras are
> not the limits of color technology - only the beginning.
Probably . . . there are quite a few advances made in video chips, then translated
for still capture. The advent of HD television may push better colour capture
technology.
>
> I think even Auntie Matilda would like a quantum leap in what could be
> advertised as a selling point such as "more life-like colors" just as Sony
> Trinitron tubes televisions have been sold because of their exceptional image
> quality. It wouldn't surprise me if tv (in whatever form) becomes the 21st
> century equivalent of the old slide projector and screen for showing images,
> though I don't think that anything will replace the tactility of a photographic
> album regardless of what types of prints those albums have - people enjoy
> holding/touching things.
Yeah, I think the direct to television approach might be a solution for some
people. I seem to remember the promises of the paperless office from computer
advancements, though there is now much more paper being generated. I also remember
the claims that the internet would eliminate the need for printed magazines, yet
there are now more magazines.
One of the great hopes of the digital camera makers was to increase computer
sales, though the published impact results fail to show a causal relationship. The
next great push was for printing images, which has yet to really become a great
reality. The hoped for increases in ink and paper sales for inkjet printers is not
nearly what was projected a couple years ago.
> Also, as you have mentioned before in some form, some
> digi video cameras can make low quality stills too, a combined digi video/still
> multi-chipped camera might be the perfect way to explore and financially
> supplement both sides (still and video) of newer more accurate color
> technology(ies).
More still digital cameras include short video, and even sound capability. I think
the consumer level will see more convergence of video and stills technology.
Unfortunately, we are back to the blinking "12:00" on the VCR syndrome, and these
things are still not very user friendly.
>
> Yes, but I still don't see why a CMYK monitor w/ Cyan Yellow Magenta and Gray
> level pixels couldn't be designed. I guess the world is stuck w/ the
> limitations of RGB monitors because that's only what the manufacturers make
> currently, or am I wrong about this and someone actually has come out w/ a CMYK
> CRT/LCD computer and/or tv display?
There are some new technologies in the works, but not much yet. There is OLED, and
another one called Light Emitting Paper (LEP). The first applications will
probably be in very small devices, like screens for mobile phones, etc. There is
much corporate investment in LCD, so I doubt the development is slowing down, or
that it will be abandoned for other technology.
> I'll go you one better Gordon - didn't the first/original tv cameras have some
> kind of moving shutter (my memory is hazy so perhaps I am imagining this) - why
> not make a video or a still camera or a combined video/still camera that uses
> its imaging sensors (RGB or CMYK, one snesor for each color in either case) as
> shutterblades that spin, but not to block light but to record an image w/ each
> of the color sensors as they take their position behind the lens.
I thought some video camcorders did have a moving shutter and prism system.
Unfortunately, I am not an engineer, though I have used several different video
systems, many of them 3 chip.
> A synched
> real shutter or LCD shutter would be used to let light onto each chip as it
> moves into position (and the chip remains stationary long enough for a
> still/clear image to be recorded in each of the RGB or CMY and K "colors").
> Alternatively you could have a moving beam splitter-like/fan-like shutter
> (almost like the type used on some movie cameras) made up out of _solid_ (_not_
> semi-transparent) mirrors to reflect light to each chip as it/the shutter
> rotated upon its central axis and stopped temporarily to record each static
> image so there would be no light loss due to semi-silvered beam splitters
> sucking up the light. If you wanted to get even wilder you could have each 3, 4
> or more sided pixel element suspended in liquid and/or a magnetic field
> spinning or possibly have their chemical nature/orientation of crystals change
> w/ whatever wavelength of light hits them so both amplitude of light (gray
> scale/"K") and hue could be recorded _continuously_ in analog form and then
> possibly (or possibly not) broken down into digital data to be recorded. Wild
> ideas, eh? (I'm using that last almost word to practice my Canadian "eh" ;-)).
Probably better to just use Photo Multiplier Tubes, though I am not so sure these
can be placed into a small and portable device. I think the power requirements
alone might make these unsuitable.
> >>
> >> The quality is a non-issue for digital now, because a handful
> >> >of
> >> >cameras are not so incredibly bad that you know from the moment you see
> >> >an image,
> >> >that it was done digitally.
>
> Which of the few portable DSLR still digi cameras, if any, look
> indistinguishable to you in color depth and contrast range (and not just
> resolution) from film?
I think several of them when worked over in PhotoShop, and printed in publications
small enough, are indistinguishable. Part of that is careful lighting. With the
larger and more well developed digital backs for medium format, and some scanning
backs, the results are quite good, though again careful lighting is often
required.
> What qualitiy(s) do they have that are they
> indistinguishable from film in image quality?
Colour accuracy, though as I stated only with careful lighting. You cannot judge
detail without being at the same place as the photographer to observe when and how
he photographed the image.
A side note on this: I graduated with one student that ended up working on print
optimization for inkjet manufacturers. He worked on enhancing image quality to
create ideal sample prints for each model of inkjet, for three different
companies. There was quite a bit of manipulation required for each inkjet output,
though the photos often originated from the same sources. Many of the photos were
quite large files from drum scanned medium format film. The largest time consumer
in his work was colour correction to optimize for each inkjet. My point is that if
you work enough on something, the differences may become invisible. However, not
all work has the luxury of limitless time to get everything right; most of us have
to work within more stringent deadlines.
>
>
> Those of us in the pre-press, publication,
> >and
> >> >design
> >> >community can tell you what software was used to layout each and every
> >> magazine
> >> >on
> >> >the market,
>
> How do you tell Quark Xpress from another layout program, what visually gives
> away the software used?
Lots of subtleties, like how they handle text, placement of text boxes, text
around images, and optical text alignment. So far, I think that Adobe InDesign has
the least layout control issues, though it has a nasty habit of crashing more
often than Quark; probably why it has a built-in file recovery feature.
>
> I thought bling bling was money, do you know what you're even trying to say
> here to me, if so, I don't, shake it don't break it, uh huh uh huh, uh huh,
> wazzzzzzzzzzzzzzzzzzzzuuuuuuuuuuuuuuuuuupppppppppppp????????? :-)
>
What?!?! You did not buy into that idea that pings on your web site will bring you
money?!?! Anyway, since you live in somewhat reality, it was just a loose
association (pun, maybe). Oh well . . . .
Gordon Moat
> Guys,
>
> I hope y'all don't mind my jumping into the middle here...
Welcome aboard . . .
> > >>
> > >> Unfortunately, that is
> > >> >only one
> > >> >problem. A pure Yellow is very tough to photograph with a digital camera,
> > >> >and any
> > >> >true Cyan rarely records properly. To make things even worse, all computer
> > >> >monitors do not display colours accurately when they are close to Yellow,
> > >> >or Cyan.
> > >>
>
> I must interject here... While it is true that cyan cannot be
> reproduced well on an RGB display, yellow is usually very well
> reproduced on RGB LCDs. The real problem area is deep greens through
> cyans. This is because, in the interest of bright white, the green
> primary of the displays are very "yellow" to start with. The whole
> upper left hand side of the 1931 CIE color coordinate chart is lopped
> off!
A major revision to the ICC standard has been approved. These guys move along slowly. More at
<http://www.color.org/> about this.
Anyway, I do much of my edits on a very good LCD now, partly because of the yellow issues. It is still quite a
ways away from printed materials, enough that I still check with the colour picker to judge saturation levels.
The other issue is that process yellow is a fairly weak printing colour. A certain amount of improvement can
be gained by substituting a Pantone Yellow choice for the process yellow in the normal CMYK print, with only a
slight increase in cost.
Glad to hear that someone is working on this stuff. However, I still think this is more printing industry
related, though it would be nice for consumer products to hit the market in the future. I wish you guys the
best of luck with this.
Gordon Moat
> Gordon Moat <mo...@attglobal.net> writes:
>
> > I think the 3 CCD (or CMOS, or whatever) idea would help a bit, but still not the
> > tonal range of film. There are quite small 3 chip video cameras already, so I do
> > not see why some company has not yet packaged a 3 chip stills camera, other than
> > perhaps too high costs.
>
> There used to be 3 CCD ENG TV cameras, but the small cost increase
> forced them out. :(
I used a few in the recent past. Many new 3 chip cameras are CMOS or HAD sensor types.
I think the colour has greatly improved, especially with some of the high end DV gear.
>
>
> Small is also a problem. Remember, you are dealing with a device that
> works by storing charge, and if the chearge get too high, it spills
> over into adjoining pixels and you get flair. It also often saturates
> things and can take a LONG time to recover. Smaller CCD equals smaller
> volume pixel, equals smaller well capacity. (And more surface effects
> and other agrivation)
The current DV format 3 chip video cameras I have worked with only seem to have
slightly less low light image capture capabilities than single chip DV cameras. Other
than that, I do not know of an observable difference. The possibility of a smear error
exists in almost all these, though is almost eliminated in the most expensive systems.
I have yet to hear about smear problems in 3 chip HD video system cameras, so perhaps
that points a better direction for the future.
>
>
> With room temperature readout noise of probably 30 e- of so (Does
> anyone have any solid numbers on this? The saleks go all gooie and the
> best I have been able to get is `better'...) from a total well
> capacity of 100K to 180K e-, there are not top line astrominy grade
> chips remember!, then we are very limited it what can be done.
Astronomy is indeed low light conditions, and will show the worst problems of 3 chip
designs. All systems are compromises that work well under a given set of conditions;
step outside those boundaries and nothing but trouble . . . .
>
>
> Interestingly, `blad has just released a 645 and has a digital back
> for it.
The old 6x6 Hasselblads have many back choices in digital. The newer Kodak for 645
fits the new H1, though the camera is actually a Fuji manufactured product of
Hasselblad design. Sinar Bron just introduced a one shot digital back that is almost a
full 645 frame (expensive).
>
>
> I will be having a look at a Sigma this afternoon. Be interesting to
> see how it goes. If they let me take photos in the darkroom that is!
Best of luck, and remember to share if you find something of interest.
Candice H. Brown Elliott
>
> Thanks for "chipping" in w/ your input Candice. I can't help but wonder in the
> rush to digital and more MP that not only color, up till recently
> (Foveon/etc.), has been largely ignored/unmarketed as a major selling point but
> whether there are analog and other yet to be disclosed or even tested
> technologies that could record more life-like info than the digital process of
> chopping up everything into bits and bytes. Chemical (film) keeps on improving
> (as does digital) in their ability to record info but digital seems (for the
> present anyways) locked into a one technology (CCD/CMOS Bayer patterned)
> mindset. Nikon's D1x and the Fuji S2's octaganol/hexagonal technologies seem
> more like kludges to get more data/detail out of the current chip resolutions.
> Foveon at least takes things in adifferent direction as far as color
> depth/accuracy are concerned. But I am left with the feeling that digital
> records, no matter how good, only a portion of the information that it could
> and that something analogue (whether film or some other medium) either instead
> of or added to digital capture might be able to overcome the limitations on
> color and resolution quality...
>
> Regards,
>
> Lewis
>
>
Lewis, actually... you have touched on what I have been working on for
years. Digitization of images suffers mostly at the reconstruction
stage, not the capture side... But first, some comments regarding what
happens with digital verses film cameras...
First, what makes film, film like? (BTW... this is good practice for
me... as I am presenting a paper touching on the topic of improving
Digital Cinema projectors at SMPTE on the 1st of March) Why is it
different than digital cameras? First, film is made with three (or
more) layers of photosensitive, randomly distributed "pixels"...
oops... "silver-halide grains". But seriously, the grains are like
pixels, each detecting light and being reduced or not reduced, by
varying amounts. This is just like pixels aren't they? The key is
that they are randomized... and that they are in different places from
layer to layer. This is very key! And... it is much like what
happens in the digital camera, save that the camera has a fixed
pattern of color dots that are not overlapping. This introduces
aliasing and moiré distortions. The second thing different is that of
quantization. Film is quantized, as the photons are quantized, as are
the atoms... but there are so many levels of quantization that it
might as well be continous as far as the human eye is concerned. But
in today's digital cameras, with only 8-12 bits... and linear at
that... it creates quantization error.
Now... what is wrong with our digital camera? Not much really... as
is improving all of the time... the biggest problem is at
reconstruction, on a monitor, or onto print film. The problem is
that, though we often talk about the Shannon-Nyquist sampling
Theory... and how we are using it when we take pictures... We simply
are not using it when we reconstruction the sampled image !!! The
Shannon-Nyquist sampling theorem states that we can sample and
reconstruct a bandlimited image (an image with no spatial frequencies
higher than half the sampling frequency... the Optical Modulation
Transfer Function of the camera lens usually provides this limitation,
if properly matched to the camera chip), perfectly... *IF* we use the
mathematically proper reconstruction function, that being an infinite
and continous function called a "sinc" function. The problem? This
sinc function has negative lobes... that must be scaled and summed,
with all of the other sample points multiplied by the sinc function.
There are only two ways to perform this... one is to have "negative
light" from a holographic laser system that performs the sinc function
optically... or using a very, very high resolution monitor / film
printer and perform the sinc function in a Digital Signal Processor at
at least 4X, and preferably much higher, resolution of the original
digital image. But... we don't have such super high resolution
monitors or film printers. Instead we use "one-to-one" mapping of
camera pixel to display or film recorder... or worse, we go the other
way... use multiple camera pixels to monitor pixels !!! The result?
Visible aliasing and moiré artifacts !
The solution? Higher resolution !!! And / or color subpixel rendering
on the flat panel display or projector, using displaced color
subpixels, mimicing the look of film. This reduces the artifacts
wonderfully. This is why film, inspite of having no built in sinc
function, still looks good... because of built in "color subpixel
rendering"... that and very high resolution.
Whew!!! Now... where's that Bayer Asprin?
Lewis Lang
>Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
>From: Gordon Moat mo...@attglobal.net
>Date: Thu, Feb 6, 2003 11:08 PM
>Message-id: <3E42EADE...@attglobal.net>
>
>Lewis Lang wrote:
>
>> Any noise reducing techniques for scanned color film images you'd care
>to
>> share? :-)
>
>I will give you a simple idea, though it takes some experience to properly
>implement this. The main idea is that digital noise shows more in one colour
>channel than another. Just like sharpening the entire image can make that
>channel
>noise worse, trying to smooth digital noise should only be done on one channel
>at
>a time. Try using very small radius Gaussian Blur on just one or two noisy
>channels. You may need to look at the image in both RGB and CMYK, to find
>which
>channels contain the most noise. You could also do this in limited areas
>by using
>the Blur Tool, though I recommend creating a selection, or custom channel
>mask
>first, to limit your corrections area.
>
>There is no such thing as a technique that works every time, nor with every
>image.
>The final output must also be considered, since you do not want your
corrections
>to be visible either. It would be bad to get rid of noise, and introduce
>flaws.
