dumb question
P§ł
I've been out of photography for many years so this digital stuff is new to
me so please forgive the dumb question. I'm confused about the maximum image
size one can get out of a camera like the Nikon d700. What's the largest
quality PRINT one can make with this beast.
Thanks
Pete
--
P㎡
Andrew Koenig
news:uktkk.554$Ht4.171@trnddc01...
> I've been out of photography for many years so this digital stuff is new
> to me so please forgive the dumb question. I'm confused about the maximum
> image size one can get out of a camera like the Nikon d700. What's the
> largest quality PRINT one can make with this beast.
How sharp do you want the print to be?
The basic image size is 2832 x 4256 pixels, which is 9.4 x 14 inches at 300
dots/inch or 14x21 inches at 200 dots/inch. For bigger prints than that,
you would probably want to use some kind of dithering/upscaling tool such as
Genuine Fractals.
John McWilliams
> "P§³" <xvzex...@verizon.net> wrote in message
Depends on the subject matter; GF may well not be necessary at larger
sizes.
Also, printers made in the last few years never print as low as 300 dpi;
you were speaking of pixels per inch, not dots. 300 ppi is a fairly
standard hi-rez number many print at, but the printers will be
outputting in excess of 720 dpi in most cases.
--
john mcwilliams
Scott W
closely at all. I have made a lot of 12x18 inch prints with a 8MP
camera that most people think look very sharp, they look a bit soft to
me, but then I am fairly near sighted.
My simple rule is this, if you have a fixed number of pixels in
general printing larger will make the photo look better. If you are
going to print at a given size then more pixels will be better then
fewer. Both of those rules clearly have limits.
D700, with a decent lens, should produce a good sharp looking 12x18
inch print. A larger print, say 20x30 inches, will still look good if
not viewed up close.
Scott
David J. Littleboy
"Scott W" <bip...@hotmail.com> wrote:
What Scott said. The above is exactly right.
The only point to be added is that prints from the D700 will be
significantly better than anything you've ever seen from 35mm film, and not
quite as good as the best an unrepentant grain-sniffer can get from 6x7 and
larger.
--
David J. Littleboy
Tokyo, Japan
Jürgen Exner
A4 is certainly no problem.
Even A3 or larger, up to poster size, is ok because typically you don't
view prints that large from close-up but from some distance.
jue
ransley
Years ago when I got a digital in 05 I saw a 12x18 blowup of a motor
from a Nikon 4mp, I thought it was great.Film is different. I did a
75% crop of a 5mp and blew it to 8x10, not good detail but it looks
good.
David J Taylor
From what distance are you viewing the print? Impossible to answer the
question otherwise. You can make a billboard-size print from a 2 or 3MP
camera, and it will look fine when viewed from the appropriate distance.
David
D-Mac
This straight from Adobe's web site and no doubt any copyright issues
that arise will see it © Copyright Adobe. you can increase the print
size by reducing the PPI. You can also double the print size by using
Interpolation with no detectable quality loss in the print.
Mega pixel File size (8bit, RGB) Pixels Example Print dimensions (at 300
ppi)
4.1 11.5 mb 2464 x 1632 Nikon D2Hs 8.213 x 5.44 in
8 34.7 3456 x 2304 Canon Digital Rebel XT 11.52 x 7.68 in
12.34 34.7 4256 x 2848 Fuji Fine Pix 3 14.187 x 9.493 in
12.4 34.9 4288 x 2848 Nikon D2X 14.293 x 9.493 in
16.7 47.5 4992 x 3328 Canon EOS 1Ds Mark II 16.64 x 11.093 in
22 62.2 5356 x 4056 Leaf Valeo 22Wi 17.853 x 13.52 in
22 65.1 5488 x 4145 Phase One P 25 18.293 x 13.817 in
--
visit www.D-Mac.info
to relieve the tension...
Usenet is after all Usenet!
HEMI-Powered
No one can answer that question correctly or intelligently unless
you provide some criteria for the answer such as the subject(s)
you want to shoot, lighting conditions, len(es) to be used, and
what your quantitative or qualitative definitions of "good" or
"bad" might be. e.g., one person may be satisfied with a print
that others abhor.
