Scanning 35mm Slides
Richard Melville
My main purpose is to put them on CD and show them on my large screen TV.
So, the question is ---- is there any advantage to scanning at a resolution
any higher than that which matches the TV, which is 1080 lines?
Since the TV is a DLP, my first inclination is to scan at about 1200dpi and
convert the result to jpg with 1080 pixel heights.
Alan Browne
As long as you're scanning, you can make archival copies at higher res.
I'd recommend at least 3000 - 4000 dpi for such purpose. Then if
somebody watching wants a print, you'll have the source material ready
to go.
Secondly, if you scan at higher res, you can "crop" out more interest
for the slide show on the television.
Thirdly, as long as you're doing this, use archival quality CD/DVD
materials (aka: "Gold") they will last for at least the rest of your
lifetime (and then some). Ordinary CD/DVD's will have a life span of
about 5 - 10 years and then start to become unreadable.
Fourthly, before scanning, winnow the slides down to a a good
collection, rather than every photo.
philo
but if you want to try a few at 1200 dpi
and 3600 dpi you can easily compare
Charles E Hardwidge
news:PMqdnfanV8gpdGTX...@giganews.com...
> As long as you're scanning, you can make archival copies at higher res.
> I'd recommend at least 3000 - 4000 dpi for such purpose. Then if somebody
> watching wants a print, you'll have the source material ready to go.
I've been trying to get a grip on the lines per inch of negatives and print,
I'm getting the impression that a 4800 line scan is the highest you need to
go for slides. I'm not sure what it is for print.
http://photo.net/film-and-processing-forum/00UorA
The other thing that leaps out is multiple scans for noise reduction and
pulling detail out of the darks. I had a play with multiple exposures last
night on my relatively poor G9 and I'm impressed.
--
Charles E Hardwidge
(PeteCresswell)
>So, the question is ---- is there any advantage to scanning at a resolution
>any higher than that which matches the TV, which is 1080 lines?
Disk space is cheap. Scanning at max rez preserves your
options: enlargement, cropping, and so-forth.
--
PeteCresswell
gpsman
wrote:
>
> Thirdly, as long as you're doing this, use archival quality CD/DVD
> materials (aka: "Gold") they will last for at least the rest of your
> lifetime (and then some).
Seems a bit of an extreme conclusion considering the lack of evidence.
-----
- gpsman
Charles E Hardwidge
news:16d01957-5126-4563...@s15g2000yqs.googlegroups.com...
Physics and statistics can do wonders. I had a whole bunch of dye based DVDs
go dud on me. As per theory, they're not great in practice.
If you're careful you need to test the media periodically, and if you're
really paranoid store the data with recovery files like PAR2.
The truly paranoid would use computers with error checking (ECC) memory but
I doubt anyone in here has anything that important.
--
Charles E Hardwidge
Wolfgang Weisselberg
> Secondly, [...]
> Thirdly, as long as you're doing this, use archival quality CD/DVD
> materials (aka: "Gold") they will last for at least the rest of your
> lifetime (and then some). Ordinary CD/DVD's will have a life span of
> about 5 - 10 years and then start to become unreadable.
And have a look at the (free) software called dvdisaster, which can
create ECC data for your CDs/DVDs/BDs. If later on you find _some_
read errors, the ECC data can be used to recover the missing data.
And *do* bitwise compare your freshly burned disks (and, if
you can, see how hard the error correction had to work, i.e. C2
scanning for CDs, PI/PO scans for DVD, ..., reading speed via
e.g. dvdisaster, ...)
I've seen too many DVDs which didn't read back correctly after
burning to blindly trust them even with dvdisaster backing me.
> Fourthly, [...]
Fifthly, higher resolution means you can zoom into an image.
Either by remote control, or creating a video stream with music,
crossfades and so on and ... zooming in and out of image areas.
Sixtly, there may come a day in your lifetime when TV changes
to yet higher resolution or we see TV mostly on wallscreens,
with beamers of ever higher resolutions or receiving the data
via internet and watching it on ever larger computer screens.
Would you really want to redo the whole scanning in 10 years?
Seventhly, should old age, water damage, fire, etc. destroy your
slides (may it never happen), you'd still have a backup copy.
You can thus ask a service to create duplicates (if you are asked
for a slide) or to recreate your originals.
-Wolfgang
Alan Browne
Not at all. The tests done in accelerated life (eg: very harsh)
conditions show conclusively that the gold disks will outlast the
ordinary disks by a considerable multiple in ordinary conditions.
http://www.conservationresources.com/Main/section_6/section6_12_files/image004.gif
Richard Melville
Thanks to all for the suggestions. I guess it is probably worthwhile to go a
high resolution for archival purposes.
Albert Ross
Agreed, scanning at the highest res possible is slow but you'd only
need to do it once.
I back my stuff up onto at least two different HDs of different makes
(NOT Seagates) and burn to Verbatim high end DVD and DVD-RAM for luck
(the latter are slower to burn but have higher error correction)
Albert Ross
IME using my now vintage Minolta, sometimes there's extra data there
at 5400 dpi over 2700 and sometimes there ain't, depending on the
emulsion, the lenses, lack of camera shake, distance blur etc.
16 bit scans and multiple scans may get more information out of a
difficult slide but I tend not to use them as routine due to the
extended scanning and processing times and huger files. Like
everything in photography it's a compromise. On most material
"domestic" kit like Minoltas and Nikons can come very close to drum
scan quality at far lower cost and faffing around.
Charles E Hardwidge
news:8hntf59ij84h86i0s...@4ax.com...
> Agreed, scanning at the highest res possible is slow but you'd only
> need to do it once.
>
> I back my stuff up onto at least two different HDs of different makes
> (NOT Seagates) and burn to Verbatim high end DVD and DVD-RAM for luck
> (the latter are slower to burn but have higher error correction)
I've been trying to find the highest sensible scanning limit. It looks like
4000 lines per inch is enough for the best slide media.
Check out PAR2 files. They're a bunch of data recovery files you can store
alongside media. If there's any corruption you can recover data.