>
Thanks for the informative tips, Gordon. What/which flaws would doing selective
Gaussian blur/etc. create that I should be on the look out for
(over-softening?, some kind of halo effect of certain colors?, others?)?
SNIP
>A
>larger viewing area is needed, meaning playback on a television, or computer
>screen. There may be more hand held (like Palm, HandSpring, etc.) that allow
>slightly larger image preview in the near future, if there is market demand
>for
>such products. Leaf makes something like that for their digital back for
>medium
>format (expensive).
>
I don't understand why the back of a digital camera can't have the equivalent
of a flip down and/or expandable screen, maybe 4-6 inches diagonal, which would
go a long ways towards eliminating the problems of judging focus and
composition on the current tiny screens...
SNIP
>Okay, time for a quote that relates to this:
>
>"Photography is the only genre that can achieve a popular effect on the
>immediate,
>visible level, and an elitist one after its initial impact on a deeper,
>more
>subtle level." - Horst Wackerbarth
>
Great/true quote! :-)
BIG SNIP
I guess the world is stuck w/ the
>> limitations of RGB monitors because that's only what the manufacturers
>make
>> currently, or am I wrong about this and someone actually has come out
>w/ a CMYK
>> CRT/LCD computer and/or tv display?
>
>There are some new technologies in the works, but not much yet. There is
>OLED, and
>another one called Light Emitting Paper (LEP). The first applications will
>probably be in very small devices, like screens for mobile phones, etc.
>There is
>much corporate investment in LCD, so I doubt the development is slowing
>down, or
>that it will be abandoned for other technology.
>
Light emitting paper, sounds interesting, I wonder if a multi-layered version
of that could be used for cheap high quality (in the far future display systems
not just as a light source but to somehow display images like LCD displays do
now. Imagine a 70 inch diagonal screen w/ (pardon the pun) "paper thin"
speakers that weighs about 5 pounds and that you could roll up and hang on your
wall ;-)...
BIG SNIP
>> Which of the few portable DSLR still digi cameras, if any, look
>> indistinguishable to you in color depth and contrast range (and not just
>> resolution) from film?
>
>I think several of them when worked over in PhotoShop, and printed in
publications
>small enough, are indistinguishable. Part of that is careful lighting. With
>the
>larger and more well developed digital backs for medium format, and some
>scanning
>backs, the results are quite good, though again careful lighting is often
>required.
>
What exactly do you mean by "careful lighting" - lighting that is w/i the
illumination levels and contrast range of the sensor to eliminate noise,
multiple light set-ups that include key, fill, background, hair and/or rim
lights? Lights used w/ gels and barn doors and grids to do complex lighting
effects equivalent to the best still photography and movie lighting set ups?
All of the above and/or something else?
>> What qualitiy(s) do they have that are they
>> indistinguishable from film in image quality?
>
>Colour accuracy, though as I stated only with careful lighting.
So careful lighting to you means white balancing for accurate/"catalog"
neutrality of color rendition?
You cannot
>judge
>detail without being at the same place as the photographer to observe when
>and how
>he photographed the image.
>
This aspect of lighting seems more a function of contrast ratios...
BIG SNIP
>> I thought bling bling was money, do you know what you're even trying to
>say
>> here to me, if so, I don't, shake it don't break it, uh huh uh huh, uh
>huh,
>> wazzzzzzzzzzzzzzzzzzzzuuuuuuuuuuuuuuuuuupppppppppppp????????? :-)
>>
>
>What?!?! You did not buy into that idea that pings on your web site will
>bring you
>money?!?!
Pings? You mean "hits" - now that I (think I understand) what you've been
saying, you've got to be kidding me, LOL. Hits are like window shoppers, though
the attention is nice, unless they come into the store and actually _buy_
something the only thing that will be pinging is my ping pong paddle (which I
haven't played that game in years). You know that old saying, "if I had a
nickle for every complement I get ;-)... But thanks for your well wishes
though.
Anyway, since you live in somewhat reality, it was just a loose
>association (pun, maybe). Oh well . . . .
>
>Ciao!
>
>Gordon Moat
>Alliance Graphique Studio
><http://www.allgstudio.com>
Ok, perhaps the association was little too loose for me to get the first time
around. I've heard of "surreality" but what is "somewhat reality" though?...
Lewis Lang
>Message-id: <1e9ffba8.030...@posting.google.com>
Hi Candice:
Let me see if I got you right... What you are saying is something that sounds
like 4 words to me "high resolution stochastic printing" (perhaps w/o the
printing aspect as such.
Also, I've heard of quantum theory but I'm not quite understanding
"quantization" and "negative light". Could you explain both terms in layman's
language for me? Thanks.
Also, negative light almost sounds like you are doing somekind of wavelength
cancellation through phase shifting to cancel out whatever wavelengths you
have, this is remarkably similar in my mind to what's doen in the
anti-reflective coatings of lenses to eliminate flare by shifting light 180
degrees? out of phase to cancel out stray/non-image forming light. Am I on the
right track here or are you suggesting something else?
Also, the "sub pixels" you describe:
- What exactly are sub-pixels, do they exist merely/only prior to output into
real pixels?
- Do sub pixels actually carry extra (very tiny?) amoutns of information or are
they just some kind of randomized digital noise and/or re-shifting of the
information already there to avoid moire patterns and other defects?
All this sounds quite fascinating... :-)
TIA
Lewis Lang
>Message-id: <3E429C32...@videotron.ca>
I see one of your eyes is slanting shut and none of the above may work for you,
have you considered opening and smelling a foil packed Kodachrome 40 Super 8
movie film, works better than walnuts ;-)
Per Inge Oestmoen
> It's not easy to see, since you have to look beyond the noise introduced by
> the scanner. But the tones on the digital shot are flat, while on the film
> shot, they are more realistic. Look at the base of the tree, and some of the
> patterns on the dig shot are merely fields of a like color, while on the
> film shot, you can see gradations. I'd like to see a better scan, as well.
There is no doubt that the image to the left in
http://www.pbase.com/image/11766189 is the one that contains the most
detail. If it had been properly post processed it would have been even
more superior to the other image.
However, there are several possible sources of error. For example,
what lens was used for the digital camera at right, was it a prime
lens or was it a cheap zoom? Was the image recorded in RAW/TIFF or in
JPEG?
I own and use a D60 myself, and my experience is that only in very few
instances can it be beaten by a film scan by a consumer-grade scanner
like the Nikon LS-40 or LS-4000. Most of the time the D60 yields
superior results.
--
Per Inge Oestmoen, Norway
http://www.coldsiberia.org/
Annika1980
>There is no doubt that the image to the left in
>http://www.pbase.com/image/11766189 is the one that contains the most
>detail. If it had been properly post processed it would have been even
>more superior to the other image.
If you consider grain and noise to be "detail" then I agree.
>
>However, there are several possible sources of error. For example,
>what lens was used for the digital camera at right, was it a prime
>lens or was it a cheap zoom? Was the image recorded in RAW/TIFF or in
>JPEG?
Same lens was used for both shots (the incredible 28-135 IS). The digital shot
was taken in RAW format and then converted to JPG.
Dennis O'Connor
side by side with the best digital ink print(s), visually... Running the
silver negative or slide through a scanner and then showing the images on
the usual computer monitor over the internet is not going to change anyone's
mind <some minds are incapable of change, of course>...
I have to say that I have been impressed with the image qualities of the top
end cameras from Canon, Nikon, et. al... It will be interesting to see what
the full size CCD chip with >10meg images, will have to offer us in the next
generation of cameras from the big three or four <depending on how you
count>, over the next couple of years...
Apparently Minolta is aggressively coming out to play... And, Canon is
obviously offloading it's current inventory of 5meg cameras before releasing
their new flagship... Nikon has an 11 meg camera out now, and I have not
heard what their rumored plans are for Photokina..
Denny
"Annika1980" <annik...@aol.com> wrote in message
news:20030207100751...@mb-md.aol.com...
Annika1980
Dennis O'Connor
frantically beating the drum of the "true religion"...
Denny
"Candice H. Brown Elliott" <cell...@clairvoyante.com> wrote in message
news:1e9ffba8.030...@posting.google.com...
Dennis O'Connor
that your lenses just got cut back to 1:1 ratio...
"Annika1980" <annik...@aol.com> wrote in message
news:20030207114253...@mb-fe.aol.com...
Simon Stanmore
Canon's the one with a 11MP body - their 1DS. Kodak will soon be releasing
the somewhat overdue 14N with 13.9MP - that has a Nikon F-mount. And the
camera that Canon is currently 'offloading' is the D60 - a 6.3MP model.
The last I heard from Minolta on DSLR's was a wait-and-see policy.
Pentax will have a new DSLR on show at the PMA.
Nikon has not announced any true double-figure MP camera yet - a few weeks
from now (at the PMA) they may well do of course.
Canon is going to announce at least one - possible two - new DSLR's as well,
and I suspect they'll be priced very keenly indeed (for a DSLR that is).
From a number of reports I've read over the past several months I can only
conclude that Canon's Japanese directorship see the DSLR market as a huge
oppurtunity to capture market share - their new models will be priced to do
just that I'm sure.
After the PMA and the subsequent reliable review sources have spoken I'm
buying into digital - I may well switch brands for this. Digital is the
(very immenent) future as far as 35mm-type work is concerned, and I'll be
going with the marque that I consider to be implementing the technology in a
fashion that I consider the most effective and long-term ... We are
photographing through a very interesting time ...
Simon
Candice H. Brown Elliott
> >Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
> >From: cell...@clairvoyante.com (Candice H. Brown Elliott)
> >Date: Thu, Feb 6, 2003 11:39 PM
> >Message-id: <1e9ffba8.030...@posting.google.com>
> >
> >cont...@aol.comnospam (Lewis Lang) wrote in message
> news:<20030206094235...@mb-fn.aol.com>...
> >>
> >>
> >
"quantization" means putting things into "counting numbers" instead of
the continous world of analog.
"negative light" is a physical impossibility... save as a mathmatical
construct in coherent optics (read: holography). I mostly mentioned
it as a completeness point... mostly it is a strawman, something we
can't do at the moment with our limited technology.
>
> Also, negative light almost sounds like you are doing somekind of wavelength
> cancellation through phase shifting to cancel out whatever wavelengths you
> have, this is remarkably similar in my mind to what's doen in the
> anti-reflective coatings of lenses to eliminate flare by shifting light 180
> degrees? out of phase to cancel out stray/non-image forming light. Am I on the
> right track here or are you suggesting something else?
You understand me perfectly... one can only get "negative light" in a
holographic reconstruction. It is theoretically possible to create a
good photoprinter that really does use the Shannon-Nyquist sinc
recontruction filter using an LCD illuminanted by a single laser,
followed by a holographic "blur" filter that implements the sinc
function optically. No one has actually done this to my knowledge...
but... frankly I think someone graduate student should try it
sometime.
My main point was that we simply are not actually doing proper
theoretical reconstructions in our real world displays or film
printers... and this is why digital photography doesn't work as well
as it should.
>
> Also, the "sub pixels" you describe:
>
> - What exactly are sub-pixels, do they exist merely/only prior to output into
> real pixels?
The "subpixels" I am refering to are the colored dots on a flat panel
screen... or the color pixels of a projector, offset from one another
by one-half pixel in the diagonal directionl. (That is what my
upcoming paper is on...) by using the color subpixels, one gets more
points with which to reconstruct the digital image... just as film
uses separate grains on each of the film layers.
>
> - Do sub pixels actually carry extra (very tiny?) amoutns of information or are
> they just some kind of randomized digital noise and/or re-shifting of the
> information already there to avoid moire patterns and other defects?
If the display is designed right, additional detail can be obtained...
note that the commonly used RGB Stripe arrangement used in LCDs is not
designed right... and one only gets the shifts that reduce moiré as
you correctly noted. However, my company has developed a novel
subpixel arrangement, the PenTile Matrix(tm), that one may indeed get
extra detail out of... to explore that further, one should visit our
website
>
> All this sounds quite fascinating... :-)
>
> TIA
>
> Regards,
>
> Lewis
>
-- Towards a clearer vision of the future --
Annika1980
>Digital is the
>(very immenent) future as far as 35mm-type work is concerned, and I'll be
>going with the marque that I consider to be implementing the technology in a
>fashion that I consider the most effective and long-term .
So have you decided which Canon body you'll be buying?
Simon Stanmore
I've a nice comprehensive Nikon system at the moment and yes ... Canon is
naturally the other option. I'm waiting for image samples and detailed
reviews of the new releases before I decide (I'm in no great rush). It's a
system thing as well - Nikon's latest DX lens idea is something I really
don't like. The F-mount is already littered with incompatabilities and this
new line of lenses is just pushing the whole mess too far for me now
Simon
Per Inge Oestmoen
> If you consider grain and noise to be "detail" then I agree.
I do not, but I see that the film scan shows more detail in the bushes.
>>However, there are several possible sources of error. For example,
>>what lens was used for the digital camera at right, was it a prime
>>lens or was it a cheap zoom? Was the image recorded in RAW/TIFF or in
>>JPEG?
> Same lens was used for both shots (the incredible 28-135 IS). The digital shot
> was taken in RAW format and then converted to JPG.
Then a major source of error is found. That is a low-grade lens when
compared to the 70-200 2.8 L and primes. If you repeat the experiment
with a prime or with the high-grade zoom the D60 image will be vastly
better.
ThomasH
>
> Hi Thomas:
>
> >Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
>
> >If stored in regular room temperature conditions, E6 film can not
> >stay in good shape for over a decade or two. Not to talk about
> >the possible impact of fungi, bacteria and dust. All my slides
> >from the 80ties look a bit faded out, especially highlights
> >are almost without detail.
> >
>
> Newest data from Henry Wilhelm suggests dark storage life of 40 years for
> Fujichromes and 120+ years from E-6 Ektachromes. I haven't had problems of
> either fading on me yet and I 've been using both since about 1983...
Thanks, I will browse than the web site! The book is already dated
and the web page was down for a long time. Latest data which I had
was in favor of Fujichrome E6 materials.
>
> And you are tied to a life time of copying and recopying huge amounts of data
> on a non-archival unstable storage medium that may or may not exist a short
> time down the road of life...
The attractive here is, that this is not an effort at all, really.
Using the external drive storage for example, you just copy over
the entire directory tree with a few mouse clicks. Let the drive
rattle than in the background and mind your other tasks. Besides
I am accustom to a daily and monthly backup anyway.
>
> >For example I found my old 1st generation DAT tape with a copy
> >of my Phd theses. My DAT DSS3 Seagate Scorpio (24gig capacity)
> >could read that tape recorded in mid 1980ties without a problem.
> >The tape was almost 18 years old. Of course a bit luck helped.
> >I used back than a PC-DOS program called Arcada Backup to store
> >my data. Arcada was than purchased by Seagate and Seagate Software
> >made a contract with Microsoft to deliver this very backup as the
> >W95 backup, what is still to this day a valid Microsoft format.