Now, it is at least fair to discuss what is the optimum or even
the minimum PPI for a "good" print IF one can also assume that
the camera to be used is up to the task, meaning that it's lens,
sensors, and software can actually provide a quality image at a
given size in pixels. i.e., an 8 MP P & S is unlikely to provide
anywhere near as good an image as a quality 8 MP DSLR.
I have found that depending on the subject and depending on the
viewing distance that as little as 100 PPI can result in an
acceptable print while 150, 200, maybe higher is desirable if
possible. Do the math yourself to translate PPI to actual print
size if you want one man's opinion of "good" quality.
--
HP, aka Jerry
"Laid off yet? Keep buying foreign, you will be!" - increasingly
popular bumper sticker
HEMI-Powered
jour ...
> "P§ł" <xvzex...@verizon.net> wrote in message
DPI is an incorrect unit designation; DPI is most commonly used
for scanning where the hardware scans X dots per linear inch of
material, where each dot turns into a pixel. In digital printing
or processing, everything is already in pixels and is re-
converted to a spot on the paper, hence the proper units are PPI
or Pixels Per Inch. Thus, one can avoide the mental gear change
and think only in pixels unless and until the image information
is actually transferred to a print.
HEMI-Powered
...
> On Jul 31, 3:06 pm, "P§³" <xvzex3qt...@verizon.net> wrote:
>> Dumb question;
>> I've been out of photography for many years so this digital
>> stuff is new to me so please forgive the dumb question. I'm
>> confused about the maximum im age size one can get out of a
>> camera like the Nikon d700. What's the large st quality
>> PRINT one can make with this beast. Thanks
> It is amazing how most people really don’t look at a
> photograph very
> closely at all. I have made a lot of 12x18 inch prints with a
> 8MP camera that most people think look very sharp, they look a
> bit soft to me, but then I am fairly near sighted.
It's probably a good thing that viewers never looked closely at
the prints of even the famous 35mm photographers as they may have
been disappointed. The original rules for DOF were based on
circles of confusion which in turn had an empirical relationship
to the ability of the human eye to discern fine detail. And, this
concept can be extended to prints by using the eye's ability to
resolve/discern detail at a given viewing distance. Thus, the
farther away a print of a given size is viewed, the
fuzzier/blurrier it can be and still be perceived as "sharp" but
if one puts their nose up againt the print or even uses a
magnifying glass, all sorts of defects can easily be seen and may
even ruin the expectations of the viewer.
I am a pragmatist not a theortician thus it matters far less to
me what can be scientifically measured or even what prevailing
wisdom may require for PPI. About the only thing that is really
important to me is if my prints or those of others LOOK sharp at
the normal distances I view them. i.e., most people tend to view
4 x 6 prints rather close up but 8 x 10, 11 x 14, 16 x 20 ...
prints at increasing distances.
Years ago I put up Super B 13 x 19 prints of cars on the outer
walls of my office at work. They were made with an HP 1220C
printer, good in its day but certainly crap by today's standards.
All of these prints were made from images that provided only 75-
100 PPI. I also put up 8.5 x 11 prints on the rear wall of my
office. These had PPIs in the range of 125-150 PPI. Many people
passing by would stare at my prints and sometimes come in to
comment on them and ask where I was getting them developed
thinking they were film enlargements. This is far less bragging
up my scanning or photographic ability and much more that the
viewers were looking at the big prints from many feet away and
usually looking at the smaller ones at a distance of 1 to 2 feet.
> My simple rule is this, if you have a fixed number of pixels
> in general printing larger will make the photo look better.
Come again? The larger one prints from any given number of
pixels, the more and more degraded the print quality will become.
Whether it is or is not objectionable is the issue.
> If you are going to print at a given size then more pixels
> will be better then fewer. Both of those rules clearly have
> limits.
>
> D700, with a decent lens, should produce a good sharp looking
> 12x18
> inch print. A larger print, say 20x30 inches, will still look
> good if not viewed up close.
You don't say how many MP are needed for a "good" 20 x 30 print
but if one assumes 200 PPI then it would have to be 4000 x 6000
or about 12 MP, a very large image even today. Yes, modern DSLRs
can do this and more but even quality cameras with quality lenses
cannot always guarantee a "good"image. Now, once you add the "if
not viewed up close" disclaimer - which I highly agree with -
then I would say that, yes, a really large print might even be
possible as low as 4 MP. Now, take a state-of-the art P & S or
EVF camera in the 8-10 MP range and try the same scene and see
the difference.