Links:
--
Charles E Hardwidge
Charles E Hardwidge
"Albert Ross" <sp...@devnull.com.invalid> wrote in message
I hear what you're saying. It just bothers me there's no definitive guide to
scanning and most of the recommendations are wild ass guesses.
I understand there can be a lot of surprisingly usable data in old prints so
a higher scan level can be useful there as well.
--
Charles E Hardwidge
Robert Peirce
"Charles E Hardwidge" <bo...@invalid.invalid> wrote:
> I hear what you're saying. It just bothers me there's no definitive guide to
> scanning and most of the recommendations are wild ass guesses.
A Few Scanning Tips by Wayne Fulton (http://www.scantimps.com)
Noons
> "Albert Ross" <sp...@devnull.com.invalid> wrote in message
> news:8hntf59ij84h86i0s...@4ax.com...
>
>> Agreed, scanning at the highest res possible is slow but you'd only
>> need to do it once.
I'd recommend you keep the film for longer than that. I've been scanning for
quite a few years - still mostly use film - and quality of scanner results is so
much better now it beats anything I did before at 4000 by a country mile. Had I
thrown away the older film, I'd be kicking myself now. As is, I'm re-scanning
some of the 4000 stuff I did years ago and the results are waaaay better. There
are no absolutes and no one has the "final technical solution" when it comes to
scanning.
> I've been trying to find the highest sensible scanning limit. It looks
> like 4000 lines per inch is enough for the best slide media.
On the other hand, modern b&w 35mm often widely go over 4000. I've got quite a
few Adox CMS20 negatives that far exceed what a Nikon Coolscan 9000 ED can do at
4000 and that is one of the best scanners in the market. Pan F and the older
Kodak Tech Pan is close to matching Adox. Same with some of the boutique new
b&w films from Rollei and others.
Some 35mm slide film gets pretty close if not over, namely Astia 100 and Velvia
50. And in colour negative the new Ektar 100 is as close as can get to 4000
definition.
But it all highly depends on how much care was taken with focusing, avoiding
camera vibration, lens quality, light quality, etcetc. Just like with a high
rez digital: same concerns. If you don't take care of those parameters with a
22Mpixel dslr, you'll never achieve what that sensor can do.
Likely, older film material will rarely exceed 4000 unless it is b&w.
Alan Browne
That's www.scantips.com, actually.
As to Hardwidge's statement, as I've scanned well over 5000 slides and
negatives (on one scanner alone) and I'm on my 3rd film scanner, I think
any statement I make on the matter is, if not authoritative, at least
supported by factual experience and large prints.
Wolfgang Weisselberg
> Not at all. The tests done in accelerated life (eg: very harsh)
> conditions show conclusively that the gold disks will outlast the
> ordinary disks by a considerable multiple in ordinary conditions.
That does not absolve you from checking your archives from time to
time and copy to new media as the old ones near their end of life.
How to detect "near their end of life" reliably is, of course,
not quite trivial.
-Wolfgang
Charles E Hardwidge
It's a claim, and I've yet to see coherent theory and hard numbers from you.
I've brought down bigger assholes than you, Alan.
"Trust but verify."
--
Charles E Hardwidge
Charles E Hardwidge
news:hdnf88$6kh$1...@news.eternal-september.org...
That sounds about right. Robert made a good suggestion to look at
http://www.scantips.com/ but it waffles and there's nothing in there on
scanning B&W print.
Personally, I don't have much stuff to scan and am in no rush to do it but
I've got some B&W prints that should benefit from a high LPI.
99.9% of my scanning is just document archival.
--
Charles E Hardwidge
swandy
wrote:
d to do it once.
>
>I'd recommend you keep the film for longer than that. I've been scanning for
>quite a few years - still mostly use film - and quality of scanner results is so
>much better now it beats anything I did before
This is so true. I spent time years ago scanning my old slides. Really
wish I still had them to do over as current scanners - even ones made
primarily for home use - are much better, as is the software.
Noons
> That sounds about right. Robert made a good suggestion to look at
> http://www.scantips.com/ but it waffles and there's nothing in there on
> scanning B&W print.
scantips is way out of date and hasn't been updated in ages.
Most of the sites you hear recommended are from folks who at some stage dabbled
around with film scanning, a long, long time ago, as a sideline. Most of them
are grossly out of date, show little images short on detail and have not kept up
with the latest film developments, newest scanning techniques and
post-processing. And remarkably few deal with scanning prints, as you likely
found out by now.
This guy: http://cjeastwd.blogspot.com/ still scans as of today and is an
advanced amateur. I don't totally agree with all his conclusions but I don't
disagree with any of his technique. He uses modern film and good quality
scanners. We've exchanged film before and compared results and I know he's got
no hidden agendas. Ask him a few questions directly, who knows?
In my public gallery - http://wizofoz2k.deviantart.com/gallery/#_browse you'll
find reasonably sized samples of what can be done with careful amateur-level
scanning, modern film and gear and careful image acquisition. Click on each
album and/or each image to see larger samples Make sure your browser has colour
profiling on and the gamma is well adjusted.
I don't know of any sites that exclusively deal with print scanning techniques
but you can try asking in http://www.hybridphoto.com/forums/home.php .
Most of the folks there are very knowledgeable and dedicated and don't engage in
silly scams or jihad wars. I'm in there as well and so is Chris.
> Personally, I don't have much stuff to scan and am in no rush to do it but
> I've got some B&W prints that should benefit from a high LPI.
I think you'd be perfectly fine scanning prints on a flatbed of good quality at
300lpi. If you really want to squeeze the last drop off a very well printed
photo, taken with extreme care of the technique, you might go for 600.
More than that and quite frankly, for photochemical prints, you're into serious
diminishing returns territory.
Just my $0.02, provided for free. ;)
Albert Ross
wrote:
>Charles E Hardwidge wrote,on my timestamp of 15/11/2009 4:17 AM:
>> "Albert Ross" <sp...@devnull.com.invalid> wrote in message
>> news:8hntf59ij84h86i0s...@4ax.com...