> >
>
> Had you waited too long, perhaps into the mid 90's you might not have been able
> to read that file. Also, relying on the "generosity" of Microsoft or any other
> company to keep alive a way(s) of reading your data is a non-prudent gamble at
> best.
I know, you are fully correct. The choice of a format might be an
issue. Nowadays the MS backup is probably as future safe format as
it gets!
The storage of files to another drive does not use any particular
storage format at all. Its just a MS file system, and these will
never be abandoned.
>
> Not yet mentioned problem? All we fever do here is talk about the obsolecsnece
> of digital storage drives/media when the subject of archival longevity of film
> vs. digital media comes up on this ng. The Fujitsu magnetic optical
> drives/media? are rated at about 10 years on their website, these seem to be
> some of the longer lasting media/drives and have been around for years though
> they are don't have "Zip disc" and "CD/DVD" popularity. I aggre in theory w/
> what you say about always choosing the most popuar media/drives however I am
> leaning towards less popular and longer lasting (both archivally/longevity and
> in terms of how long the company/format has been in business selling like
> products to the public) digital media such as magnetic optical. Frankly I feel
> that all digi solutions for both archival longevity and
> brand/company/drive/media longevity in terms of sales availability of the
> product (or similar products years down the line to read the older products)
> are piss poor to poor at best at this time. I am looking at external hard
> drives also as a _longer_ but not long term solution not only for its cheapness
> but for its longevity. My own computer's hard drive is going on 9 years, though
> I have no info on how long other external hard drives as well as removable
> media last, I wish Henry Wilhelm would do something more on digital sotrage
> media archival longevity as well as an update to his film/paper longevity
> info...
The optomagnetic RW CD's are not reliable at all, period! Nobody
stores data on them, unless ignorance is the reason.
The durability of the magnetic storage is known already from the
condition of the countless large reel tapes dated 196x or 197x.
Otherwise my instinct says: the smaller the data density, the
more molecules memorize the data bits and the better is a chance
for a good durability.
While speaking about the CD's and their expected life span, let
me also add that the audio CD's store information with redundant
data blocks allowing automatic damage repair to an extend.
Basics of such error-correcting codes are known for a long time
already and they have been investigated in the new discipline
called Information Theory. One of the simplest codes of such
kind is the Hamming code.
Such principles were applied from the very beginning for the
storage of the audio signal because the CD was first used as
an audio media, and became only later adopted by the computer
industry. First level of repair: Digital, lossless, Second
level or repair: Analog interpolation. This makes the CD's so
resilient to fingerprints and even to some minor scratches in
an amazing amount.
I must admit that I completely do not know if these principles
are being used similarly to protect digital data, the CDFS or
Joliet for example.
For photography storing we would surely need a storage format
allowing such two-level recovery: First digitally, a complete
recovery possible. Appearance of damages of such kind would be
a warning bell: "Copy your data, media is deteriorating." The
second level would be some sort of interpolation, which would
deliver "restored" images with some degree of deterioration.
[...]
>
> Just back up the film to digital for copies as needed (_not every image needs
> to be scanned_, just the best/most useful) or have optical or digital dupes
> made of only the very best of theimages onto film again.
The point here is though that the emulsion in every film
was exposed to the developing liquids, which in turn were
active and caused complex chemical reactions. No matter how
well you will flush and clean the film, some sort of a rest
reaction will continue to happen. Even the chemicals from
our surrounding air may diffuse into the emulsion and cause
some further changes over a time. All materials age and
deteriorate somehow.
I store my slides in regular slide magazine boxes, by Leica
or Reflecta. I like to assume that the used plastics for these
boxes were are well chosen for the film storage (no fume
emission etc.) These boxes are being kept in a dark closet in
a dry room, albeit in a room temperature. Its just a regular
home storage, as good as it gets without cooling.
All my slides from the 80 ties show a significant loss of
details in the highlights: Kodak, Fuji and Agfa alike. This
is just a subjective observation without trying to analyze
if Fuji E6 films are indeed so much more durable than the
others. Maybe the recent advances in the E6 materials were
so great that these old slides simply represent the past
technology and did not aged so much? I do not know, but I
know that my pixels in raw tiff's will not age ! :-)
Thomas.
Gordon Moat
> Thanks for the informative tips, Gordon. What/which flaws would doing selective
> Gaussian blur/etc. create that I should be on the look out for
> (over-softening?, some kind of halo effect of certain colors?, others?)?
The easiest one to see is that one colour channel will have some parts of the
colour spread a little too far. This would look a little like colour fringing, or
a colour halo, or edge at some detail areas of the image. The other thing is that
you can just lose sharpness, or lose detail. You could enlarge to 400% on your
screen, and examine those areas, but a print would be the best way to judge it.
> I don't understand why the back of a digital camera can't have the equivalent
> of a flip down and/or expandable screen, maybe 4-6 inches diagonal, which would
> go a long ways towards eliminating the problems of judging focus and
> composition on the current tiny screens...
Probably battery power consumption, which is already high enough. Another aspect
would be increased costs. The separate display idea does not seem like a bad
option, and an extra cost that some photographers may not feel too bad about
incurring. Another option would be to drag a laptop to the shooting location, but
that is just more to carry. Come to think of it, perhaps Polaroid proofs still
have some life in them after all . . . .
> Light emitting paper, sounds interesting, I wonder if a multi-layered version
> of that could be used for cheap high quality (in the far future display systems
> not just as a light source but to somehow display images like LCD displays do
> now. Imagine a 70 inch diagonal screen w/ (pardon the pun) "paper thin"
> speakers that weighs about 5 pounds and that you could roll up and hang on your
> wall ;-)...
One of the hoped for aspects of this was flexible displays. It may be possible to
have a window shade type of display, and it would need very little storage area.
The last thing I read about these is that the colour accuracy and contrast are not
good enough for computer use . . . yet.
> What exactly do you mean by "careful lighting" - lighting that is w/i the
> illumination levels and contrast range of the sensor to eliminate noise,
> multiple light set-ups that include key, fill, background, hair and/or rim
> lights? Lights used w/ gels and barn doors and grids to do complex lighting
> effects equivalent to the best still photography and movie lighting set ups?
> All of the above and/or something else?
>
Sure . . . contrast range mostly. Colour quality (or accuracy) is probably
slightly less of an issue, except with human skin tones. Anyway, it takes just a
little more effort to set up properly than do the same for film shooting. This is
not a huge issue, but it is a consideration.
>
> >> What qualitiy(s) do they have that are they
> >> indistinguishable from film in image quality?
> >
> >Colour accuracy, though as I stated only with careful lighting.
>
> So careful lighting to you means white balancing for accurate/"catalog"
> neutrality of color rendition?
Actually, I would think that setting the white point, and setting the black point,
would get an even range. I do not think there are that many higher contrast photo
shoots making it into print, at least I do not see many examples of them. Colour
accuracy can be very important with product photography, though a good graphic
designer, or printing place, could get the final results closer if needed. Pantone
colour systems are good for this.
The downside is that the speed advantage direct digital gives in an immediate
file, could be negated by more labour time spent in post processing the images. It
is easier to get it right the first time, than to correct the inadequacies later.
>
>
> You cannot
> >judge
> >detail without being at the same place as the photographer to observe when
> >and how
> >he photographed the image.
> >
>
> This aspect of lighting seems more a function of contrast ratios...
Sure . . . film still has a greater contrast ratio capability than direct digital
capture. However, this can also be an issue with scanning, though there are ways
to improve that. I wish I could tell you what I do with some high contrast scans
to get the greatest contrast range, but it takes a long time to write that much.
Besides, that is one of my specialities that I get paid to perform. I mention
brief bits about doing double scans, blending of channels and/or layers, tricks
with the maximum, minimum, and highpass filters, double scanning with
overexposing/underexposing on the scanner, etc. Anyway, I think I only have one
image like that posted at the moment, since it is a better technique for printed
images. Sorry for not explaining lengthy processes, but take a look at:
<http://www.allgstudio.com/gallery/corporate/Amtrack_LAoldyard.jpg> This is an
image of the Amtrak yard in Los Angeles. The large rig in the centre of the image
is for turning a train engine around. The detail in the shadows of the train on
the left was tough to bring out, without completely blowing out the nice sky and
clouds. Getting the cloud detail nice, without the loss of almost all shadow
detail, was beyond the dynamic range of the scanner. Using VueScan with multiple
passes still did not give this much contrast range, nor detail. There were many
"tricks" done on this one image to get the details closer to the original chrome.
It may be slightly dark on gamma still, if you are viewing on a Windows based
monitor, though a MacOS monitor gamma should be okay. I am in no way claiming it
is a great image, but it held some interest for me, and fits with some others in
my photo gallery (in other words, this is not a portfolio image).
> Pings? You mean "hits" - now that I (think I understand) what you've been
> saying, you've got to be kidding me, LOL. Hits are like window shoppers, though
> the attention is nice, unless they come into the store and actually _buy_
> something the only thing that will be pinging is my ping pong paddle (which I
> haven't played that game in years). You know that old saying, "if I had a
> nickle for every complement I get ;-)... But thanks for your well wishes
> though.
Yes . . . I was kidding . . . .
> Ok, perhaps the association was little too loose for me to get the first time
> around. I've heard of "surreality" but what is "somewhat reality" though?...
Sort of like "Never, never Land" . . . .
Annika1980
>> Same lens was used for both shots (the incredible 28-135 IS). The digital
>shot
>> was taken in RAW format and then converted to JPG.
>
>Then a major source of error is found. That is a low-grade lens when
>compared to the 70-200 2.8 L and primes. If you repeat the experiment
>with a prime or with the high-grade zoom the D60 image will be vastly
>better.
I kinda doubt it. You'd have to look pretty closely to see any real
differences between those two zooms. The Super 70-200 f/2.8L is slightly
sharper, but not "vastly better."
Even you agreed that the incredible 28-135 IS did a good job with my cat pic.
I don't think the "L" lens would've been much sharper.
Lewis Lang
>Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
>From: ThomasH tho...@coco.net
>Date: Fri, Feb 7, 2003 11:42 PM
>Message-id: <3E444468...@oracle.com>
>
>Lewis Lang wrote:
>>
>> Hi Thomas:
>>
>> >Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
>[...]
>>
>> >If stored in regular room temperature conditions, E6 film can not
>> >stay in good shape for over a decade or two. Not to talk about
>> >the possible impact of fungi, bacteria and dust. All my slides
>> >from the 80ties look a bit faded out, especially highlights
>> >are almost without detail.
>> >
>>
>> Newest data from Henry Wilhelm suggests dark storage life of 40 years
>for
>> Fujichromes and 120+ years from E-6 Ektachromes. I haven't had problems
>of
>> either fading on me yet and I 've been using both since about 1983...
>
>Thanks, I will browse than the web site! The book is already dated
>and the web page was down for a long time. Latest data which I had
>was in favor of Fujichrome E6 materials.
>
You're welcome. Plus, if you ever have a chance to get to a library that has
back issues of Pop Photo check out the April '99 issues article "HOW LONG WILL
YOUR PRINTS LAST"? - it has a table w/ updated film longevity figures. Wilhelm
hasn't done a major mag. update of his 1990 "GOING... GOING... GONE!..."
article, since about, well, 1990 ;-) and my guess would be that his 'Image
Permanence' book hasn' been update sine ''92/93? or whenever it came out - a
real pity...
>
>>
>> And you are tied to a life time of copying and recopying huge amounts
>of data
>> on a non-archival unstable storage medium that may or may not exist a
>short
>> time down the road of life...
>
>The attractive here is, that this is not an effort at all, really.
>
>Using the external drive storage for example, you just copy over
>the entire directory tree with a few mouse clicks. Let the drive
>rattle than in the background and mind your other tasks. Besides
>I am accustom to a daily and monthly backup anyway.
>
Yes, you're right about this from one aspect, w/ external and extra internal
harddrives, this is probably very easy to do/update, however <BEGIN MR. T
VOICE/"I pity the fool who has a multitude of images on many CDs"/END MR. T.
VOICE> :-). Still, digi requires constant backing up throughout your life
(regardless of how often you do it) and then after your gone, if you care about
its longevity after you live, you probably have about 5-10 years max before
that info may be lost if not recopied, whilst Kodachrome, Ektachrome and
possibly Fujichrome (we have yet to see how real the projected figures of
archival life willpan out in reality) will most likely be little diminished or
easily recoverable (whatever dye changes have taken place) through scanning and
reoutput on a film recorder (or somesuch) to film, Evercolor or other archival
paper process or some other more stabile future medium.
>
>>
>> >For example I found my old 1st generation DAT tape with a copy
>> >of my Phd theses. My DAT DSS3 Seagate Scorpio (24gig capacity)
>> >could read that tape recorded in mid 1980ties without a problem.
>> >The tape was almost 18 years old. Of course a bit luck helped.
>> >I used back than a PC-DOS program called Arcada Backup to store
>> >my data. Arcada was than purchased by Seagate and Seagate Software
>> >made a contract with Microsoft to deliver this very backup as the
>> >W95 backup, what is still to this day a valid Microsoft format.
>> >
>>
>> Had you waited too long, perhaps into the mid 90's you might not have
>been able
>> to read that file. Also, relying on the "generosity" of Microsoft or any
>other
>> company to keep alive a way(s) of reading your data is a non-prudent gamble
>at
>> best.
>
>I know, you are fully correct. The choice of a format might be an
>issue. Nowadays the MS backup is probably as future safe format as
>it gets!
>
In an era of corporate downsizing, possible war and great economic uncertainty
even "the gold standards" are not so golden anymore, hence my healthy mistrust
of a medium such as digital whose economic heartbeat is based on planned
obsolecence.
>The storage of files to another drive does not use any particular
>storage format at all. Its just a MS file system, and these will
>never be abandoned.
>
I'm not sure what you mean by file system - operating system, file format,
program? I have heard proposals for non-hirchal computer operating systems,
everything is subject to change and 50 years from now, if we are fortunate
enough to still be alive, we might have machines that might not even rsemble
modern day computers much less be compatable with them and their operating
systems.
The Fujitsu is a magnetic optical drive that is more of a Zip disc kind of
cartridge configuration. Regular CDRs use dyes only, I thought, written on and
read by lasers. What format is the optomagnetic RW CDs you are talking about? -
I thought CD RWs use mereyl a (unstable ) dyeset and not some kind of magnetic
optical configuration...
>The durability of the magnetic storage is known already from the
>condition of the countless large reel tapes dated 196x or 197x.
>Otherwise my instinct says: the smaller the data density, the
>more molecules memorize the data bits and the better is a chance
>for a good durability.
>
>
>While speaking about the CD's and their expected life span, let
>me also add that the audio CD's store information with redundant
>data blocks allowing automatic damage repair to an extend.