HEMI-Powered
jour ...
> P㎡ wrote:
A very good analogy, David. If one looks at even a modern
billboard (if you can find one!) the size of the printed "dots"
is often so large that viewing close up makes the entire picture
look like just so much mush.
Scott W
12,18 and 20x30 and see which print people prefer.
> > If you are going to print at a given size then more pixels
> > will be better then fewer. Both of those rules clearly have
> > limits.
>
> > D700, with a decent lens, should produce a good sharp looking
> > 12x18
> > inch print. A larger print, say 20x30 inches, will still look
> > good if not viewed up close.
>
> You don't say how many MP are needed for a "good" 20 x 30 print
> but if one assumes 200 PPI then it would have to be 4000 x 6000
> or about 12 MP, a very large image even today.
A really good 20x30 inch print need somewhere around 54MP, assuming
300 PPI.
With stitched images a 54MP image is on the small side.
Scott
fr...@home.com
>Andrew Koenig added these comments in the current discussion du
>jour ...
>
>> "P㎡" <xvzex...@verizon.net> wrote in message
>> news:uktkk.554$Ht4.171@trnddc01...
>>
>>> I've been out of photography for many years so this digital
>>> stuff is new to me so please forgive the dumb question. I'm
>>> confused about the maximum image size one can get out of a
>>> camera like the Nikon d700. What's the largest quality PRINT
>>> one can make with this beast.
>>
>> How sharp do you want the print to be?
>>
>> The basic image size is 2832 x 4256 pixels, which is 9.4 x 14
>> inches at 300 dots/inch or 14x21 inches at 200 dots/inch. For
>> bigger prints than that, you would probably want to use some
>> kind of dithering/upscaling tool such as Genuine Fractals.
>>
>DPI is an incorrect unit designation; DPI is most commonly used
>for scanning where the hardware scans X dots per linear inch of
>material, where each dot turns into a pixel. In digital printing
>or processing, everything is already in pixels and is re-
>converted to a spot on the paper, hence the proper units are PPI
>or Pixels Per Inch.
Wow are YOU wrong!!!
Pixels are reserved for MONITOR DISPLAYS!
Printers are ALLWAYS rated dots per inch! My Dad agrees, and he's been a
printer for 40 years... look up halftone mask...
John McWilliams
There are a lot of folk in the printing business who always speak in
dpi, and with good reason.
A digital image, is pixels only, and remains pixels until it's sent to
the RIP or printer driver. There the pixels are converted to commands to
produce spurts of ink, usually in the 600 dpi range and up on inkjets.
Halftones are way less than that, but doesn't play into these equations.
Once on paper, the correct term is indeed dots per inch.
--
John McWilliams
Ray Fischer
It's been my experience that the limiting factor isn't the camera but
the quality of the lens and the ability of the photographer. Focus
problems, lens sharpness, and steadiness all tend to have a bigger
effect than the camera.
That said, it depends upon the subject matter and your fussiness.
I've printed a 6MP image at 20" by 30" and it looks quite nice.
Subjectively, the larger the print the lower the resolution the eye
can accept. With bigger prints people tend to stand back further.
For a big print you can get away with 100 pxels per inch. For a
small print you'd want 300 PPI.
--
Ray Fischer
rfis...@sonic.net
David J Taylor
[]
> There are a lot of folk in the printing business who always speak in
> dpi, and with good reason.
>
> A digital image, is pixels only, and remains pixels until it's sent to
> the RIP or printer driver. There the pixels are converted to commands
> to produce spurts of ink, usually in the 600 dpi range and up on
> inkjets.
> Halftones are way less than that, but doesn't play into these
> equations. Once on paper, the correct term is indeed dots per inch.
When you talk about a dot, are you talking about a constant-sized dot, or
a variable-sized dot? A variable-coloured dot (i.e. a mixture of CMYK) or
a fixed-colour dot (i.e. a single ink colour)?
Methinks that not all "dots" are created equal.
David
Me
DPI of inkjet printers (the number on the sales brochure/box) gives a
totally unreasonable figure, but my "4800x2400 dpi" printer will resolve
about 400 lines per inch horizontally and about 200 lines per inch
vertically of dithered half-tones. So, it's potentially around about
600dpi (about double what most commercial wet-process prints are)
depending what you measure and how.