>>
>>> Agreed, scanning at the highest res possible is slow but you'd only
>>> need to do it once.
>
>I'd recommend you keep the film for longer than that. I've been scanning for
>quite a few years - still mostly use film - and quality of scanner results is so
>much better now it beats anything I did before at 4000 by a country mile. Had I
>thrown away the older film, I'd be kicking myself now. As is, I'm re-scanning
>some of the 4000 stuff I did years ago and the results are waaaay better. There
>are no absolutes and no one has the "final technical solution" when it comes to
>scanning.
Oh yeah, over time some of my slides (and some I inherited from my
father) have faded and/or grown fungus, I wouldn't even consider
ditching the originals though.
>> I've been trying to find the highest sensible scanning limit. It looks
>> like 4000 lines per inch is enough for the best slide media.
>
>On the other hand, modern b&w 35mm often widely go over 4000. I've got quite a
>few Adox CMS20 negatives that far exceed what a Nikon Coolscan 9000 ED can do at
>4000 and that is one of the best scanners in the market. Pan F and the older
>Kodak Tech Pan is close to matching Adox. Same with some of the boutique new
>b&w films from Rollei and others.
>
>Some 35mm slide film gets pretty close if not over, namely Astia 100 and Velvia
>50. And in colour negative the new Ektar 100 is as close as can get to 4000
>definition.
>
>But it all highly depends on how much care was taken with focusing, avoiding
>camera vibration, lens quality, light quality, etcetc. Just like with a high
>rez digital: same concerns. If you don't take care of those parameters with a
>22Mpixel dslr, you'll never achieve what that sensor can do.
>Likely, older film material will rarely exceed 4000 unless it is b&w.
Agreed! This is very much what I'm also finding with digital, my Canon
5D Mk 2 is pretty much up to medium format quality *as long as all
other factors are controlled* and a colleague's Nikons don't fall far
short - the photographer and the circumstances become the limiting
factor rather than the equipment. IS/VR lenses can improve camera
shake almost as much as a tripod but don't do much for subject
movement or heat/distance haze for example.
Albert Ross
wrote:
>I think you'd be perfectly fine scanning prints on a flatbed of good quality at
>300lpi. If you really want to squeeze the last drop off a very well printed
>photo, taken with extreme care of the technique, you might go for 600.
>More than that and quite frankly, for photochemical prints, you're into serious
>diminishing returns territory.
Quite, I recently did a B&W scan of a textured paper print (fairly
old) and a small and rather faded colour print and was amazed how well
they came out at A2.
I did have some trouble with the colour print which had taken on a
pronounced magenta cast I couldn't entirely eliminate without wrecking
the other colours, but the owner was pleased - a photographer himself,
he'd only given me the job because he couldn't make a decent go of it
himself. I was somewhat disappointed with the result until I saw his -
you can't really put back information when it's gone but you can do a
fair imitation given a few hours to tinker. 600 dpi, some curves and
some sharpening. I'm sure I could have done better - and so could he -
given enough hours to become totally cost-ineffective.
Noons
>> Older films may not be worth scanning at very high res. though, for
>> instance scans of some shots I took in the 1980's on ISO100 Agfachrome were
>> unusable at over 1800x1200 (about 1250dpi).
>
> Yes I'd pretty much go along with that from my experience, and from
> other stuffs I recall reading, including scanner reviews and user
> experience as well as more technical information.
Hmmm, not really. Well: Agfachrome yes, I agree. Depends on the film and
quality of the shot.
This: http://wizofoz2k.deviantart.com/art/smiling-rock-63139798
and this: http://wizofoz2k.deviantart.com/art/Antipodeans-48682464
were taken in 1987 and 1983 respectively, K64. Click on image to see larger. I
have Fujichromes of the same era that are as good. I can confidently put the
full size (around 18MP) images against anything dslr of similar resolution. Of
course: those links show downressed images. But they ain't half bad either.
A lot depends on the lenses used and the care taken with initial exposure and
conservation of the slide.
> Any of you ever find a satisfactory way of dealing with Fuji's
> notorious "pepper grain"?
Yes. Use modern film. ;)
Seriously though: I have never had that problem with either the Epson 4990,
Coolscan VED or 9000ED and any Fuji slide film, except with old Provia film.
The pepper grain was caused by scan aliasing of the "rough" emulsion side of the
older Provia film. "Hard light" sources such as in some LED scanners also
compounded the problem. Some folks used "fog" filters to reduce the problem.
But it never went totally away, no matter what.
The solution for me was to use newer film, which doesn't have this problem. With
any modern Fuji slide film such as Provia 400X, Astia 100 or one of the Velvias
you'll notice the emulsion side is nowadays as shiny as the other side, to the
point only the letters and frame numbers now show which is which.
This has completely eliminated the pepper grain problem, or at least I have not
suffered from it in those films and with my scanners. Kodak's slide and colour
reversal hasn't had this problem for a long time: they "shined up" both sides
ages ago.
The problem is also present in some of the old colour neg Fuji films. Notably,
Reala and Pro160 (NPH). Newer batches don't show it anymore so it appears Fuji
has fixed it there as well. Superia Xtra doesn't have it: again, a modern film.
> My only solution so far has been application of blur and sharpening in
> various degrees, and not making large prints from the affected slides.
Might want to try a Coolscan? ;)
> Plan B is to grow some more arms and legs so I can afford to have them
> put through an Imacon or drum scanner.
Too expensive, too much time to do anything. Coolscans and Epson V7xx series do
a very good job nowadays. Particularly if you refine the post-processing.
Gordon Freeman
> Any of you ever find a satisfactory way of dealing with Fuji's
> notorious "pepper grain"?
>
> If you haven't come across this, a slide which projects perfectly
> appears on scanning to have a very clumpy grain structure mainly in
> skies, bodies of water etc. with fairly even tone.