>
Audi CDs are also much more archival because their data dimple pits are pressed
from glass masters as opposed to being recorded on dyes whose
contrast/readability may fade w/i just a couple of years...
>Basics of such error-correcting codes are known for a long time
>already and they have been investigated in the new discipline
>called Information Theory. One of the simplest codes of such
>kind is the Hamming code.
>
>Such principles were applied from the very beginning for the
>storage of the audio signal because the CD was first used as
>an audio media, and became only later adopted by the computer
>industry. First level of repair: Digital, lossless, Second
>level or repair: Analog interpolation. This makes the CD's so
>resilient to fingerprints and even to some minor scratches in
>an amazing amount.
>
Were that the digital data CDRs and CD RWs we all use were as archival as the
audio CDs - a real pity...
>I must admit that I completely do not know if these principles
>are being used similarly to protect digital data, the CDFS or
>Joliet for example.
>
>For photography storing we would surely need a storage format
>allowing such two-level recovery: First digitally, a complete
>recovery possible. Appearance of damages of such kind would be
>a warning bell: "Copy your data, media is deteriorating." The
>second level would be some sort of interpolation, which would
>deliver "restored" images with some degree of deterioration.
>
I don't have the technical knowledge to tell you whether this is possible or
not, but what I do know is that a deteriorated film image can easily be
adjusted for faded color and missing pieces of data due to the emulsion flaking
off/spots can at least be cloned/rubber stamped over to give the impression of
a flawless print. Once a digital file loses some of its data does it not become
completely corrupted or am I missing something here?
>
>[...]
>>
>> Just back up the film to digital for copies as needed (_not every image
>needs
>> to be scanned_, just the best/most useful) or have optical or digital
>dupes
>> made of only the very best of theimages onto film again.
>
>The point here is though that the emulsion in every film
>was exposed to the developing liquids, which in turn were
>active and caused complex chemical reactions. No matter how
>well you will flush and clean the film, some sort of a rest
>reaction will continue to happen. Even the chemicals from
>our surrounding air may diffuse into the emulsion and cause
>some further changes over a time. All materials age and
>deteriorate somehow.
>
True, see comments above.
>I store my slides in regular slide magazine boxes, by Leica
>or Reflecta. I like to assume that the used plastics for these
>boxes were are well chosen for the film storage (no fume
>emission etc.) These boxes are being kept in a dark closet in
>a dry room, albeit in a room temperature. Its just a regular
>home storage, as good as it gets without cooling.
>
Might be a good idea, instead of assuming, to contact the companies in
questionto ask them exactly what _are_ the materials their slide boxes are made
of. Then find out from Wilhelm's, Light Impressions, and other websites (and
google.com's archive, whether these material will out gas or react in some
other way w/ your slides. If they do, it would be wise to invest in non-PVC
archival slide pages or some similar non-out gassing, non-reactive archival
storage method. Never assume anything, especially when it comes to your own
precious slides.
>All my slides from the 80 ties show a significant loss of
>details in the highlights: Kodak, Fuji and Agfa alike.
My, oh, let me take a stab at it, wild off the top of my head guess is that
either the boxes you have been storing them in and/or any wood/paint/etc.
conditions that may outgassing around them may have effected them - 20 some odd
years time is _way_ too short for slides to start significantly fading (3M
slides may be another story, however, this brand is known to be piss poor for
slide archival longevity).
This
>is just a subjective observation without trying to analyze
>if Fuji E6 films are indeed so much more durable than the
>others. Maybe the recent advances in the E6 materials were
>so great that these old slides simply represent the past
>technology and did not aged so much?
Between '78? and about 82/83? there might have been some changes in the E-6
process which would significantly affect archivability (no to not as much
stabilizer used and/or some other factor(s)?) as well as emulsion imporvements
themselves. But I have seen older Kodacolor negs that lasted longer so I really
wonder if it has anything to do w/ the film emulsion and something more to do
with processing and/or storage medium (your slide boxes) or storage
environmental conditions (out-gassing from wood/plastic/paint/etc. near bye...
I do not know, but I
>know that my pixels in raw tiff's will not age ! :-)
>
>Thomas.
No, they will just deteriorate and become unreadable unless readily and
constantly copied every now and then, unfortunately ;-(... Look Thomas, I'm
very sorry about what happened to your slides but digital is merely a stop gap
solution, not a total archival solution, just a part of it, you should really
consider, if/when you have the money, keeping different forms of the best
slides in your collection in different locations (to safe guard against
damage/theft,etc.) and making at least two sets of duplicate slides (in
addition to whatever sets of backups you have done/scanned digitally off of
your best slides) of your best images (optical dupes (corrected w/ whatever
filtration/etc. could help optically to restore color balance/density/contrast)
and/or adjusted for color/density digital outputted to a film recorder) and
possibly store one set of copy slide dupes in the freezer (according to
archival storage methods that will lock out any moisture (consult Wilhelm/Kodak
and possibly others for more info on how to do this _properly_ - you can't just
shove slide pages in a baggy and hope everything will come out OK/unaffected by
moisture) and store another dupe slide set undr normal temp/humidity in some
real archival slide pages and not those boxes you've been using. Keep all
slides away from furniture, wood, mothballs, paint and other solvents that
might out gas and cause chemical changes. An air purifier (in addition to
central air/heating to maintain constant temperatures might also not hurt. You
also might want to store your affected/faded slides separately from your dupes
(just being extra cautious here). At the very least, check on the materials
used in your slide boxes and when you have the cash, consider transferring
these slides (and all other dupe slides, even if at a different location) into
archival non-reactive, non out-gassing slide pages. You've spent I'm sure a lot
of time and money on your images. it would be a shame, if for the lack of a
little more time and money (by using archival storage materials/conditions and
dupes of the best lsides) if you would not at least consider extending whatever
time/life your slides have left.
Lewis Lang
>Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
>From: Gordon Moat mo...@attglobal.net
>Date: Sat, Feb 8, 2003 12:30 AM
>Message-id: <3E444F83...@attglobal.net>
>
>Lewis Lang wrote:
>
>> Thanks for the informative tips, Gordon. What/which flaws would doing
>selective
>> Gaussian blur/etc. create that I should be on the look out for
>> (over-softening?, some kind of halo effect of certain colors?, others?)?
>
>The easiest one to see is that one colour channel will have some parts of
>the
>colour spread a little too far. This would look a little like colour fringing,
>or
>a colour halo, or edge at some detail areas of the image. The other thing
>is that
>you can just lose sharpness, or lose detail. You could enlarge to 400% on
>your
>screen, and examine those areas, but a print would be the best way to judge
>it.
>
Makes sense.
>> I don't understand why the back of a digital camera can't have the
equivalent
>> of a flip down and/or expandable screen, maybe 4-6 inches diagonal, which
>would
>> go a long ways towards eliminating the problems of judging focus and
>> composition on the current tiny screens...
>
>Probably battery power consumption, which is already high enough. Another
>aspect
>would be increased costs. The separate display idea does not seem like a
>bad
>option, and an extra cost that some photographers may not feel too bad about
>incurring. Another option would be to drag a laptop to the shooting location,
>but
>that is just more to carry. Come to think of it, perhaps Polaroid proofs
>still
>have some life in them after all . . . .
>
Uggggggh (shrugs like Lurch on the Adams Family) don't mention Polaroids (I had
to trade in my Minox to be able to afford a couple of boxes of B&W 4x5"
Polaroids for an old Brooks (Institute of Photography) class!)
>> Light emitting paper, sounds interesting, I wonder if a multi-layered
>version
>> of that could be used for cheap high quality (in the far future display
>systems
>> not just as a light source but to somehow display images like LCD displays
>do
>> now. Imagine a 70 inch diagonal screen w/ (pardon the pun) "paper thin"
>> speakers that weighs about 5 pounds and that you could roll up and hang
>on your
>> wall ;-)...
>
>One of the hoped for aspects of this was flexible displays. It may be possible
>to
>have a window shade type of display, and it would need very little storage
>area.
>The last thing I read about these is that the colour accuracy and contrast
>are not
>good enough for computer use . . . yet.
>
Sounds neato, great for the den or the airplane where they could be pulled down
like the old ceiling mounted pull down/slide/movie screens of yore..
SNIP
>The downside is that the speed advantage direct digital gives in an immediate
>file, could be negated by more labour time spent in post processing the
>images. It
>is easier to get it right the first time, than to correct the inadequacies
>later.
>
Agreed.
>>
>>
>> You cannot
>> >judge
>> >detail without being at the same place as the photographer to observe
>when
>> >and how
>> >he photographed the image.
>> >
>>
>> This aspect of lighting seems more a function of contrast ratios...
>
>Sure . . . film still has a greater contrast ratio capability than direct
>digital
>capture. However, this can also be an issue with scanning, though there
>are ways
>to improve that. I wish I could tell you what I do with some high contrast
>scans
>to get the greatest contrast range, but it takes a long time to write that
>much.
>Besides, that is one of my specialities that I get paid to perform. I mention
>brief bits about doing double scans, blending of channels and/or layers,
>tricks
>with the maximum, minimum, and highpass filters, double scanning with
>overexposing/underexposing on the scanner, etc.
What I suspected anyways, layering, multiple scans optomized for highlights,
shadows and/or midtones, high pass filter for sharpening (learned about this at
a Photo Expo seminar), etc. If you ever have the time and/or desire in the
future you might want to think about posting a web page on these techniques and
how each affects the exact same image... :-)
This is definitely a shot that would have benefited from combining a tripoded
bracket of 3 different exposures of the same image (for shadows, midtones and
brighteest hilights). Something to consider... Its a bitch trying to get the
most out of slide film as most scanners poop out in the shadows (unless you are
doing some high dynamic range drum scans) and detail lost in the hilights is,
well, lost. Conventional masking (using a moderately out of focus B&W maks of
the same image) then printing a conventional pirnt (or slide dupe) then
scanning either of those in for further adjustment might have overcome the
limitations of trying to get the whole range scanned in/manipulated from one
locked in density range on a single slide, however, when/if time/opportunity
permit, on another image perhaps, unless this image is especially important to
you, a 3 image bracketed exposure combined in PShop might work best.
>> Pings? You mean "hits" - now that I (think I understand) what you've been
>> saying, you've got to be kidding me, LOL. Hits are like window shoppers,
>though
>> the attention is nice, unless they come into the store and actually _buy_
>> something the only thing that will be pinging is my ping pong paddle (which
>I
>> haven't played that game in years). You know that old saying, "if I had
>a
>> nickle for every complement I get ;-)... But thanks for your well wishes
>> though.
>
>Yes . . . I was kidding . . . .
>
>> Ok, perhaps the association was little too loose for me to get the first
>time
>> around. I've heard of "surreality" but what is "somewhat reality" though?...
>
>Sort of like "Never, never Land" . . . .
>
>Ciao!
>
>Gordon Moat
>Alliance Graphique Studio
><http://www.allgstudio.com>
This almost sounds like an insult but I'll let it pass bye since I'm not sure
exactly how you meant it. I live in this reality (not some imaginative/fantasy
one) and have a pretty clear view of it most times even when I'm not shooting
it w/ my Zeiss lenses ;-), however I may, to others at least, seem to have an
offbeat/different view of it - Never Never Land? - you must have me confused w/
Michael Jackson ;-)
Lewis Lang
BIG SNIP OF PREVIOUS POST
>> Let me see if I got you right... What you are saying is something that
>sounds
>> like 4 words to me "high resolution stochastic printing" (perhaps w/o
>the
>> printing aspect as such.
>>
>> Also, I've heard of quantum theory but I'm not quite understanding
>> "quantization" and "negative light". Could you explain both terms in
layman's
>> language for me? Thanks.
>
>"quantization" means putting things into "counting numbers" instead of
>the continous world of analog.
>
Makes perfect sense.
>"negative light" is a physical impossibility...
Possibly what a black hole gives off in an alternative universe? Or perhaps an
anti-matter equivalent to light? - why not anti-photons? (are there such
things? - I believe there are anti-portons and other anti-sub-atomic particles
so why not "anti-light" unless the negative light you are talking about is just
18o depgree wavelength phase shifting cancellation of light) :-)
save as a mathmatical
>construct in coherent optics (read: holography). I mostly mentioned
>it as a completeness point... mostly it is a strawman, something we
>can't do at the moment with our limited technology.
>
Perhaps it can be mathematically/electronically simulated if not actually
possible in physical reality?
>
>>
>> Also, negative light almost sounds like you are doing somekind of wavelength
>> cancellation through phase shifting to cancel out whatever wavelengths
>you
>> have, this is remarkably similar in my mind to what's doen in the
>> anti-reflective coatings of lenses to eliminate flare by shifting light
>180
>> degrees? out of phase to cancel out stray/non-image forming light. Am
>I on the
>> right track here or are you suggesting something else?
>
>You understand me perfectly... one can only get "negative light" in a
>holographic reconstruction.
Please explain how negative light is achieved in holography, if you will...
Thanks. :-)
It is theoretically possible to create a
>good photoprinter that really does use the Shannon-Nyquist sinc
>recontruction filter using an LCD illuminanted by a single laser,
>followed by a holographic "blur" filter that implements the sinc
>function optically. No one has actually done this to my knowledge...
>but... frankly I think someone graduate student should try it
>sometime.
>
>My main point was that we simply are not actually doing proper
>theoretical reconstructions in our real world displays or film
>printers... and this is why digital photography doesn't work as well
>as it should.
>>
>> Also, the "sub pixels" you describe:
>>
>> - What exactly are sub-pixels, do they exist merely/only prior to output
>into
>> real pixels?
>
>The "subpixels" I am refering to are the colored dots on a flat panel
>screen... or the color pixels of a projector, offset from one another
>by one-half pixel in the diagonal directionl. (That is what my
>upcoming paper is on...) by using the color subpixels, one gets more
>points with which to reconstruct the digital image... just as film
>uses separate grains on each of the film layers.
>
Sub-pixels don't sound like they are real/physical extra pixels are they?, just
some kind of interpolated/guessed at extra data to get larger apparent
resolutions than real extra detail. This is where you lose me - a projector or
a flat panel screen can have only so many _physical_ pixels, are the sub-pixels
w/ the extra info on them some kind of mathematical interpolation of pixel data
for use on larger screens/monitors/printers etc. How is more resolution
achieved if the sub-pixels are not physical pixels, just interpolated/filled in
data? Sounds alot like Genuine Fractals where still digital images are resed up
gaining in pixel amounts but not in pixel info...
>>
>> - Do sub pixels actually carry extra (very tiny?) amoutns of information
>or are
>> they just some kind of randomized digital noise and/or re-shifting of
>the
>> information already there to avoid moire patterns and other defects?
>
>If the display is designed right, additional detail can be obtained...
See my questions above...