So, unless printing at a very small size, while DPI may be a "standard",
it's irrelevant when at the size of print ("largest?") the limiting
factor is pixels, not printer resolution, unless you're using something
very old for printing.
Anyway, IMO the most critical factor in "how many megapixels is enough"
(apart from the obvious ones like viewing distance) is composition.
Something drawing the viewer in (landscape) then 300 ppi is desirable,
but portrait/macro/nature etc, it doesn't really matter very much at
all. I've produced (IMO) great prints from 6mp at 19"x13" where
resolution wasn't an obvious limiting factor, but others where above
12x8" approx, 6mp really starts to fall apart.
Colin.D
That figure is not unreasonable. Good printers print at so many pixels
per inch (PPI), the 'native' PPI for the printer; Canon uses 600PPI, and
Epson uses 720 PPI. That is pixels per inch delivered to the printer.
The printer driver software resamples whatever the actual image PPI
happens to be up to the native PPI the printer requires.
The printer, though, lays down on the paper its native dots per inch,
DPI, for my Canon it's 4,800*2,400 dots per inch.
This means that a single pixel at 600 PPI will have 4800/600, or 8 dots
horizontally by 2400/600 or 4 dots vertically, a total of 32 dots of ink
to represent one pixel of the image. The printer will vary the color of
each dot to produce the required blend of ink to reproduce the correct
image color.
Colin D.
HEMI-Powered
> On Fri, 01 Aug 2008 13:03:08 GMT, "HEMI-Powered"
> <no...@none.en> wrote:
>
>>Andrew Koenig added these comments in the current discussion
>>du jour ...
>>
>>> "P§ł" <xvzex...@verizon.net> wrote in message
>>> news:uktkk.554$Ht4.171@trnddc01...
>>>
>>>> I've been out of photography for many years so this digital
>>>> stuff is new to me so please forgive the dumb question. I'm
>>>> confused about the maximum image size one can get out of a
>>>> camera like the Nikon d700. What's the largest quality
>>>> PRINT one can make with this beast.
>>>
>>> How sharp do you want the print to be?
>>>
>>> The basic image size is 2832 x 4256 pixels, which is 9.4 x
>>> 14 inches at 300 dots/inch or 14x21 inches at 200 dots/inch.
>>> For bigger prints than that, you would probably want to use
>>> some kind of dithering/upscaling tool such as Genuine
>>> Fractals.
>>>
>>DPI is an incorrect unit designation; DPI is most commonly
>>used for scanning where the hardware scans X dots per linear
>>inch of material, where each dot turns into a pixel. In
>>digital printing or processing, everything is already in
>>pixels and is re- converted to a spot on the paper, hence the
>>proper units are PPI or Pixels Per Inch.
>
> Wow are YOU wrong!!!
>
> Pixels are reserved for MONITOR DISPLAYS!
>
> Printers are ALLWAYS rated dots per inch! My Dad agrees, and
> he's been a printer for 40 years... look up halftone mask...
>
Printers are NOT rated in DPI except for their max rated print
resolution which has no meaning for real printing because of the
use of dithering to turn CMYK into millions of colors. Again, it
is ONLY PPI that matters, period.
--
HP, aka Jerry
"Don't be a foppish blooter, make only pithy comments on Usenet"
HEMI-Powered
jour ...
> John McWilliams wrote:
You are obviously correct, David. Printers these days often
advertise resolutions in the multiple thousands of DPI but that
is far from a 1-to-1 conversion from a pixel in an image to a
"dot" on the printed page. In it's simplest form, printers use 4
inks - CMYK - to print tens of millions of colors. Photo printers
use 6 to 12 ink cartridges but still variants of CMYK thus they
must use advanced methods of dithering to turn these basic ink
shades into a full range of colors on the final print.
Again, the ONLY thing that matters really if one has a "quality"
photo printer is the image PPI since it is pixels laid down on a
page that produces the image. And, of course, it still matters
what the viewer perceives as "good" or "bad" quality
John McWilliams
Yes, indeed, I am. :-)
>
> Methinks that not all "dots" are created equal.