I'm not sure if this is the same thing, but I tend to have problems with
seemingly grainy blue skies in slide scans despite there being no grain
evident in the original. I put this down to the scanner CCD introducing
noise, since CCDs are not very sensitive to blue light so the blue channel
has to be amplified a lot. (I suppose negs might also suffer grainy skies
since a bright sky will be very dense so you might get photon noise?)
Anyhow the only solution I've found is to blur the blue channel for that
part of the image in photoshop using gaussian blur. It's not perfect, if
you fully eliminate the noise it can give a processed look to the image,
but it helps to some extent. I find blurring noise etc generally is best
done on the R,G,B channels separately since different channels are often
affected to radically different degrees. Another mode worth trying is Lab
mode, where you can blur the chroma more than the luminance and so preserve
detail.
Charles E Hardwidge
"Gordon Freeman" <Gor...@freeman.invalid> wrote in message
news:Xns9CC6C3...@127.0.0.1...
You could try multiple scans like multiple exposures. Align and average
them, and it will kill the noise. I've read that some people have got
results. Let us know if it works out for you.
--
Charles E Hardwidge
Charles E Hardwidge
"Gordon Freeman" <Gor...@freeman.invalid> wrote in message
> "Charles E Hardwidge" <bo...@invalid.invalid> wrote:
>> them, and it will kill the noise. I've read that some people have got
>> results. Let us know if it works out for you.
>>
>
> in the film transport mechanism so the result looked slightly blurred as
> successive scans didn't line up with pixel-level accuracy. If I could do
> it be scanning each line multiple times before moving the film on, rather
> than repeated whole scans, then it would probably be ok, but I don't know
> if the scanner can be made to do this.
That's tricky without some funky maths.
Vuescan has a multiple exposure option. It may be worth a try.
Links:
http://www.hamrick.com/vuescan/html/vuesc28.htm#inputmultiexposure
--
Charles E Hardwidge
Albert Ross
wrote:
>Albert Ross wrote,on my timestamp of 17/11/2009 5:26 AM:
>
>>> Older films may not be worth scanning at very high res. though, for
>>> instance scans of some shots I took in the 1980's on ISO100 Agfachrome were
>>> unusable at over 1800x1200 (about 1250dpi).
>>
>> Yes I'd pretty much go along with that from my experience, and from
>> other stuffs I recall reading, including scanner reviews and user
>> experience as well as more technical information.
>
>Hmmm, not really. Well: Agfachrome yes, I agree. Depends on the film and
>quality of the shot.
>This: http://wizofoz2k.deviantart.com/art/smiling-rock-63139798
>and this: http://wizofoz2k.deviantart.com/art/Antipodeans-48682464
>were taken in 1987 and 1983 respectively, K64. Click on image to see larger. I
>have Fujichromes of the same era that are as good. I can confidently put the
>full size (around 18MP) images against anything dslr of similar resolution. Of
>course: those links show downressed images. But they ain't half bad either.
>A lot depends on the lenses used and the care taken with initial exposure and
>conservation of the slide.
Nice! Yes agreed, some of my stuff comes up well despite its age.
>> Any of you ever find a satisfactory way of dealing with Fuji's
>> notorious "pepper grain"?
>
>Yes. Use modern film. ;)
>
>Seriously though: I have never had that problem with either the Epson 4990,
>Coolscan VED or 9000ED and any Fuji slide film, except with old Provia film.
>The pepper grain was caused by scan aliasing of the "rough" emulsion side of the
>older Provia film. "Hard light" sources such as in some LED scanners also
>compounded the problem. Some folks used "fog" filters to reduce the problem.
>But it never went totally away, no matter what.
>
>The solution for me was to use newer film, which doesn't have this problem. With
>any modern Fuji slide film such as Provia 400X, Astia 100 or one of the Velvias
>you'll notice the emulsion side is nowadays as shiny as the other side, to the
>point only the letters and frame numbers now show which is which.
Yes I've had it on some films but not others apparently of the same
emulsion, what you write makes sense.
>This has completely eliminated the pepper grain problem, or at least I have not
>suffered from it in those films and with my scanners. Kodak's slide and colour
>reversal hasn't had this problem for a long time: they "shined up" both sides
>ages ago.
>
>The problem is also present in some of the old colour neg Fuji films. Notably,
>Reala and Pro160 (NPH). Newer batches don't show it anymore so it appears Fuji
>has fixed it there as well. Superia Xtra doesn't have it: again, a modern film.
Never used colour neg but I could dig out some of my late father's
stuff and see how that compares.
>> My only solution so far has been application of blur and sharpening in
>> various degrees, and not making large prints from the affected slides.
>
>Might want to try a Coolscan? ;)
A possibility, the Minolta gives the impression of sharper and more
accurate scans which is why I bought it, but in this case it's too
sharp in the wrong way. I might upgrade my flatbed at some stage and
see if that can be used instead.
>> Plan B is to grow some more arms and legs so I can afford to have them
>> put through an Imacon or drum scanner.
>
>Too expensive, too much time to do anything. Coolscans and Epson V7xx series do
> a very good job nowadays. Particularly if you refine the post-processing.
Plan C was to buy a Canon 5d Mk 2. NO MORE SPOTTING OUT!!!!!
Albert Ross
Excellent point, it mostly seems to appear in skies and bodies of
water which wouldn't be so affected by being oversmoothed. Where
there's detail the grain is not so evident so leaving the other
channels unblurred would probably help a lot. Thanks!
Noons
>
> Plan C was to buy a Canon 5d Mk 2. NO MORE SPOTTING OUT!!!!!
Personally, I prefer the A900. ¦D
Albert Ross
wrote:
>On Nov 20, 3:19�am, Albert Ross <s...@devnull.com.invalid> wrote:
>
>>
>> Plan C was to buy a Canon 5d Mk 2. NO MORE SPOTTING OUT!!!!!
>
>Personally, I prefer the A900. �D
Looks like good kit
http://www.luminous-landscape.com/reviews/cameras/a900-5dmkii.shtml
It was a friend/colleague's Nikons that actually convinced me digital
had come of age, despite they being cropped sensor jobs in the Real
World there's very little difference in the quality of the output, and
IMO they (and probably other kit also) now comfortably exceed the
quality of 35mm film.