>note that the commonly used RGB Stripe arrangement used in LCDs is not
>designed right... and one only gets the shifts that reduce moiré as
>you correctly noted. However, my company has developed a novel
>subpixel arrangement, the PenTile Matrix(tm), that one may indeed get
>extra detail out of... to explore that further, one should visit our
>website
>
"www.clairvoyante.com" - is this the URL I should explore to, pardon the pun,
"fill in the gaps" in my understanding about subpixels?
Gordon Moat
> > I mention
> >brief bits about doing double scans, blending of channels and/or layers,
> >tricks
> >with the maximum, minimum, and highpass filters, double scanning with
> >overexposing/underexposing on the scanner, etc.
>
> What I suspected anyways, layering, multiple scans optomized for highlights,
> shadows and/or midtones, high pass filter for sharpening (learned about this at
> a Photo Expo seminar), etc. If you ever have the time and/or desire in the
> future you might want to think about posting a web page on these techniques and
> how each affects the exact same image... :-)
Unfortunately, I constantly seem to be behind in what I would like to accomplish
on my web site. It would be nice to share more of these techniques, so perhaps
something in the future.
>
>
> ><http://www.allgstudio.com/gallery/corporate/Amtrack_LAoldyard.jpg> This
> >is an
> >image of the Amtrak yard in Los Angeles.
>
> This is definitely a shot that would have benefited from combining a tripoded
> bracket of 3 different exposures of the same image (for shadows, midtones and
> brighteest hilights). Something to consider... Its a bitch trying to get the
> most out of slide film as most scanners poop out in the shadows (unless you are
> doing some high dynamic range drum scans) and detail lost in the hilights is,
> well, lost. Conventional masking (using a moderately out of focus B&W maks of
> the same image) then printing a conventional pirnt (or slide dupe) then
> scanning either of those in for further adjustment might have overcome the
> limitations of trying to get the whole range scanned in/manipulated from one
> locked in density range on a single slide, however, when/if time/opportunity
> permit, on another image perhaps, unless this image is especially important to
> you, a 3 image bracketed exposure combined in PShop might work best.
Yeah, a drum scan would have been better. Also, if I could afford one of the newer
bodies that does 3 shot auto bracketing, that would have worked nicely. Probably a
slight historical significance to the shot, since that train turning rig is the
last of its kind. There is also a chance that it may not exist anymore. Anyway,
like I stated before, it is not a portfolio quality image.
> I've heard of "surreality" but what is "somewhat reality" though?...
> >
> >Sort of like "Never, never Land" . . . .
> >
>
> This almost sounds like an insult but I'll let it pass bye since I'm not sure
> exactly how you meant it.
I was making it up on the fly, so it really has little meaning . . . poor attempt
at humour from being too tired, though I am a fan of the Walt Disney version of
Peter Pan. The internet has failed to live up to the hype, so it is the eternal
land of never going to happen, and that may be about as close as it gets to
"somewhat reality" . . . though I would rather be an idealistic dreamer,
travelling through reality ;-)
> I live in this reality (not some imaginative/fantasy
> one) and have a pretty clear view of it most times even when I'm not shooting
> it w/ my Zeiss lenses ;-), however I may, to others at least, seem to have an
> offbeat/different view of it - Never Never Land? - you must have me confused w/
> Michael Jackson ;-)
The one gloved one? Nah . . . not a chance . . . . Anyway, I do not follow any of
his stupidity, though I am sure he is still yet to top his own ineptitude.
I think with your images you live in a better than reality state. I find reality
too boring on occasion, so it is nice to see creative expression, and different
ideas. Offbeat is a term I would not ascribe to you, though unique would fit . . .
. So the lesson here is to keep creating, and keep photographing. Do not get
bogged down, or too caught up in technical gibberish.
Ciao!
Gordon Moat
Alliance Graphique Studio
<http://www.allgstudio.com>
>
>
William E. Graham
"Lewis Lang" <cont...@aol.comnospam> wrote in message
news:20030208012511...@mb-ct.aol.com...
> Hi Candice:
>
> >"negative light" is a physical impossibility...
>
> Possibly what a black hole gives off in an alternative universe? Or
perhaps an
> anti-matter equivalent to light? - why not anti-photons? (are there
such
> things? - I believe there are anti-portons and other anti-sub-atomic
particles
> so why not "anti-light" unless the negative light you are talking
about is just
> 18o depgree wavelength phase shifting cancellation of light) :-)
Physicists claim that the photon is it's own anti-particle......
Lewis Lang
some of the issues discussed in this thread:
<A HREF="http://www.kenrockwell.com/tech/filmdig.htm">Film vs. Digital Š 2003
KenRockwell.com</A>
Enjoy!
Lewis Lang
>Date: Sat, Feb 8, 2003 7:50 AM
>Message-id: <zF21a.30022$tq4.2272@sccrnsc01>
I wonder if that's the truth or just speculation on their part for something
they have very little information (and/or theories?) about...
Lewis Lang
>Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
>From: Gordon Moat mo...@attglobal.net
>Date: Sat, Feb 8, 2003 7:00 AM
>Message-id: <3E44AAFD...@attglobal.net>
>
>Lewis Lang wrote:
>
>> > I mention
>> >brief bits about doing double scans, blending of channels and/or layers,
>> >tricks
>> >with the maximum, minimum, and highpass filters, double scanning with
>> >overexposing/underexposing on the scanner, etc.
>>
>> What I suspected anyways, layering, multiple scans optomized for highlights,
>> shadows and/or midtones, high pass filter for sharpening (learned about
>this at
>> a Photo Expo seminar), etc. If you ever have the time and/or desire in
>the
>> future you might want to think about posting a web page on these techniques
>and
>> how each affects the exact same image... :-)
>
>Unfortunately, I constantly seem to be behind in what I would like to
accomplish
>on my web site. It would be nice to share more of these techniques, so perhaps
>something in the future.
>
Not enough days in the hour ;-) (never enough time)
Manual bracketing (on a tripod) would have worked as well :-)
>Probably a
>slight historical significance to the shot, since that train turning rig
>is the
>last of its kind. There is also a chance that it may not exist anymore.
>Anyway,
>like I stated before, it is not a portfolio quality image.
>
I understand.
>> I've heard of "surreality" but what is "somewhat reality" though?...
>> >
>> >Sort of like "Never, never Land" . . . .
>> >
>>
>> This almost sounds like an insult but I'll let it pass bye since I'm not
>sure
>> exactly how you meant it.
>
>I was making it up on the fly, so it really has little meaning . . . poor
>attempt
>at humour from being too tired, though I am a fan of the Walt Disney version
>of
>Peter Pan. The internet has failed to live up to the hype, so it is the
>eternal
>land of never going to happen, and that may be about as close as it gets
>to
>"somewhat reality" . . .
Thanks for explaining it to me, sorry for wondering if it was an insult, I
should have given you more of the benefit of the doubt... :-)
though I would rather be an idealistic dreamer,
>travelling through reality ;-)
>
I don't see why idealism isn't as much part of reality as any other unseen but
important underlying thought/principle in reality, ie. just because the guy
next to you can't see or feel or touch your headache doesn't make it any less
real...
>> I live in this reality (not some imaginative/fantasy
>> one) and have a pretty clear view of it most times even when I'm not
shooting
>> it w/ my Zeiss lenses ;-), however I may, to others at least, seem to
>have an
>> offbeat/different view of it - Never Never Land? - you must have me confused
>w/
>> Michael Jackson ;-)
>
>The one gloved one? Nah . . . not a chance . . . . Anyway, I do not follow
>any of
>his stupidity, though I am sure he is still yet to top his own ineptitude.
>
I saw the Brittish documentary on him a day or two ago on tv, and amongst many
other off topic things, which I wont mention here lest the net Nazis jump down
my throat, I find it intriguing, that for one interview segment he actually
hired his own separate lighting director to light for him so he would literally
be seen in the best "light" (this was during the part of the show at the end
when they were asking him the most incriminating questions). I saw at least two
soft boxes w/ grids/louvers? above him and one soft box w/ some kind of grid
spotting or louvers on it possibly as fill underneath. I couldn't tell but the
shots they did of him shopping in Las Vegas seemed to be on video. I wonder how
much of the documentary was on film and how much if any was on video. Later,
after the documentary on the next show (20/20?, Nightline? or somesuch) they
did a fascinating morphing of his face from childhood to adulthood and
interviewed in depth a plastic surgeon about his "transformation" over the
years. She commented something like "he has an end stage nose" which is kind of
freaky because I kind of got the impression that his nose was on the verge of
collapse or disappearing from her comments due to too much and/or poor surgery.
I believe she may have been the plastic surgeon that actually refused to do the
plastic surgery on Jackson. Joe Jackson on the earlier documentary? came off
looking like a greedy abusive b*st*rd in the way he treated Michael growing up
so no wonder that affected Michael later on in life in wanting to remain a
Peter Pan-like child. One of his brothers (at least) used to make fun of his
wide nose and at least one brother seemed to be doing hanky panky in the room
w/ a girlfriend virtually right in fron of him as a kid and told him to shut
his eyes. Sadly, words (and motion pictures/tv) can paint a picture far more
disturbing than some photographs. Anyway I wouldn't trade my childhood for his
anyday and mine was hardly "Disneyesque" to say the least...
>I think with your images you live in a better than reality state.
Thanks, I think of it more as "core of reality"/too real ;-).
I find
>reality
>too boring on occasion,
the surface of it is definitely at times, I guess the key is to be able to see
beyond the surface... The rare times I am actually, sort to speak "bored", I
find it kind of relaxing as I don't think reality exists to entertain anyone,
though there is much entertainment, info and insight to be found in even the
most boring moments. And most "boring" times, its just enough to "be" -
breathing and being alive are two things/equivalencies we as Americans much
under-rate. I don't see anyone in grave yards doing hip hop/break dance moves
dancing in joy at being alive (neither do I see the dead complain about being
too bored either ;-)).
so it is nice to see creative expression, and different
>ideas. Offbeat is a term I would not ascribe to you, though unique would
>fit . . .
Double thanks. :-)
>. So the lesson here is to keep creating, and keep photographing. Do not
>get
>bogged down, or too caught up in technical gibberish.
>
>Ciao!
>
>Gordon Moat
>Alliance Graphique Studio
><http://www.allgstudio.com>
>
Don't worry about that, Gordon. I must love photography because people sure
aren't paying me enough to do it just for the money ;-). Actually, as much as I
think I deserve what I get when I do my photography for the occaisonal paying
client (because I give it my all regardless), I think photographers getting
paid (just like pop stars and movie actors) is almost crazy in the sense that
we love it so much we would do it regardless if anybody paid us, and in at
least my case, regardless if anybody ever saw the finished work. I have
thousands of good photgraphs and perhaps hundreds of excellent ones that will
never be seen on my website (as I'm sure many photographers who have dedicated
themselves to photography over many years like most people devote themselves to
their kids or eating or breathing have). I love not only looking at good
technical info but using good techniques in my photography, but like good
writing skills, technique is just a start, its the ideas and the execution of
the photographs that are my main focus. Technique is just one of the lanes on a
road I take to the finished creative photograph.
Candice H. Brown Elliott
> Hi Candice:
>
> BIG SNIP OF PREVIOUS POST
>
> >> Let me see if I got you right... What you are saying is something that
> sounds
> >> like 4 words to me "high resolution stochastic printing" (perhaps w/o
> the
> >> printing aspect as such.
> >>
> >> Also, I've heard of quantum theory but I'm not quite understanding
> >> "quantization" and "negative light". Could you explain both terms in
> layman's
> >> language for me? Thanks.
> >
> >"quantization" means putting things into "counting numbers" instead of
> >the continous world of analog.
> >
>
> Makes perfect sense.
>
> >"negative light" is a physical impossibility...
>
> Possibly what a black hole gives off in an alternative universe? Or perhaps an
> anti-matter equivalent to light? - why not anti-photons? (are there such
> things? - I believe there are anti-portons and other anti-sub-atomic particles
> so why not "anti-light" unless the negative light you are talking about is just
> 18o depgree wavelength phase shifting cancellation of light) :-)
You understand completely. I mean that one reconstruction point's
negative light (from the negative going parts of the sinc function (
sinc = sin/tan )) are cancelling part of the positive going parts of
nearby reconstruction point's sinc function. This can only happen in
coherent optics.... not in real world film printers... thus, my point,
all reall world digital film printers and display monitors cannot
accurately reconstruct what digital cameras capture. The fault lies
in the film printers, not the digital camera (save for the low dynamic
range and course quantization of most of them...)
>
> save as a mathmatical
> >construct in coherent optics (read: holography). I mostly mentioned
> >it as a completeness point... mostly it is a strawman, something we
> >can't do at the moment with our limited technology.
> >
>
> Perhaps it can be mathematically/electronically simulated if not actually
> possible in physical reality?
>
Again, you now understand me correctly. One can mathmatically, in a
computer, simulate/perform the proper Shannon-Nyquist interpolation /
reconstruction sinc function. But then... one needs a
super-high-resolution film printer to output the results... which must
be four times higher, in each axis (thus 16 times more dots), at the
very least, the resolution of the original digital camera image, to
properly reconstruct the original light pattern that was focused on
the camera chip....
... and again... my point is that we simply do not perform this
neccessary step in the digital image processing sequence... and that
is why digital photography is not as good as film photography.
(Frankly, I use a digital camera all of the time... not just my
Minolta SLR 35mm camera... but I recognize, because of my professional
career in electronic displays, where they fail to measure up.)
> >
> >>
> >> Also, negative light almost sounds like you are doing somekind of wavelength
> >> cancellation through phase shifting to cancel out whatever wavelengths
> you
> >> have, this is remarkably similar in my mind to what's doen in the
> >> anti-reflective coatings of lenses to eliminate flare by shifting light
> 180
> >> degrees? out of phase to cancel out stray/non-image forming light. Am
> I on the
> >> right track here or are you suggesting something else?
> >
> >You understand me perfectly... one can only get "negative light" in a
> >holographic reconstruction.
>
> Please explain how negative light is achieved in holography, if you will...
> Thanks. :-)
Certainly... a hologram causes light wavefronts to be constructively
interfered to redirect light into a given pattern, from point to
point. If light from one point on the hologram then is 180° out of
phase with the light from another point, as it hits the film, it
destructively interferes and fails to land there at all ! Now... if
we were to record a fraunhaufer hologram (one whose image is "focused
at infinity") such that it scatters light into a pattern matching that
of the sinc function, it would be laminated onto an LCD or other two
dimensional, transmission, spatial light modulator. Photo sensitive
film or print paper could then be placed a given distance from the
LCD/hologram. Laser light would be modulated by the LCD, which is
displaying the digital image to be printed, then the hologram performs
the proper interpolation between the digitized samples, reconstructing
the original, bandlimited image that landed on camera chip... but now
reconstructed onto the film or print paper.