I'd quite agree, but this is outside the scope of my 'treatise', and I
am not a dot counter. Pixel peeper sometimes, but not dots......
How printer Mfgs. count their dots I am sure varies, and is probably
done at the smallest sized dot and with each of the 4-9 (or more tanks?)
pulses of separate colors counted as a dot, to get the maximum number
of dpi.
--
john mcwilliams
HEMI-Powered
du jour ...
>> When you talk about a dot, are you talking about a
>> constant-sized dot, or a variable-sized dot? A
>> variable-coloured dot (i.e. a mixture of CMYK) or a
>> fixed-colour dot (i.e. a single ink colour)?
>
> Yes, indeed, I am. :-)
Fine, then how does one convert from on-screen pixels to this
sort of DPI? i.e., how does one convert from PPI as all of us
understand the term to a dithered DPI?
>>
>> Methinks that not all "dots" are created equal.
>
> I'd quite agree, but this is outside the scope of my
> 'treatise', and I am not a dot counter. Pixel peeper
> sometimes, but not dots......
>
> How printer Mfgs. count their dots I am sure varies, and is
> probably done at the smallest sized dot and with each of the
> 4-9 (or more tanks?) pulses of separate colors counted as a
> dot, to get the maximum number of dpi.
>
One thing is sure, John: printer manufacturers measure "dots" in
the same way as they measure "throughput" - by any way they can
that maximizes what they can claim their printer produces. In the
final analysis, it is only what the eye can perceive that counts.
Or, if one wants to be scientific about it, what instruments can
measure. I doubt anyone could make an intelligent buying decision
for any printer based on its theoretical specs.
John McWilliams
> John McWilliams added these comments in the current discussion
> du jour ...
>
>>> When you talk about a dot, are you talking about a
>>> constant-sized dot, or a variable-sized dot? A
>>> variable-coloured dot (i.e. a mixture of CMYK) or a
>>> fixed-colour dot (i.e. a single ink colour)?
>> Yes, indeed, I am. :-)
>
> Fine, then how does one convert from on-screen pixels to this
> sort of DPI? i.e., how does one convert from PPI as all of us
> understand the term to a dithered DPI?
One doesn't! There used to be a somewhat close correlation between ppi
and dpi, but as printers have improved, and/or as marketing types
started seeing more dots, there is none, except what you set in your
printer dialog. Dunno your printer, or driver, my my Espons give dpi
numbers, and some other mfgs give Hi, Med. or Lo quality instead.
>>> Methinks that not all "dots" are created equal.
>> I'd quite agree, but this is outside the scope of my
>> 'treatise', and I am not a dot counter. Pixel peeper
>> sometimes, but not dots......
>>
>> How printer Mfgs. count their dots I am sure varies, and is
>> probably done at the smallest sized dot and with each of the
>> 4-9 (or more tanks?) pulses of separate colors counted as a
>> dot, to get the maximum number of dpi.
>>
> One thing is sure, John: printer manufacturers measure "dots" in
> the same way as they measure "throughput" - by any way they can
> that maximizes what they can claim their printer produces.
That's what I said.
> In the
> final analysis, it is only what the eye can perceive that counts.
> Or, if one wants to be scientific about it, what instruments can
> measure. I doubt anyone could make an intelligent buying decision
> for any printer based on its theoretical specs.
Agreed. Same as buying a camera or stereo based solely on specs.
--
john mcwilliams
HEMI-Powered
du jour ...
>>>> When you talk about a dot, are you talking about a
>>>> constant-sized dot, or a variable-sized dot? A
>>>> variable-coloured dot (i.e. a mixture of CMYK) or a
>>>> fixed-colour dot (i.e. a single ink colour)? Yes, indeed,
>>>> I am. :-)
>>
>> Fine, then how does one convert from on-screen pixels to this
>> sort of DPI? i.e., how does one convert from PPI as all of us
>> understand the term to a dithered DPI?
>
> One doesn't! There used to be a somewhat close correlation
> between ppi and dpi, but as printers have improved, and/or as
> marketing types started seeing more dots, there is none,
> except what you set in your printer dialog. Dunno your
> printer, or driver, my my Espons give dpi numbers, and some
> other mfgs give Hi, Med. or Lo quality instead.