Alan Browne
> "Charles E Hardwidge" <bo...@invalid.invalid> wrote:
> It just bothers me there's no definitive guide to
>> scanning and most of the recommendations are wild ass guesses.
>> I understand there can be a lot of surprisingly usable data in old prints
>
> examining a photo of a test target with a microscope or very powerful
> magnifier. Examining slides I took of a test target a few years ago showed
> me that in ideal circumstances (tripod, flash, good lens, accurate focus,
> fine grain film) there were certainly test target details at visible at
> around 4000 dpi in a 35mm Velvia slide (ie. around 75 line pairs per mm).
>
> OTOH I have scanned hand-held shots of general scenes taken with the same
> film using my 2800 dpi film scanner and the resultant images needed a fair
> bit of sharpening, there didn't appear to be any detail at the 1 pixel
> level, probably due to camera shake. However it may be that some quality is
> lost during the scanning process too: according to Nyquist theory you need
> to sample at slightly more than twice the resolution of the detail you are
> trying to capture, so in theory to capture 2800dpi detail in a slide you
> would need to scan at closer to 6000dpi.
>
> Older films may not be worth scanning at very high res. though, for
> instance scans of some shots I took in the 1980's on ISO100 Agfachrome were
> unusable at over 1800x1200 (about 1250dpi).
Does not compute. That film would easily print to 12x8" so scanning at
4000 dpi should produce fine results.
I've scanned film from the 50's and 60's recently at 4000 dpi with
wonderful results - even wish I still had my Minolta 5400 to go a little
further.
Noons
> It was a friend/colleague's Nikons that actually convinced me digital
> had come of age, despite they being cropped sensor jobs in the Real
> World there's very little difference in the quality of the output, and
> IMO they (and probably other kit also) now comfortably exceed the
> quality of 35mm film.
Not at all. In speed, yes and that has been the case for a long time. It
remains their main advantage.
In colour saturation and plain old resolution, they don't. 24MP of Bayer mask
is not the same as 24MP of resolution, not by a very long shot. The
post-processing looks "clean" but that is of importance for clothes, not photos.
I still beat regularly anything made with a 5dm2 with my plain old 1964 Nikon
F and Astia 100. But that requires attention from the 5dm2 user and myself. If
we aren't careful with all the other dependencies, it won't matter a bit.
Most folks using these high rez cameras continue to do so as if they were point
and shoot, using the same old lenses. Which is probably fine, mind you. But to
get the best out of this gear we need the best of everything, not just the
camera body. Otherwise, it's wasted. Same applies to film, by the way.
On the other hand, not every shot requires absolute maximum resolution. It's
all about the moment, the image, the composition, etcetc. Resolution is one
amongst many things that have to come together for a shot that one can be happy
with. Yet again, not every shot has to be a competition winner in order to be
satisfactory:
http://wizofoz2k.deviantart.com/art/two-religions-two-cathedrals-142906985
(good old noisy, "unclean", 35mm Fuji Pro 800Z, hand-held at 1/10)
Wolfgang Weisselberg
> I still beat regularly anything made with a 5dm2 with my plain old 1964 Nikon
> F and Astia 100.
How hard can you push your Astia 100? :-)
And does it include a video mode?
-Wolfgang
Noons
> How hard can you push your Astia 100? :-)
As hard as it can go. And I can prove it and have done so multiple times, with
the unvarying result that those who issue such comments usually shut up.
I thought I made it VERY CLEAR that I was not talking about high ISOs, when I
said speed was the major advantage of a dslr. I see that was completely and
conveniently lost...
> And does it include a video mode?
I thought we were talking about photography. Shall I bring in 70mm as well?
Alan Browne
> Alan Browne <alan....@FreelunchVideotron.ca> wrote:
>
>> Gordon Freeman wrote:
>>> Older films may not be worth scanning at very high res. though, for
>>> instance scans of some shots I took in the 1980's on ISO100
>>> Agfachrome were unusable at over 1800x1200 (about 1250dpi).
>> Does not compute. That film would easily print to 12x8" so scanning
>> at 4000 dpi should produce fine results.
>
> same as a 150dpi inkjet print; the grain is visible though unobtrusive.
> This is stuff shot around 1982 give or take a year, by the end of the
> decade I agree you'd get more resolution than that.
Care to mail me the film?
Wolfgang Weisselberg
> Wolfgang Weisselberg wrote,on my timestamp of 22/11/2009 1:06 PM:
>> > I still beat regularly anything made with a 5dm2 with my plain old
>> > 1964 Nikon F and Astia 100.
[citation restored]
>> How hard can you push your Astia 100? :-)
> As hard as it can go. And I can prove it and have done so multiple times, with
> the unvarying result that those who issue such comments usually shut up.
Oh, I should have added "with usable results". My fault.
> I thought I made it VERY CLEAR that I was not talking about high ISOs, when I
> said speed was the major advantage of a dslr.
Speed can refer to sooo many things. Like "time from the event to
the printable result in the computer" --- not having to develop the
shots chemically and scanning them is a *distinct* speed advantage.
Ask any newspaper.
> I see that was completely and conveniently lost...
If you meant "I regularly beat anything made with a 5D2 *at ISO
100 or 200* with ... etc.", wouldn't you write that?
>> And does it include a video mode?
> I thought we were talking about photography.
"Reverie" was made with a 5D2.
If you meant "I regularly beat any *still photograph* from a 5D2
at ISO 100 or 200 with ... etc.", why didn't you say so?
> Shall I bring in 70mm as well?
70mm what? Lenses (short portrait range for 35mm film)?
Lenses (similar view angle to a 112mm for 35mm film in 1.6x
crop cameras)? Lenses (image circle: just shy of 5x5cm² film)?
Film size (for example 7x10cm)? Film strip width (as in "sometimes
used in moving picture shooting")? Bullet diameter ('please
kill me very messily with your artillery')??
And how would that work with your 1964 Nikon F, unless it's a
70mm focal length lens?