As I said, this is what should happen... and doesn't !!!
You are missing my point here.... these subpixels do exist, as the
red, green, and blue color dots of flat panel displays...
>
>> >
>
> "www.clairvoyante.com" - is this the URL I should explore to, pardon the pun,
> "fill in the gaps" in my understanding about subpixels?
Yes, this is the website for my company... and at that website is an
explaination of how our technology rearranges the color dots, the red,
green, and blue colored subpixels of LCDs, so that they are more
efficient at reconstructed images, including computer generated text.
For a general understanding.. though not completely accurate
(especially inaccurate is the history comments)... one may also visit:
www.grc.com
William E. Graham
"Lewis Lang" <cont...@aol.comnospam> wrote in message
> >Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
> >From: "William E. Graham" we...@attbi.com
> >Date: Sat, Feb 8, 2003 7:50 AM
> >Message-id: <zF21a.30022$tq4.2272@sccrnsc01>
> >
> >
> >"Lewis Lang" <cont...@aol.comnospam> wrote in message
> >news:20030208012511...@mb-ct.aol.com...
> >> Hi Candice:
> >>
> >> >"negative light" is a physical impossibility...
> >>
> >> Possibly what a black hole gives off in an alternative universe? Or
> >perhaps an
> >> anti-matter equivalent to light? - why not anti-photons? (are there
> >such
> >> things? - I believe there are anti-portons and other
anti-sub-atomic
> >particles
> >> so why not "anti-light" unless the negative light you are talking
> >about is just
> >> 18o depgree wavelength phase shifting cancellation of light) :-)
> >
> >Physicists claim that the photon is it's own anti-particle......
>
> I wonder if that's the truth or just speculation on their part for
something
> they have very little information (and/or theories?) about...
>
> Lewis
>
have believed it for the last 30 years at least......
NickC
Nick
Lewis Lang wrote:
> I found this a rather interesting/informative web page and it covers at least
> some of the issues discussed in this thread:
>
> <A HREF="http://www.kenrockwell.com/tech/filmdig.htm">Film vs. Digital © 2003
Annika1980
>Certainly... a hologram causes light wavefronts to be constructively
>interfered to redirect light into a given pattern, from point to
>point. If light from one point on the hologram then is 180° out of
>phase with the light from another point, as it hits the film, it
>destructively interferes and fails to land there at all ! Now... if
>we were to record a fraunhaufer hologram (one whose image is "focused
>at infinity") such that it scatters light into a pattern matching that
>of the sinc function, it would be laminated onto an LCD or other two
>dimensional, transmission, spatial light modulator.
I'll bet you're a barrel of laughs at parties.
William E. Graham
Lewis Lang
>Date: Sat, Feb 8, 2003 11:54 PM
>Message-id: <3E4598C7...@attbi.com>
>
>Interesting, Lewis. Thanks
>
>Nick
>
You're welcome, Nick :-)
>Lewis Lang wrote:
>> I found this a rather interesting/informative web page and it covers at
>least
>> some of the issues discussed in this thread:
>>
>> Film vs. Digital © 2003
>> KenRockwell.com
>>
Lewis Lang
ULTRA BIG SNIP OF PREVIOUS POST:
Thanks for explaining about negative light and holograms though I am more of a
"visual" person and I think I'll have to see a few diagrams of what goes on
before I get both the ideas and the terminolgy straight in my head ;-)
But red, green and blue dots _are_ pixels - what am I missing here? Are the
pixels you are shifting real data/pixels but you are shifting by half pixels
mathematically instead of in reality? Since a display is made up of whole
pixels I really don't see/understand how one could shift any info diagonally
half a pixel at all unless its done mathematically in some ways then
reoutputted to whole/real pixels again in some way - what am I missing here?
:-)
>>
>> "www.clairvoyante.com" - is this the URL I should explore to, pardon the
>pun,
>> "fill in the gaps" in my understanding about subpixels?
>
>Yes, this is the website for my company... and at that website is an
>explaination of how our technology rearranges the color dots, the red,
>green, and blue colored subpixels of LCDs, so that they are more
>efficient at reconstructed images, including computer generated text.
>
>For a general understanding.. though not completely accurate
>(especially inaccurate is the history comments)... one may also visit:
> www.grc.com
>
>
>
>Candice H. Brown Elliott
>Chief Technology Officer
>ClairVoyante Laboratories
>www.clairvoyante.com
>
Thanks, I'll check this out..
Regards,
Lewis Lang
>From: "William E. Graham" we...@attbi.com
>Message-id: <T5f1a.39101$HN5.1...@rwcrnsc51.ops.asp.att.net>
Thanks, regardless of what they accept/believe, I still wonder if its true -
don't particles and antiparticles cancel each other out? If a particle is its
own anti-particle how could it maintain its own existence?
Candice H. Brown Elliott
Actually... I am... I play the applachian mountain dulcimer and
guitar, and sing folk tunes... and even american pop to rock... My
interests are wide ranging, from history (ancient to modern) to the
decorative arts (Arts & Crafts movement, Deco, etc). My number one
hobby is restoring old houses. I am passionate about progressive
politics, even spent six months as an unpaid legislative lobbyist
fighting for the employment rights of medical minorities and
differently-abled people. I am a prolific writer... and I have taught
classes on a number of non-techie, as well as techie subjects. Most
people who meet me at parties think that I am an artist or something
like that. They are quite surprised to discover that I am a techie as
well.
-- Candice
Candice H. Brown Elliott
Wrong gender... check it out... how many men do you know who are named
"Candice". Further, Mr. Claven was ignorante of the subjects upon
which he pontificated... and covered his ignorance by making things
up. I am perfectly willing to admit ignorance... which alas, is
often, as the universe is an awfully big place with many secrets yet
untold...
I was asked to clarify some points... and made an honest effort to do
so to the best of my ability. It is mean spirited of the two posters
here to make fun of that... and of me.
-- Candice
Candice H. Brown Elliott
Given the two mean spirited posts from two other individuals, I will
just have to trust that I am not being egged on...
When we speak of "subpixels" we are using an accepted convention, one
in which we speak of "whole pixels" as being full color, with three or
four primaries. The term "subpixel" is refering to the color
primaries that make up the "whole pixel", or for short, "pixel".
On an LCD screen, the colored dots are not overlapping each other.
There is a spatial displacement. In conventional image rendering, the
computer makes the assumption that the colored dots are overlapping
each other. But, since this is not true, the conventional rendering
is missing an opportunity to take advantage of the spatial
displacement for increasing the "addressability" of the panel, to
reduce the aliasing and moiré found in so many digital image
reconstructions.
About fifteen years ago, some guys at IBM decided to take advantage of
that displacement... and even patented the technique... but IBM never
commercialized it. Before the patent was issued, I independly
invented very similar ideas... but then the patent issued, and I
couldn't do anything with it. Several years ago, that patent was
allowed to lapse, due to failure to pay the maintanance fees. Twelve
days later, Microsoft announced ClearType(tm) a text quality
improvement technology that was actually the very same as the IBM
technique... and of course... just after that, I launched my company
to further develop the technology. So... now ya know.
Lewis Lang
>Message-id: <1e9ffba8.03020...@posting.google.com>
Candice:
You do realise that at least one of your posters is the resident (bad imitation
of a) troll on this newsgroup (my apologies to the real trolls). And youre
right, they both shouldn't have made fun of you. But arguing w/ Annika is like
arguing w/ a Smurf (blue cartoon-type character), that's why he's in my
killfile, not so much for his lack of intellect and lack of talent (he hasn't
any and uses equipment worship to gloss over that fact) but for how he
compensates for both of those factors w/ his willfull stupidity. This may
_sound_ like an insult, but unfortunately it is the accurate truth - just read
(or better yet, spare your eyes and don't read) any of his thousands of inane
postings on how great his Canon cameras are and how they have god-like powers,
or worse, take a look at what great photos he posts to see the accuracy of my
assesment of him. He's not worthy of your time/talents/intellect/attention,
though it was interesting hearing all those things you are/have been involved
with in replying to him :-)
Regards,
Lewis Lang
>Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
>From: cell...@clairvoyante.com (Candice H. Brown Elliott)
>Date: Sun, Feb 9, 2003 8:20 PM
>Message-id: <1e9ffba8.03020...@posting.google.com>
Candice, as I've said in so many words in the other post I made just now that
I'm sad you had to deal w/ their meanness, can't you assume by now that what I
am asking is genuine/sincere? Don't let your bad experience of others color
what you think of me (or anybody else on this newsgroup). If you have been on
this newsgroup awhile, you will come to notice who are the real posters and who
are the trolls :-). By the way, I'm into humor, not egging on, the question(s)
were/are sincere.
>When we speak of "subpixels" we are using an accepted convention, one
>in which we speak of "whole pixels" as being full color, with three or
>four primaries. The term "subpixel" is refering to the color
>primaries that make up the "whole pixel", or for short, "pixel".
>
OK, now I understand. I usually think of what you would call sub-pixels (red,
green and blue pixels) as pixels themselves...
>On an LCD screen, the colored dots are not overlapping each other.
>There is a spatial displacement. In conventional image rendering, the
>computer makes the assumption that the colored dots are overlapping
>each other.
Really, why? Don't the people who design these things assume that the pixels
will either be displayed on a monitor or outputted onto paper/etc. w/
non-overlapping pixels? It would only make sense to me that they do assume this
spacing since that's the way things usually work.
But, since this is not true, the conventional rendering
>is missing an opportunity to take advantage of the spatial
>displacement for increasing the "addressability" of the panel, to
>reduce the aliasing and moiré found in so many digital image
>reconstructions.
>
What do you mean by "increased addressability" -overlapping of pixels (or what
you would call overlapping of subpixels)?
>About fifteen years ago, some guys at IBM decided to take advantage of
>that displacement... and even patented the technique... but IBM never
>commercialized it.
Reminds me of the WYSIWYG technology that both Apple and IBM capitalized on
from Xerox park's lack of foresight as to the value of its innovations...
Before the patent was issued, I independly
>invented very similar ideas... but then the patent issued, and I
>couldn't do anything with it. Several years ago, that patent was
>allowed to lapse, due to failure to pay the maintanance fees. Twelve
>days later, Microsoft announced ClearType(tm) a text quality
>improvement technology that was actually the very same as the IBM
>technique... and of course... just after that, I launched my company
>to further develop the technology. So... now ya know.
>
:-)
May I assume you took out a patent too or is that technique now just considered
"in the public domain" due to the patent lapse?
SNIP
Regards (and don't let the trolls get to you),
Annika1980
>> I'll bet you're a barrel of laughs at parties.
>
>Actually... I am... I play the applachian mountain dulcimer and
>guitar, and sing folk tunes... and even american pop to rock... My
>interests are wide ranging, from history (ancient to modern) to the
>decorative arts (Arts & Crafts movement, Deco, etc). My number one
>hobby is restoring old houses. I am passionate about progressive
>politics, even spent six months as an unpaid legislative lobbyist
>fighting for the employment rights of medical minorities and
>differently-abled people. I am a prolific writer...
I don't doubt that!
Nice resume, btw.
I play golf.
I hope you continue to post on this forum.
After dealing with all the psuedo-intellectuals up in here, it is refreshing to
meet someone who actually is one.
You'll find that this place is filled with posers and no-talent photographic
wannabes like Loser Lang, but there are a few contributors like yourself who do
have something meaningful to say and actually know a bit about the subjects
they discuss.
NickC
Annika1980 wrote:
>>From: cell...@clairvoyante.com (Candice H. Brown Elliott)
>
>
>>>I'll bet you're a barrel of laughs at parties.
>>
>
>>Actually... I am... I play the applachian mountain dulcimer and
>>guitar, and sing folk tunes... and even american pop to rock... My
>>interests are wide ranging, from history (ancient to modern) to the
>>decorative arts (Arts & Crafts movement, Deco, etc). My number one
>>hobby is restoring old houses. I am passionate about progressive
>>politics, even spent six months as an unpaid legislative lobbyist
>>fighting for the employment rights of medical minorities and
>>differently-abled people. I am a prolific writer...
>
>
> I don't doubt that!
> Nice resume, btw.
> I play golf.
>
> I hope you continue to post on this forum.
> After dealing with all the psuedo-intellectuals up in here, it is refreshing to
> meet someone who actually is one.
<VBG>
Now there's a line. "...refreshing to meet someone who actually is one."
After a dull day, reading this line could perk up one's evening.
Nick
Lewis Lang
Please do me a favor and don't quote Anniktrash so I won't have to read his
gentically malformed meaningless ---> LIES and TOTAL(LY) ----> idiocy.
I was trying to hold a meaningful conversation w/ Candice but this butt
munching bedlamite, demented, madman, maniac, non compos, nut, demoniac,
energumen; raver; neuropath, neurotic, paranoid, psycho, psychoneurotic
<A HREF="aol://4344:1708.T0004652.40243758.674318644">CRACKPOT</A>, crackbrain,
crank, cuckoo, ding-a-ling, harebrain, kook, nut, screwball, Section 8
(Annikrap) stuck his doggie nose in our business and started sniffing at our
backsides...
Please don't quote him/her/it (Annikrap1980), either start a new thread or
bring it to email if you will.
Thanks for understanding, which is more than Annikrap ever will.
>Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
>From: NickC n-c...@attbi.com
>Date: Mon, Feb 10, 2003 2:34 AM
>Message-id: <3E470FBB...@attbi.com>
Takes one to know one and as far as being any kind of talent or even a wannabe
you _are_ no one.
but there are a few contributors like yourself
>who do
>> have something meaningful
What would you know about "meaning"? Annikrap? All you do is (_for years now_)
spout meaningless crap and odes to your Canon camera and post crappy pictures
of dog's crapping, golf and other, and I'd use the word loosely, "pictures".
You contribute nothing. You are a black hole of knowledge and stupidity in and
I pity the poor dog who gets sucked in. With you, meaning gets sucked in and
dog crap gets spouted out.
to say and actually know a bit about the subjects
>> they discuss.
>>
>>
If you don't like what I say you can get off this thread Annican't - better
yet, I personally invite you to leave this newsgroup, it would be a _much_
better place w/o you (as it was before you muddied it w/ your
meaningless/pointless commercials for the D60, 1v and other Canon cameras). I
just figured out what you are, you're not just a worthless no talent troll but
a spammer. Go spam another ng w/ your Canon ng commercials and your links to
your no talenthood. W/ ever photo you make and every image you post you just
show exactl what a loser _YOU_ are. You would be funny if your posts weren't
such mindless yuppie chants/spams for Canon and your boasting about your
photography which is so pathetic and limited in scope (forget about meaning, it
hasn't any) and that ranges from the low depths of ubelievably suicidal bad to
the stellar hights of mind-numbing mediocrity w/ your "shots" of your dog, cat
and the poor unfortunate golf crap you spew at us. The only thing you've ever
contributed to this group is your ego, your mindless commercialism and
unfailing loyalty to brown nosing Canon as your "God" and your own babyish
rants. Go find your own sandbox to play in Anniktrash, the only loser, Brett,
is you.