For all the lurkers who may be wondering what all the controversy
is, let me explain. The only way I know of to help folks like the
OP here make a rational decision as to how big they can print
from a given size digital image, assuming the camera used can
generate a high enough quality image, is to talk about PPI since
it has a direct correlation to final quality. For those who
assert that the proper term is DPI because that's what printers
output, it would be necessary to understand how the printer
actually lays down "dots". Since modern ink jets, especially
photo printers, hardly have a 1-to-1 relationship between pixels
and dots, it IS necessary to either talk only in pixels, i.e.,
PPI, or be able to computer the conversion to dots, i.e., DPI.
Since that isn't possible, as you observe, it is useless to talk
in terms of DPI. Yes, in the dot-matrix days, there was a
correlation but no more. Also, as you observe, the printer driver
often has a setting for draft, normal, and high quality and it
also alters how it lays down dots and spreads ink according to
what type of paper you tell it you're using.
But, back to the OP's original question, what he or she really
wants to know is how many mega pixels is it likely to take to
print to a given size, or at least some guidelines for same. But,
as I have talked about and others have also, that's not enough
without considering the camera and the printer's inherent
quality, the viewing distance, the subject(s) being printed and,
gasp!, the photographer's definition of what is or is not "good"
quality. This age of instant gratification leads newbies to fall
for the "more mega pixels the better the picture" crap and also
leads them to ask for simplistic answers to very complex
questions.
>
>
>>>> Methinks that not all "dots" are created equal.
>>> I'd quite agree, but this is outside the scope of my
>>> 'treatise', and I am not a dot counter. Pixel peeper
>>> sometimes, but not dots......
>>>
>>> How printer Mfgs. count their dots I am sure varies, and is
>>> probably done at the smallest sized dot and with each of the
>>> 4-9 (or more tanks?) pulses of separate colors counted as a
>>> dot, to get the maximum number of dpi.
>>>
>> One thing is sure, John: printer manufacturers measure "dots"
>> in the same way as they measure "throughput" - by any way
>> they can that maximizes what they can claim their printer
>> produces.
>
> That's what I said.
But, you said it in a way that made me look stupid. Nothing I've
said is technically incorrect, and hopefully, all of it is
technical enough but understandable enough for the OP to figure
out the answer to their original question. As a pragmatist, I'm
not interested in satisfying the theorists here, just the normal
folks who want to take, view, and print "good" pictures. The
others can argue until the cows come home on the finer technical
points, but it won't matter much to most camera buyers/users.
>
>> In the
>> final analysis, it is only what the eye can perceive that
>> counts. Or, if one wants to be scientific about it, what
>> instruments can measure. I doubt anyone could make an
>> intelligent buying decision for any printer based on its
>> theoretical specs.
>
> Agreed. Same as buying a camera or stereo based solely on
> specs.
>
Yes, but all too many people in fact DO buy on specs or on the
opinions expressed here and possibly the biased reviews in camera
mags. IMO, when I'm out to buy something, I look to
"independent" tests 5-10% tops and personal recommendations about
the same. The reason I rate them so low is that I've never found
testing people or, in general, other owners, who take the same
kind of pictures I do. Hence, I rely on my own eyes to make the
determination by buying only from a store that will let me return
the camera if I don't like it. So, I nose around long enough to
create a short list of cameras I'm considering, handle them in
the store to see which ones I like or dislike by their size,
weight, and ergonomics, buy, and shoot test pictures. Of course,
once I went Canon with my Rebel XT a couple of years ago and
invested in glass, I'm unlikely to jump ship and go with Nikon or
other in the future.
Have a good weekend.
--
HP, aka Jerry
Don't be a fop or a blooter, make only pithy comments on Usenet
fr...@home.com
"Pixel" is a term coined a relatively short time ago for digital displays, there
is no such thing as a 'pixel of ink'.
Home style printers such as inkjet's use a more or less random spray pattern of
various size dots of ink, which is why their quality is not as good as a
commercial printer, which uses diamond or other set patterns of dots.
They use Dots... NOT pixels...
The fact that printers use 3 or 4 or 6 colors has nothing to do with 'pixels'.
You really should stop passing yourself off as a graphics 'expert' when in
reality you're just a tourist who bought a cheap Canon camera, and is so psycho,
you think you are now some kind of authority...