If your messages have multiple possible meanings, don't be too
surprised when not everyone picks the one you meant.
-Wolfgang
Noons
>> I thought I made it VERY CLEAR that I was not talking about high ISOs, when I
>> said speed was the major advantage of a dslr.
>
> Speed can refer to sooo many things. Like "time from the event to
> the printable result in the computer" --- not having to develop the
> shots chemically and scanning them is a *distinct* speed advantage.
> Ask any newspaper.
True. And I failed to note I was talking about ISO, not speed of processing.
Ask a newspaper? Whatever for? I don't work for one nor will I ever and their
constraints are not mine.
> If you meant "I regularly beat anything made with a 5D2 *at ISO
> 100 or 200* with ... etc.", wouldn't you write that?
If I say Astia 100, I'd expect it to be implicit I'm talking about 100 ISO film,
but one never knows with the likes of you...
> "Reverie" was made with a 5D2.
As if no one knew that. And it is a royal POS, as videos go.
Learn a little bit about video technique and you'll realize why.
No wonder Canon pulled the second one: the first was laughed about long and hard
in the video making circles.
> If you meant "I regularly beat any *still photograph* from a 5D2
> at ISO 100 or 200 with ... etc.", why didn't you say so?
Because - once again - I was talking about film. Familiar with the concept of
context?
> 70mm what?
Film. As in motion picture. Similar to video. Context.
> If your messages have multiple possible meanings, don't be too
> surprised when not everyone picks the one you meant.
If you made a small effort to understand that any message has a context and is
not to be taken on literal individual meaning of each word, you'd be noted as a
little bit more than a troll.
As is, <plonk>
Wolfgang Weisselberg
> Wolfgang Weisselberg wrote,on my timestamp of 23/11/2009 6:23 AM:
>> If you meant "I regularly beat anything made with a 5D2 *at ISO
>> 100 or 200* with ... etc.", wouldn't you write that?
> If I say Astia 100, I'd expect it to be implicit I'm talking about 100 ISO film,
> but one never knows with the likes of you...
Sure, you have an ISO 100 film. How does that constrain the
5D2 to low ISO settings?
>> "Reverie" was made with a 5D2.
> As if no one knew that. And it is a royal POS, as videos go.
> Learn a little bit about video technique and you'll realize why.
Sure, you could do much better, which is why you post under your
real and full name. Anyway, all you'd need would be a couple
dozen millions, a large team, huge lighting trucks, tons of other
gear ...
Do it all in 72 hours with no advance warning, no storyboard
prepared, no location scouted and so on, around busy workdays for
everyone, in only available light --- and the worst examples of
that, sodium and mercury vapor lights --- at night, for peanuts,
then we talk.
> No wonder Canon pulled the second one:
Canon *asked* for Reverie to be pulled --- temporarily.
Canon *asked* for Nocture to be pulled --- temporarily.
You want to make something from Canon stumbling over their
own size, again? Feel free.
> the first was laughed about long and hard in the video making circles.
Only by those who could not grasp what they were seeing.
The same people who laugh long and hard about testchart shots
showing unpreceded resolution. The same kind of people who
laughed about 35mm cameras and mobility, as they were shooting
daguerreotypes on full plate size, to be telefaxed --- oops,
fax and remote image transmission will never take off and has no
use whatsoever, I meant "pony expressed" --- to the next newspaper.
The rest of them know what an affordable full frame sensor
with high sensitivity means.
>> If you meant "I regularly beat any *still photograph* from a 5D2
>> at ISO 100 or 200 with ... etc.", why didn't you say so?
> Because - once again - I was talking about film. Familiar with the concept of
> context?
I see. That probably means you really meant "I regularly beat
*myself* anytime when shooting still photographs with a 5D2 at
ISO 100 or 200 with ...". Now, context, are you doing something
wrong when using the 5D2?
>> 70mm what?
> Film. As in motion picture. Similar to video. Context.
I see. I'd like you to do a Nocturne or Reverie --- you even get
to write your own storyboard and cheesy story --- in available
light at night with your 70mm gear. I'll see results on friday,
which is *more* than ample time, even with the developing and
scanning needed.
>> If your messages have multiple possible meanings, don't be too
>> surprised when not everyone picks the one you meant.
> If you made a small effort to understand that any message has a context and is
> not to be taken on literal individual meaning of each word, you'd be noted as a
> little bit more than a troll.
Trying to redefine words and sentences to mean what you want
them to mean is the hallmark of a troll, and I don't pander
to them. I read *exactly* what you write, not more, not
less.
And yes, I aim higher than "a little bit more than a troll".
> As is, <plonk>
Oh, plonked by a "No-ones". How devastating!
He didn't like critique.
-Wolfgang
Albert Ross
wrote:
>Albert Ross wrote,on my timestamp of 22/11/2009 3:39 AM:
>
>> It was a friend/colleague's Nikons that actually convinced me digital
>> had come of age, despite they being cropped sensor jobs in the Real
>> World there's very little difference in the quality of the output, and
>> IMO they (and probably other kit also) now comfortably exceed the
>> quality of 35mm film.
>
>Not at all. In speed, yes and that has been the case for a long time. It
>remains their main advantage.
Partly agree, I'm getting results at 200ASA on the Canon that IMO
equals or exceeds Velvia 100, and results at higher ISOs that don't
fall that far short.
>In colour saturation and plain old resolution, they don't. 24MP of Bayer mask
>is not the same as 24MP of resolution, not by a very long shot. The
>post-processing looks "clean" but that is of importance for clothes, not photos.
> I still beat regularly anything made with a 5dm2 with my plain old 1964 Nikon
>F and Astia 100. But that requires attention from the 5dm2 user and myself. If
>we aren't careful with all the other dependencies, it won't matter a bit.
Partly disagree, there may be fewer pixels than a 5400 dpi scan of a
slide but IME the quality of the pixels, ie. the information, is
higher (using camera RAW and converting to TIFF). Add in the reduced
need for post processing.