Oops, I forgot, you're in my kill filter and you no longer exist so poof and be
off w/ you.
Kinon
hell will I buy lenses that will quickly become obsolete when larger sensor
cameras become more commonplace.
"Simon Stanmore" <nom...@thanks.com> wrote in message
news:104465659...@iris.uk.clara.net...
> : )
>
> I've a nice comprehensive Nikon system at the moment and yes ... Canon is
> naturally the other option. I'm waiting for image samples and detailed
> reviews of the new releases before I decide (I'm in no great rush). It's a
> system thing as well - Nikon's latest DX lens idea is something I really
> don't like. The F-mount is already littered with incompatabilities and
this
> new line of lenses is just pushing the whole mess too far for me now
>
> Simon
>
>
>
> > >Digital is the
> > >(very immenent) future as far as 35mm-type work is concerned, and I'll
be
> > >going with the marque that I consider to be implementing the technology
> in a
> > >fashion that I consider the most effective and long-term .
> >
> > So have you decided which Canon body you'll be buying?
>
>
Annika1980
>I was trying to hold a meaningful conversation w/ Candice but this butt
>munching bedlamite, demented, madman, maniac, non compos, nut, demoniac,
>energumen; raver; neuropath, neurotic, paranoid, psycho, psychoneurotic
>CRACKPOT, crackbrain,
>crank, cuckoo, ding-a-ling, harebrain, kook, nut, screwball, Section 8
>(Annikrap) stuck his doggie nose in our business and started sniffing at our
>backsides...
For someone who supposedly has me in his little killfile ("real world filter"),
you sure do get torn up by what I post.
Who Rules your butt?
Magnus W
news:38P1a.2712$c6....@bos-service2.ext.raytheon.com:
> Man, I couldn't agree more about those DX lenses! What a joke! No way
> in hell will I buy lenses that will quickly become obsolete when
> larger sensor cameras become more commonplace.
Why do you think they will? The Nikon idea of creating "those DX lenses"
makes us who want our long, fast teles become longer fast teles happy,
while offering a wideangle alternative. Smaller sensors equals lower price.
Everyone's happy. The DX lens (there is only one yet AFAIK?) has actually
restored much of my faith in Nikon as a digital player.
Candice H. Brown Elliott
> Hi Candice:
<BIG SNIP>
> >About fifteen years ago, some guys at IBM decided to take advantage of
> >that displacement... and even patented the technique... but IBM never
> >commercialized it.
>
> Reminds me of the WYSIWYG technology that both Apple and IBM capitalized on
> from Xerox park's lack of foresight as to the value of its innovations...
>
Don't be too hard on IBM. First, the LCD technology during that time
wasn't good enough to support it. That is only now being fixed.
Second, no one else seemed to be very interested in the technology.
(though, of course, they were... but were also waiting for the LCDs to
get better)
> Before the patent was issued, I independly
> >invented very similar ideas... but then the patent issued, and I
> >couldn't do anything with it. Several years ago, that patent was
> >allowed to lapse, due to failure to pay the maintanance fees. Twelve
> >days later, Microsoft announced ClearType(tm) a text quality
> >improvement technology that was actually the very same as the IBM
> >technique... and of course... just after that, I launched my company
> >to further develop the technology. So... now ya know.
> >
>
> :-)
>
> May I assume you took out a patent too or is that technique now just considered
> "in the public domain" due to the patent lapse?
The general concept is now in the public domain, but not improvements.
Its like this: The old patents on electricity are in the public
domain, but modern integrated circuits are still patentable.
Microsoft improved the subpixel rendering technique as applies to text
rendering on conventional RGB Stripe LCD. ClairVoyante has improved
it with respect in increasing the general resolution of flat panel
displays, improving the arrangement of subpixels and matching subpixel
rendering algorithms. We have both applied for improvement patents.
ClairVoyante has something over two dozen applications pending.
>
> SNIP
>
> Regards (and don't let the trolls get to you),
Thank you, and to the others that spoke kindly. I guess there is
still a bit of that ugly duckling little girl in me... the one that
used to cry when she was teased in grade school for sounding like
"Encyclopedia Brown", a geeky character in a popular book series. I
was the most geeky kid in the school. I am still touchy about it.
Of course, one of my high school classmates used to be considered a
geek and a nerd. His name is Steve Jobs (Founder and CEO of Apple
Computer).
Candice H. Brown Elliott
Chief Technology Officer
ClairVoyante Laboratories
www.clairvoyante.com
Annika1980
>I guess there is
>still a bit of that ugly duckling little girl in me... the one that
>used to cry when she was teased in grade school for sounding like
>"Encyclopedia Brown", a geeky character in a popular book series. I
>was the most geeky kid in the school.
Shocking!
Lewis Lang
>Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
>From: cell...@clairvoyante.com (Candice H. Brown Elliott)
>Date: Tue, Feb 11, 2003 12:54 AM
>Message-id: <1e9ffba8.0302...@posting.google.com>
>
>cont...@aol.comnospam (Lewis Lang) wrote in message
news:<20030209183404...@mb-ct.aol.com>...
>> Hi Candice:
><BIG SNIP>
>
>> >About fifteen years ago, some guys at IBM decided to take advantage of
>> >that displacement... and even patented the technique... but IBM never
>> >commercialized it.
>>
>> Reminds me of the WYSIWYG technology that both Apple and IBM capitalized
>on
>> from Xerox park's lack of foresight as to the value of its innovations...
>>
>
>Don't be too hard on IBM. First, the LCD technology during that time
>wasn't good enough to support it. That is only now being fixed.
>Second, no one else seemed to be very interested in the technology.
>(though, of course, they were... but were also waiting for the LCDs to
>get better)
>
Actually, I was referring to (or at least thinking about ;-)) the use of the
GUI interface (use of icons, windows, etc. that were later incorporated into
what would become the "Lisa"? and Apple Macintoshes and various Microsoft
versions of Windows TM).
Ideas are like bubbles in ginger ale, there are so many there floating around
that we don't see and a few that float away entirely and never get used and a
few that cling to the glass. Actually I like ginger ale but I much prefer cream
soda or Root Beer and I prefer juices to sodas but that is an entirely 'nother
story :-).
>
>
>> Before the patent was issued, I independly
>> >invented very similar ideas... but then the patent issued, and I
>> >couldn't do anything with it. Several years ago, that patent was
>> >allowed to lapse, due to failure to pay the maintanance fees. Twelve
>> >days later, Microsoft announced ClearType(tm) a text quality
>> >improvement technology that was actually the very same as the IBM
>> >technique... and of course... just after that, I launched my company
>> >to further develop the technology. So... now ya know.
>> >
>>
>> :-)
>>
>> May I assume you took out a patent too or is that technique now just
considered
>> "in the public domain" due to the patent lapse?
>
>The general concept is now in the public domain, but not improvements.
> Its like this: The old patents on electricity are in the public
>domain,
Has air been patented yet? ;-) Can Perrier patent its water? Just how far can
patenting go?
but modern integrated circuits are still patentable.
>Microsoft improved the subpixel rendering technique as applies to text
>rendering on conventional RGB Stripe LCD. ClairVoyante has improved
>it with respect in increasing the general resolution of flat panel
>displays, improving the arrangement of subpixels and matching subpixel
>rendering algorithms. We have both applied for improvement patents.
>ClairVoyante has something over two dozen applications pending.
>
ClairVoyante sounds like someone who has E.S.P. (and/or to perceive matters
beyond the ordinary range of perception - perfect name for a company that deals
in "sub-pixel" technology) and can see the future, it also sounds something
like "clear vision", how was this name chosen? It has (to me) similar
connotations as "Lucent Technologies" ("Lucent as also in "to see
clearly"/'marked by clarity'/'translucent').
>
>>
>> SNIP
>>
>> Regards (and don't let the trolls get to you),
>
>Thank you, and to the others that spoke kindly. I guess there is
>still a bit of that ugly duckling little girl in me... the one that
>used to cry when she was teased in grade school for sounding like
>"Encyclopedia Brown", a geeky character in a popular book series.
I know, either I used to read them and/or my mom used to read them to us when
we were kids (its been a long time, and like my computer, memory is the first
to go ;-)).
I
>was the most geeky kid in the school. I am still touchy about it.
>
I can never understand why intelligence is considered "uncool" - I guess some
people put down what they envy the most. This (popular) society seems to
cherish looks and stupidity/conformity and intelligence/perceptiveness (that
can't be either understood and/or assimilated by the masses) sticks out and is
given a "stigma".
>Of course, one of my high school classmates used to be considered a
>geek and a nerd. His name is Steve Jobs (Founder and CEO of Apple
>Computer).
>
I'm using his computer ;-). As much as I hate sloganeering and mindless
commercialism, I think/feel/believe the "think different(ly)" Apple campaign
perfectly sums up what Apple/Macintoshes are about. I've, I hesitate to use the
word in a materialistic context, but anyways, I love what the Macintosh is all
about, a machine for people who want to concentrate on what those machines can
do, as perfect/tool extension of self-expression. Pure intuitiveness. D*mn,
somebody stop me before an ad agency signs me up ;-)!
>
>Candice H. Brown Elliott
>Chief Technology Officer
>ClairVoyante Laboratories
>www.clairvoyante.com
>candice_hellen_brown.tripod.com
Regards,
Kinon
small sensor cameras, like the AF lenses for the F3 were a stop gap. See any
of those around?
Nikon should be spending their limited resources on a full-frame D-SLR
instead of wasting their time on stupid lenses that are useless to anyone
who doesn't own one of the small sensor cameras. And those people really
don't need anything but a wide angle that's special, so I'm hoping that this
is a one-shot lens.
"Magnus W" <va...@death-star.spam-trap.com> wrote in message
news:Xns931EE79AC676...@212.83.64.229...
Kinon
Rocco Kinon
"Candice H. Brown Elliott" <cell...@clairvoyante.com> wrote in message
news:1e9ffba8.03020...@posting.google.com...
Simon Stanmore
seem to be the case. They have more in the pipeline apparently. Yes they're
small / light, and yes the tele factor (crop factor actually) of smaller
sensors is appealing ... but ... the price (in writing) from my local Nikon
Professional dealer for the impending 12-24 DX is £1000.00 (UK). They often
overshoot on these 'reserve yours now' prices but even so... if it's say
actually £800.00 for this variable (modest) apertured optic with a uncertain
future then Nikon have put the final nail in their coffin as far as this
consumer with several thousands of pounds of their gear is concerned ... but
Canon may do that for them anyway...
Look at the 80-400 VR - Nikon catching up with Canon at last ... Well not by
a long shot: No AF-S (it is a slloowww AF'er) and no AF teleconverter -
Features I value with long tele's (zoom or not) for my wildlife stuff. I
love my Nikon body's - both old and new. I regard their flash system as the
best. The optics are optically spot on ... But genuinely beneficial advances
are being made by Canon and Nikon seem unable to offer us a credible
alternative
OK - Rant over . . .
Simon
Kinon
Do you have the real answers scribed on the bottom of your clubs?
Magnus W
news:mhb2a.2717$c6....@bos-service2.ext.raytheon.com:
> I think this lens, and any othere like it are a stop gap measure for
> their small sensor cameras, like the AF lenses for the F3 were a stop
> gap. See any of those around?
Your analogy is flawed. There were no advantages with using a F3AF camera
with matching lenses above using a more "modern" AF system. There are,
however, significant advantages with sticking to a smaller sensor format
(although there are disadvantages, too). My opinion is that the advantages
with the smaller sensor format are bigger than the disadvantages. You have
not shown otherwise. Whining about Nikon "wasting their time on stupid
lenses" isn't much of an argument, when it's easy to see that they in fact
are catering to professional's needs with the very same lens.
ThomasH
since the discussion loses itself a bit on a variety of
topics, let me answer today evening with a new thread.
My impression is that some issues regarding computers, file
systems and the very nature of digital signals as such are
not sufficiently valued here among the photographic community.
I would hate to say "resistance if futile," I prefer to say
"resistance is wrong." Digital storage has all the advantage
you can dream of, you just have to embrace it more!
I just restored yesterday some raw-tiff files from tape to a
secondary drive, 10 gig of them in some 2h timespan. My own work
effort on this was maybe a minute or so. Price for the tapes:
approx. $10 for 2 DDS2 tapes, DAT is ultra, ultra cheap per gig,
you may get as many as you want without hurting budget at all
and even make some redundant copies of the data on tapes by
a different manufacturer. One possible source for all sorts of
tapes is:
http://www.tape.com/cgi-bin/SoftCart.exe/dat.html?L+cassette+cpyf8723+1045208834
Thomas
Lewis Lang wrote:
>
> Hi Thomas:
>
> >Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
> >From: ThomasH tho...@coco.net
> >Date: Fri, Feb 7, 2003 11:42 PM
> >Message-id: <3E444468...@oracle.com>
> >
> >Lewis Lang wrote:
> >>
> >> Hi Thomas:
> >>
> >> >Subject: Re: Digital vs. Film Quality Now A Non-Issue? - L
> >[...]
> >>
> >> >If stored in regular room temperature conditions, E6 film can not
> >> >stay in good shape for over a decade or two. Not to talk about
> >> >the possible impact of fungi, bacteria and dust. All my slides
> >> >from the 80ties look a bit faded out, especially highlights
> >> >are almost without detail.
> >> >
> >>
> >> Newest data from Henry Wilhelm suggests dark storage life of 40 years
> >for
> >> Fujichromes and 120+ years from E-6 Ektachromes. I haven't had problems
> >of
> >> either fading on me yet and I 've been using both since about 1983...
> >
> >Thanks, I will browse than the web site! The book is already dated
> >and the web page was down for a long time. Latest data which I had
> >was in favor of Fujichrome E6 materials.
> >
>
> You're welcome. Plus, if you ever have a chance to get to a library that has
> back issues of Pop Photo check out the April '99 issues article "HOW LONG WILL
> YOUR PRINTS LAST"? - it has a table w/ updated film longevity figures. Wilhelm
> hasn't done a major mag. update of his 1990 "GOING... GOING... GONE!..."
> article, since about, well, 1990 ;-) and my guess would be that his 'Image
> Permanence' book hasn' been update sine ''92/93? or whenever it came out - a
> real pity...
>
> >
> >>
> >> And you are tied to a life time of copying and recopying huge amounts
> >of data
> >> on a non-archival unstable storage medium that may or may not exist a
> >short
> >> time down the road of life...
> >
> >The attractive here is, that this is not an effort at all, really.
> >
> >Using the external drive storage for example, you just copy over
> >the entire directory tree with a few mouse clicks. Let the drive
> >rattle than in the background and mind your other tasks. Besides
> >I am accustom to a daily and monthly backup anyway.