You arguing with me is quite a joke, since I and my family members have been
working in graphics for a combined period of over 100 years!
What are you anyway, a retired laborer?
Now please PISS OFF!
fr...@home.com
When a printer manufacturer talks about his DPI, he is referring to the
positioning possibilities of the print head. This is WAY different from a
commercial press.
So when they say 2400 DPI they mean they can move the head 1/2400 of an inch,
and then spray out a dot of 'unknown' size. That is why you will NEVER be able
to print 2400 separate dots in an inch space. Try it if you don't believe me.
You should, however, be able to place that dot exactly to 1/2400".
As for multi color dots, it doesn't really matter if they spray 300 magenta
ones, 300 cyan ones, and 300 yellow ones, in an inch. It is still 300 dots per
inch... a dot meaning a color element of the picture. If you use a magnifier on
your local newspaper, you will see a pattern of dots, each coming from one color
drum on the press. They are designed to NOT overlap the inks. So each press drum
is maybe 85 DPI and the total colors you get is considered to also be 85 DPI.
One of my best friends is a color checker, he makes sure the results are
registered and true.
There are other forms of printer, but it is always rated DPI. Note that home
inkjet's are usually rated 300 DPI "equivalent", regardless of what you set the
software to. I would guess that this is because the quality doesn't go up...
As for correlation between the original picture, the monitor, and the printer,
that is all up to the various drivers.
Lastly, pixel stands for "picture element', and even if you could use it for
printers, it just isn't done.
John McWilliams
> On Sat, 02 Aug 2008 13:38:48 GMT, "HEMI-Powered" <no...@none.en> wrote:
>
>> added these comments in the current discussion du jour ...
>>
>>> On Fri, 01 Aug 2008 13:03:08 GMT, "HEMI-Powered"
>>> <no...@none.en> wrote:
>>>
>>>> Andrew Koenig added these comments in the current discussion
>>>> du jour ...
>>>>
That's a bit much, dontcha think? I haven't seen Jerry saying he was
expert at this.
When he says only ppi matters, that's true insofar as what most of us
photographers deal with (except for setting the printer driver to x y or
z # of dpi).
--
john mcwilliams
Scott W
> On Sat, 02 Aug 2008 13:38:48 GMT, "HEMI-Powered" <n...@none.en> wrote:
> > added these comments in the current discussion du jour ...
>
> >> On Fri, 01 Aug 2008 13:03:08 GMT, "HEMI-Powered"
> >> <n...@none.en> wrote:
>
> >>>Andrew Koenig added these comments in the current discussion
> >>>du jour ...
>
time ago.
And it was not first used for digital displays.
> Home style printers such as inkjet's use a more or less random spray pattern of
> various size dots of ink, which is why their quality is not as good as a
> commercial printer, which uses diamond or other set patterns of dots.
A home printer is not all all random, at the very least it uses error
diffusion dithering, which is a pretty need way to get good high
resolution detail but also smooth colors for larger areas.
You might want to look at this
http://en.wikipedia.org/wiki/Dither
You will see that a random dither is total crap and no one would use
it.
No one really uses fixed dither patterns anymore, they just don't work
well.
There are also advances in halftones for printing, but most prints
done do halftones so this is of little interest.
Scott
Me
>
> Home style printers such as inkjet's use a more or less random spray pattern of
> various size dots of ink, which is why their quality is not as good as a
> commercial printer, which uses diamond or other set patterns of dots.
>
>
Really?
Do you have some information on this?
David J Taylor
[]
> Lastly, pixel stands for "picture element', and even if you could use
> it for printers, it just isn't done.
From your discussion, it's clear that it's pixels-per-inch which matter
rather than dots-per-inch, though.
Put in another way, it's the MTF of the printer and paper combination
which matters - how that might degrade the ultimate image quality, and the
other defects which printers introduce (non-linear transfer
characteristic, banding, smudges etc.).
Cheers,
David
HEMI-Powered
jour ...
>> Lastly, pixel stands for "picture element', and even if you
>> could use it for printers, it just isn't done.
>
> From your discussion, it's clear that it's pixels-per-inch
> which matter rather than dots-per-inch, though.
I think it's clear, David, but apparently not everyone does. As
to what you quoted, I can't understand anyone NOT believing that
all prints start with pixels, but then ...