>Most folks using these high rez cameras continue to do so as if they were point
>and shoot, using the same old lenses. Which is probably fine, mind you. But to
>get the best out of this gear we need the best of everything, not just the
>camera body. Otherwise, it's wasted. Same applies to film, by the way.
Quite. These sort of resolutions are very unforgiving of lens and
photographer quality, especially camera shake, subject movement,
distance blur/haze etc.
>On the other hand, not every shot requires absolute maximum resolution. It's
>all about the moment, the image, the composition, etcetc. Resolution is one
>amongst many things that have to come together for a shot that one can be happy
>with. Yet again, not every shot has to be a competition winner in order to be
>satisfactory:
>http://wizofoz2k.deviantart.com/art/two-religions-two-cathedrals-142906985
>(good old noisy, "unclean", 35mm Fuji Pro 800Z, hand-held at 1/10)
Definitely! I've produced A2 prints from about a quarter of a frame
which have had rave reviews, and prints/cards which are not
technically/photographically top quality but make people go "wow!"
At the end of the day it's all a compromise. What I want is 5x4
quality, with 35mm equivalent lenses from 12 to 2000 mm f2, at film
speeds of 200 - 800ASA, which folds up and goes in my pocket . . .
Noons
> Partly agree, I'm getting results at 200ASA on the Canon that IMO
> equals or exceeds Velvia 100, and results at higher ISOs that don't
> fall that far short.
Sure. But Velvia 100 is not my preferred film, so I can't comment there.
With higher ISOs than about 800, I'm totally sold on dslrs. They can achieve
much better than film. Is that enough for me to switch? Not really.
> Partly disagree, there may be fewer pixels than a 5400 dpi scan of a
> slide but IME the quality of the pixels, ie. the information, is
> higher (using camera RAW and converting to TIFF). Add in the reduced
> need for post processing.
What you just described *is* post-processing! There is no way in the world
you'll be able to do anything with the RAW file unless you post-process it - be
that with ACR + PS or whatever.
I do the same with scanned film: it's post-processed, a lot of the same ACR and
PS does with the exception of de-mosaic, which I don't need. But noise
reduction, rotation/cropping, curve and other adjustments, save in Tiff and
final output downress/sharpening? Exactly the same as I do with my dslr RAW files.
> At the end of the day it's all a compromise. What I want is 5x4
> quality, with 35mm equivalent lenses from 12 to 2000 mm f2, at film
> speeds of 200 - 800ASA, which folds up and goes in my pocket . . .
See? Forget dslrs. What you need is micro 4/3!
;)
Michael
By that do you include the Olympus Pen Digital?
--
Michael
Noons
I said micro 4/3, didn't I?
And, what part of " ;)" do you find hard to grasp?
Albert Ross
wrote:
>Albert Ross wrote,on my timestamp of 29/11/2009 3:40 AM:
>
>> Partly agree, I'm getting results at 200ASA on the Canon that IMO
>> equals or exceeds Velvia 100, and results at higher ISOs that don't
>> fall that far short.
>
>Sure. But Velvia 100 is not my preferred film, so I can't comment there.
>With higher ISOs than about 800, I'm totally sold on dslrs. They can achieve
>much better than film. Is that enough for me to switch? Not really.
So use both! I still have my film body but I simply haven't been using
it.
>> Partly disagree, there may be fewer pixels than a 5400 dpi scan of a
>> slide but IME the quality of the pixels, ie. the information, is
>> higher (using camera RAW and converting to TIFF). Add in the reduced
>> need for post processing.
>
>What you just described *is* post-processing! There is no way in the world
>you'll be able to do anything with the RAW file unless you post-process it - be
>that with ACR + PS or whatever.
>
>I do the same with scanned film: it's post-processed, a lot of the same ACR and
>PS does with the exception of de-mosaic, which I don't need. But noise
>reduction, rotation/cropping, curve and other adjustments, save in Tiff and
>final output downress/sharpening? Exactly the same as I do with my dslr RAW files.
Perhaps you should learn to read what I meant, not what I wrote <G> I
meant there's LESS post-processing required. Principally no spotting
out of dust, but because the sensor is more consistent than different
brands of film the tweaks I need to apply are not so dissimilar, I end
up doing much the same sequence to every frame with relatively small
differences.
John Sisker
news:LslKm.6155$rE5....@newsfe08.iad...
>I have many old (going back 50 years) slides that I am planning to
>digitize.
>
> My main purpose is to put them on CD and show them on my large screen TV.
>
> So, the question is ---- is there any advantage to scanning at a
> resolution any higher than that which matches the TV, which is 1080 lines?
>
> Since the TV is a DLP, my first inclination is to scan at about 1200dpi
> and convert the result to jpg with 1080 pixel heights.
>
>
>
Richard,
I don't want to get into the technical aspects of such a scanner, for others
have already answered that far better than I could anyway. However, less
then a year ago, I purchased such a slide scanner to, like you, convert
hundreds of slides into a digital format. While this scanner does have many
options for choosing the resolution, I did forget about one thing, that
actually made a big difference. And that was the color.
Most of my slides were taken on a cross-country trip in 1973, and include
such things as the Grand Canyon, Carlsbad Caverns, Walt Disney World, to
name a few, in addition to many of the towns and scenery from a trip to
Colorado. The one thing that really stood out about my original slides was
the vivid colors. In spite of the fact that I had these slides stored in a
dark closest, apparently the colors had faded somewhat over time. Even
though I could enhance the colors somewhat with the slide scanner and
programs such as Photoshop, it wasn't quite the same. Besides, I am
certainly not going to devote the time necessary it would take to bring
hundreds of slides 'back to life.' They came out alright, but not like the
original originals... when I first took them.On the other hand, I'm sure
no-one will notice, for they never saw the originals for comparison. But I
have, and still remain somewhat disappointed even today.
John Sisker
"Pose & Go" Photography
www.tinplatedesign.com
Noons
> So use both! I still have my film body but I simply haven't been using
> it.
Yup, what I do. The D80 before and now the D200 get a lot of use for
macro stuff as well as some of the kid's sports. The film gear is
used for just about everything else.