> >
>
> Yes, you're right about this from one aspect, w/ external and extra internal
> harddrives, this is probably very easy to do/update, however <BEGIN MR. T
> VOICE/"I pity the fool who has a multitude of images on many CDs"/END MR. T.
> VOICE> :-). Still, digi requires constant backing up throughout your life
> (regardless of how often you do it) and then after your gone, if you care about
> its longevity after you live, you probably have about 5-10 years max before
> that info may be lost if not recopied, whilst Kodachrome, Ektachrome and
> possibly Fujichrome (we have yet to see how real the projected figures of
> archival life willpan out in reality) will most likely be little diminished or
> easily recoverable (whatever dye changes have taken place) through scanning and
> reoutput on a film recorder (or somesuch) to film, Evercolor or other archival
> paper process or some other more stabile future medium.
>
> >
> >>
> >> >For example I found my old 1st generation DAT tape with a copy
> >> >of my Phd theses. My DAT DSS3 Seagate Scorpio (24gig capacity)
> >> >could read that tape recorded in mid 1980ties without a problem.
> >> >The tape was almost 18 years old. Of course a bit luck helped.
> >> >I used back than a PC-DOS program called Arcada Backup to store
> >> >my data. Arcada was than purchased by Seagate and Seagate Software
> >> >made a contract with Microsoft to deliver this very backup as the
> >> >W95 backup, what is still to this day a valid Microsoft format.
> >> >
> >>
> >> Had you waited too long, perhaps into the mid 90's you might not have
> >been able
> >> to read that file. Also, relying on the "generosity" of Microsoft or any
> >other
> >> company to keep alive a way(s) of reading your data is a non-prudent gamble
> >at
> >> best.
> >
> >I know, you are fully correct. The choice of a format might be an
> >issue. Nowadays the MS backup is probably as future safe format as
> >it gets!
> >
>
> In an era of corporate downsizing, possible war and great economic uncertainty
> even "the gold standards" are not so golden anymore, hence my healthy mistrust
> of a medium such as digital whose economic heartbeat is based on planned
> obsolecence.
>
> >The storage of files to another drive does not use any particular
> >storage format at all. Its just a MS file system, and these will
> >never be abandoned.
> >
>
> I'm not sure what you mean by file system - operating system, file format,
> program? I have heard proposals for non-hirchal computer operating systems,
> everything is subject to change and 50 years from now, if we are fortunate
> enough to still be alive, we might have machines that might not even rsemble
> modern day computers much less be compatable with them and their operating
> systems.
>
> >
> >>
> >> Not yet mentioned problem? All we fever do here is talk about the
> obsolecsnece
> >> of digital storage drives/media when the subject of archival longevity
> >of film
> >> vs. digital media comes up on this ng. The Fujitsu magnetic optical
> >> drives/media? are rated at about 10 years on their website, these seem
> >to be
> >> some of the longer lasting media/drives and have been around for years
> >though
> >> they are don't have "Zip disc" and "CD/DVD" popularity. I aggre in theory
> >w/
> >> what you say about always choosing the most popuar media/drives however
> >I am
> >> leaning towards less popular and longer lasting (both archivally/longevity
> >and
> >> in terms of how long the company/format has been in business selling like
> >> products to the public) digital media such as magnetic optical. Frankly
> >I feel
> >> that all digi solutions for both archival longevity and
> >> brand/company/drive/media longevity in terms of sales availability of
> >the
> >> product (or similar products years down the line to read the older products)
> >> are piss poor to poor at best at this time. I am looking at external hard
> >> drives also as a _longer_ but not long term solution not only for its
> >cheapness
> >> but for its longevity. My own computer's hard drive is going on 9 years,
> >though
> >> I have no info on how long other external hard drives as well as removable
> >> media last, I wish Henry Wilhelm would do something more on digital sotrage
> >> media archival longevity as well as an update to his film/paper longevity
> >> info...
> >
> >The optomagnetic RW CD's are not reliable at all, period! Nobody
> >stores data on them, unless ignorance is the reason.
> >
>
> The Fujitsu is a magnetic optical drive that is more of a Zip disc kind of
> cartridge configuration. Regular CDRs use dyes only, I thought, written on and
> read by lasers. What format is the optomagnetic RW CDs you are talking about? -
> I thought CD RWs use mereyl a (unstable ) dyeset and not some kind of magnetic
> optical configuration...
>
> >The durability of the magnetic storage is known already from the
> >condition of the countless large reel tapes dated 196x or 197x.
> >Otherwise my instinct says: the smaller the data density, the
> >more molecules memorize the data bits and the better is a chance
> >for a good durability.
> >
> >
> >While speaking about the CD's and their expected life span, let
> >me also add that the audio CD's store information with redundant
> >data blocks allowing automatic damage repair to an extend.
> >
>
> Audi CDs are also much more archival because their data dimple pits are pressed
> from glass masters as opposed to being recorded on dyes whose
> contrast/readability may fade w/i just a couple of years...
>
> >Basics of such error-correcting codes are known for a long time
> >already and they have been investigated in the new discipline
> >called Information Theory. One of the simplest codes of such
> >kind is the Hamming code.
> >
> >Such principles were applied from the very beginning for the
> >storage of the audio signal because the CD was first used as
> >an audio media, and became only later adopted by the computer
> >industry. First level of repair: Digital, lossless, Second
> >level or repair: Analog interpolation. This makes the CD's so
> >resilient to fingerprints and even to some minor scratches in
> >an amazing amount.
> >
>
> Were that the digital data CDRs and CD RWs we all use were as archival as the
> audio CDs - a real pity...
>
> >I must admit that I completely do not know if these principles
> >are being used similarly to protect digital data, the CDFS or
> >Joliet for example.
> >
> >For photography storing we would surely need a storage format
> >allowing such two-level recovery: First digitally, a complete
> >recovery possible. Appearance of damages of such kind would be
> >a warning bell: "Copy your data, media is deteriorating." The
> >second level would be some sort of interpolation, which would
> >deliver "restored" images with some degree of deterioration.
> >
>
> I don't have the technical knowledge to tell you whether this is possible or
> not, but what I do know is that a deteriorated film image can easily be
> adjusted for faded color and missing pieces of data due to the emulsion flaking
> off/spots can at least be cloned/rubber stamped over to give the impression of
> a flawless print. Once a digital file loses some of its data does it not become
> completely corrupted or am I missing something here?
>
> >
> >[...]
> >>
> >> Just back up the film to digital for copies as needed (_not every image
> >needs
> >> to be scanned_, just the best/most useful) or have optical or digital
> >dupes
> >> made of only the very best of theimages onto film again.
> >
> >The point here is though that the emulsion in every film
> >was exposed to the developing liquids, which in turn were
> >active and caused complex chemical reactions. No matter how
> >well you will flush and clean the film, some sort of a rest
> >reaction will continue to happen. Even the chemicals from
> >our surrounding air may diffuse into the emulsion and cause
> >some further changes over a time. All materials age and
> >deteriorate somehow.
> >
>
> True, see comments above.
>
> >I store my slides in regular slide magazine boxes, by Leica
> >or Reflecta. I like to assume that the used plastics for these
> >boxes were are well chosen for the film storage (no fume
> >emission etc.) These boxes are being kept in a dark closet in
> >a dry room, albeit in a room temperature. Its just a regular
> >home storage, as good as it gets without cooling.
> >
>
> Might be a good idea, instead of assuming, to contact the companies in
> questionto ask them exactly what _are_ the materials their slide boxes are made
> of. Then find out from Wilhelm's, Light Impressions, and other websites (and
> google.com's archive, whether these material will out gas or react in some
> other way w/ your slides. If they do, it would be wise to invest in non-PVC
> archival slide pages or some similar non-out gassing, non-reactive archival
> storage method. Never assume anything, especially when it comes to your own
> precious slides.
>
> >All my slides from the 80 ties show a significant loss of
> >details in the highlights: Kodak, Fuji and Agfa alike.
>
> My, oh, let me take a stab at it, wild off the top of my head guess is that
> either the boxes you have been storing them in and/or any wood/paint/etc.
> conditions that may outgassing around them may have effected them - 20 some odd
> years time is _way_ too short for slides to start significantly fading (3M
> slides may be another story, however, this brand is known to be piss poor for
> slide archival longevity).
>
> This
> >is just a subjective observation without trying to analyze
> >if Fuji E6 films are indeed so much more durable than the
> >others. Maybe the recent advances in the E6 materials were
> >so great that these old slides simply represent the past
> >technology and did not aged so much?
>
> Between '78? and about 82/83? there might have been some changes in the E-6
> process which would significantly affect archivability (no to not as much
> stabilizer used and/or some other factor(s)?) as well as emulsion imporvements
> themselves. But I have seen older Kodacolor negs that lasted longer so I really
> wonder if it has anything to do w/ the film emulsion and something more to do
> with processing and/or storage medium (your slide boxes) or storage
> environmental conditions (out-gassing from wood/plastic/paint/etc. near bye...
>
> I do not know, but I
> >know that my pixels in raw tiff's will not age ! :-)
> >
> >Thomas.
>
> No, they will just deteriorate and become unreadable unless readily and
> constantly copied every now and then, unfortunately ;-(... Look Thomas, I'm
> very sorry about what happened to your slides but digital is merely a stop gap
> solution, not a total archival solution, just a part of it, you should really
> consider, if/when you have the money, keeping different forms of the best
> slides in your collection in different locations (to safe guard against
> damage/theft,etc.) and making at least two sets of duplicate slides (in
> addition to whatever sets of backups you have done/scanned digitally off of
> your best slides) of your best images (optical dupes (corrected w/ whatever
> filtration/etc. could help optically to restore color balance/density/contrast)
> and/or adjusted for color/density digital outputted to a film recorder) and
> possibly store one set of copy slide dupes in the freezer (according to
> archival storage methods that will lock out any moisture (consult Wilhelm/Kodak
> and possibly others for more info on how to do this _properly_ - you can't just
> shove slide pages in a baggy and hope everything will come out OK/unaffected by
> moisture) and store another dupe slide set undr normal temp/humidity in some
> real archival slide pages and not those boxes you've been using. Keep all
> slides away from furniture, wood, mothballs, paint and other solvents that
> might out gas and cause chemical changes. An air purifier (in addition to
> central air/heating to maintain constant temperatures might also not hurt. You
> also might want to store your affected/faded slides separately from your dupes
> (just being extra cautious here). At the very least, check on the materials
> used in your slide boxes and when you have the cash, consider transferring
> these slides (and all other dupe slides, even if at a different location) into
> archival non-reactive, non out-gassing slide pages. You've spent I'm sure a lot
> of time and money on your images. it would be a shame, if for the lack of a
> little more time and money (by using archival storage materials/conditions and
> dupes of the best lsides) if you would not at least consider extending whatever
> time/life your slides have left.
Lewis Lang
>Subject: Start a new talk about digital image storage? Re: Digital vs. Film
>
>From: ThomasH tho...@coco.net
>Date: Thu, Feb 13, 2003 10:55 PM
>Message-id: <3E4C227F...@oracle.com>
>
>Lewis,
>
>since the discussion loses itself a bit on a variety of
>topics, let me answer today evening with a new thread.
>
>My impression is that some issues regarding computers, file
>systems and the very nature of digital signals as such are
>not sufficiently valued here among the photographic community.
>
I believe it depends upon the individual(s)...
>I would hate to say "resistance if futile," I prefer to say
>"resistance is wrong." Digital storage has all the advantage
>you can dream of, you just have to embrace it more!
>
Its not a matter of embracing - I recognize it exists and I see its advantages
and disadvantages. If a medium that had even better quality than film and more
archivability (longevity) than film (w/o the need for constant back ups and
planned obsolescnece of formats/equipment/etc.) and it was affordable, I'd be
shooting it in a heartbeat. Digital is _not_ that ultimate medium - neither is
film (unless you freeze it for a few thousand years to prevent the "vinegar
syndrome" from taking place as the emulsion will eventually vinegarize). I am
not against digital, I am against certain aspects of digital (mainly quality of
color depth, high price, and longevity of a tse tse fly ;-)). I am also for
certain aspects of digital (easy to copy and recopy w/o loss and the ability to
make ultra high quality enlargements w/o the optical degradation of having to
go through an enlarger lens). So what I see, I believe anyways, is a
cautious/balanced view in which I try to use the best of both worlds - film for
image capture for high quality and low immediate expense sake and ultra long
longevity and digital for disaster backups of my images and superb
Frontier/etc. digital prints. I am very happy to be living at least
technologically during this time because of the rapid advancements that were
just idle dreams but a few years back. But I'm not jumping on anyone's
bandwagon, yours or anybody elses who loves digital (even though they may have
good reasons to love certain aspects of digital such as easy copyability w/o
degradation and low (once you have the gear/computer/etc.) per shot costs and
reusability) because it doesn't meet allof my requirements. Neither does film.
But that's where it stands for the moment. In the meantime, as technology
progresses in both films and digi cameras/etc. I just keep an eye out for
whatever comes along and seems useful to me, and if I can use it and afford it,
well I do. I just wont be convinced that digital is a long term storage medium
because, pardon the pun, I've seen no real evidence "to back it up" ;-). Show
me a 40-50 year old digital tape or CD medium that still retains all its data
and is readable w/o problems on current machines and doesn't need to be
constantly backed up every several years and I'd be happy to consider it as an
adjunct to film shooting for image capture. Film is proven as an archival
storage medium. Digital is not. Even archivally poor lifed color negs from 50
years ago are still printable to some degree (though not perfect in their
colors). Its on a thread I started at/on rec.photo.film+labs newsgroup whose
title begins anyways as "Oldest color negs w/o significant fading" or somesuch.
Modern color negs, at least Kodacolor negs are likely to lasyt at least that
long if stored properly. My hope/expectation is that Fujichromes, despite what
Henry Wilhelm predicts, will/might last towards the century mark (or at least
80 years) w/o showing significant fading. Way before that time I hopefully
could make back up copies on a medium not yet existing that would put both film
and digital storage longevity to shame. "We shall see..." :-)
>I just restored yesterday some raw-tiff files from tape to a
>secondary drive, 10 gig of them in some 2h timespan. My own work
>effort on this was maybe a minute or so. Price for the tapes:
>approx. $10 for 2 DDS2 tapes, DAT is ultra, ultra cheap per gig,
>you may get as many as you want without hurting budget at all
>and even make some redundant copies of the data on tapes by
>a different manufacturer. One possible source for all sorts of
>tapes is:
>
>
http://www.tape.com/cgi-bin/SoftCart.exe/dat.html?L+cassette+cpyf8723+1045
208834
>
>Thomas
Thanks for the URL Thomas - is this the same DAT tape format that some
connoiseurs try to store music on? How long has this tape format been around?