> Perhaps you should learn to read what I meant, not what I wrote <G> I
Ooops... Sorry!
;)
> meant there's LESS post-processing required. Principally no spotting
> out of dust, but because the sensor is more consistent than different
> brands of film the tweaks I need to apply are not so dissimilar, I end
> up doing much the same sequence to every frame with relatively small
> differences.
One of the things I found out the hard way is when using RAW I spend
easily as long in each dslr frame as I do post-processing a scanned
frame. Mind you, perhaps with things like Capture-NX and such it'd be
faster? I know with ufraw it isn't. I must try vuescan for RAW one
of these days. Then again, I'm not necessarily after speed of
workflow as I'm an amateur: workflow time is largely immaterial and
all part of the fun!
¦D
Albert Ross
wrote:
>One of the things I found out the hard way is when using RAW I spend
>easily as long in each dslr frame as I do post-processing a scanned
>frame. Mind you, perhaps with things like Capture-NX and such it'd be
>faster? I know with ufraw it isn't. I must try vuescan for RAW one
>of these days. Then again, I'm not necessarily after speed of
>workflow as I'm an amateur: workflow time is largely immaterial and
>all part of the fun!
>�D
The thing I like about the Canon software is that the changes are
non-destructive.
I backup the original RAWS and jpgs and work on copies anyway, but
it's easy to back out of the changes you've made and retry if you
change your mind later. Particularly useful is to back out of
sharpening and redo it for different print sizes.
Albert Ross
wrote:
> I did forget about one thing, that
>actually made a big difference. And that was the color.
>
>Most of my slides were taken on a cross-country trip in 1973, and include
>such things as the Grand Canyon, Carlsbad Caverns, Walt Disney World, to
>name a few, in addition to many of the towns and scenery from a trip to
>Colorado. The one thing that really stood out about my original slides was
>the vivid colors. In spite of the fact that I had these slides stored in a
>dark closest, apparently the colors had faded somewhat over time. Even
>though I could enhance the colors somewhat with the slide scanner and
>programs such as Photoshop, it wasn't quite the same. Besides, I am
>certainly not going to devote the time necessary it would take to bring
>hundreds of slides 'back to life.' They came out alright, but not like the
>original originals... when I first took them.On the other hand, I'm sure
>no-one will notice, for they never saw the originals for comparison. But I
>have, and still remain somewhat disappointed even today.
That's sadly rather inevitable, all dyes (even those used for CDs and
DVDs) tend to degrade over time.
Different emulsions from my collection have degraded differently, and
some of my old man's original Kodachromes still look pretty good while
his Ektachromes and Agfas etc. have gone downhill to varying degrees.
Mother had a collection of postcards from around the world stuck on
her kitchen wall and there the effect was even more extreme, some
faded rapidly and turned cyan while others retained their original
vividness for years despite the rather hostile environment.
Alan Browne
That's no different than preparing and keeping an edited, but non
sharpened full sized image and then always going to that one for display
or print at any re-size.
Noons
> The thing I like about the Canon software is that the changes are
> non-destructive.
>
> I backup the original RAWS and jpgs and work on copies anyway, but
> it's easy to back out of the changes you've made and retry if you
> change your mind later. Particularly useful is to back out of
> sharpening and redo it for different print sizes.
I don't think that's just the Canon software.
Certainly ufraw lets me backout as many times as I want and try other
combinations. Haven't used Capture that much, but I'd say it's exactly the same
and so would be ACR?
As for sharpening, I do it as many times as I want with different settings, with
Focus Magic. All I have to do is call the plug-in on the image.
Albert Ross
Indeed, that's what I do with my slide scans, save a nearly finished
TIFF alongside the original unaltered scan.
Albert Ross
wrote:
>Albert Ross wrote,on my timestamp of 4/12/2009 7:24 AM:
>
>> The thing I like about the Canon software is that the changes are
>> non-destructive.
>>
>> I backup the original RAWS and jpgs and work on copies anyway, but
>> it's easy to back out of the changes you've made and retry if you
>> change your mind later. Particularly useful is to back out of
>> sharpening and redo it for different print sizes.
>
>I don't think that's just the Canon software.
>Certainly ufraw lets me backout as many times as I want and try other
>combinations. Haven't used Capture that much, but I'd say it's exactly the same
>and so would be ACR?
I haven't (yet) played with any other RAW software. I suppose I
should. What's your recommendation, ufraw?
I've reached the age/stage where I have more problems learning a
similar but different software than I do learning something completely
different. I haven't even explored a lot of the camera's features yet.
It seems the Canon software also does non-destructive editing of TIFFs
and jpgs, I haven't really explored this yet. It has a neatly
integrated curves/histogram display, where the histogram is tilted
such that getting it under the curve gets most of the job done, but it
has a serious lack of bells and whistles.
So many features, so few brain cells . . .
Noons
>
> I haven't (yet) played with any other RAW software. I suppose I
> should. What's your recommendation, ufraw?
I use ufraw mostly for financial reasons: most of the RAW processors for Nikon
are exceedingly expensive, while that one is free. It's far from perfect and
it's hard to get the correct profiles for it, but it does the job and I'm now
very familiar with its little quirks. Wouldn't recommend it over and above a
manufacturer's own solution like Canon's. Also use vuescan every once in a
while - I have a full licence for film scanning: it does a passable job with RAW
NEF and seems to have less problems with profiling my D200.
> I've reached the age/stage where I have more problems learning a
> similar but different software than I do learning something completely
> different. I haven't even explored a lot of the camera's features yet.
Exactly. Same here. Hence why I don't recommend anything outside what one is
already comfortable or familiar with.
> It seems the Canon software also does non-destructive editing of TIFFs
> and jpgs, I haven't really explored this yet. It has a neatly
> integrated curves/histogram display, where the histogram is tilted
> such that getting it under the curve gets most of the job done, but it
> has a serious lack of bells and whistles.
Sorry, not familiar with it.
> So many features, so few brain cells . . .
Bingo! ;)