> Put in another way, it's the MTF of the printer and paper
> combination which matters - how that might degrade the
> ultimate image quality, and the other defects which printers
> introduce (non-linear transfer characteristic, banding,
> smudges etc.).
Again, I agree. I side stepped all the stuff you talked about in
your last paragraph because I was assuming (yeah, I know what
happens when one ass/u/me s) that people wanting "good" prints
would start with a "good" camera/lens, use "good" technique, and
print using a "good" photo printer on "good" paper. Skipping any
of those vital ingrediants makes the entire discussion of "how
big can I print from a given image?" total nonsense.
R. Mark Clayton
"P§ł" <xvzex...@verizon.net> wrote in message
news:uktkk.554$Ht4.171@trnddc01...
> Dumb question;
> I've been out of photography for many years so this digital stuff is new
> to me so please forgive the dumb question. I'm confused about the maximum
> image size one can get out of a camera like the Nikon d700. What's the
> largest quality PRINT one can make with this beast.
> Pete
> --
> P§ł
>
The first question is what is the size of the image you capture?
Some are full frame (35mm), some APS (24mm with variations) and now "four
thirds", which is actually more like one third. OTOH internal magnification
may make the same amount of real world image on a smaller sensor, which in
turn cuts down on the size and weight of optics.
Then there is the number of pixels in the sensor - the more the merrier.
The compression (if any) performed by the camera
and the resolution of your printer.
For actual [process] printing (e.g. offset litho) the target on page
resolution is only about 120 lines (dots) per inch. (~10M pixels on A4)
Most inkjet and laser printers well exceed this and some can even control
the size of the dots and for instance an A4 page at 600dpi will have 100M
pixels.
Given the size of Nand flash cards and the speed of interfaces, there is
little need to compress before onward transmission and this can be optimised
at that stage if required. Images can be stored and transferred from the
cameras with either no compression or lossless compression.
So the key factor is the number of pixels in the sensor. Views vary on what
the equivalent number of pixels to 35mm film is from ~10MP to 100MP,
although it also depends on the speed of the film and quality of the optics.
So a Nikon d700 should produce and image sufficient for an A4 size (or 10x8)
print or a landscape photo across the front page of a newspaper. As a full
from camera there won't be any compromise over the size of image captured.
HEMI-Powered
du jour ...
>
Beyond the actual quality of the image, no matter how it is
produced, about the only thing that ultimately limits print size is
the number of pixels available to lay down on the paper.
Compression is only an issue if it is set too high, thus
introducting JPEG artifacts which detract from the appearance of
the image but in an entirely different way than does the aliasing
in a print caused by too few pixels to support the chosen print
size. I can't see how the frame size is at all relevant unless it
is somehow so bad that the sensors cannot produce a decent image in
the first place, but the OP would still likely be able to print
reasonably large and probably like what they get. Again, the rating
of a printer in DPI must be understood in the context of how many
ink squirts equate to one discernative "dot" of true image
information. It's highly doubtful that more than 4-6 MP are
required for an 8.5 x 11 print, a far cry of the 100 MP you appear
to be advocating. Now, whether any of this has anything to do with
what film can theoretically do is totally irrelevant since the OP
isn't talking film and said nothing about it, only asking a very
general question about digital but neglecting to include even a
mimimum of info and criteria to help those trying to answer the
question determine what's in their minds as to maximum print size.
Some say that 200 PPI is enough while others think 300 is enough.
You appear to be saying that 600 is the goal. Maybe, but doubtful
as all but the finest cameras and lenses under ideal conditions can
produce an image that good. For the 8 x 10 example you cite, 200
PPI, which most folk think at least meets the def of "good", would
require only about 3.2 MP while 300 about 7.2 MP. If anything is
amiss, i.e, the camera isn't up to snuff, lighting was poor, or the
photographer was less than excellently skilled and no amount of
additional pixels is going to do much to help.
Now, back to the OP after all this debate. I don't recall anything
about the camera make and model or whether the OP already owns it
or is in buy mode, but the most practical method of finding out
what it can do are some simple tests by taking several pictures of
subjects representative of what is the intended use for the camera,
perhaps with varying lighting conditions, then trying each of the
mega pixel sizes available for that camera on the chosen printer at
each print size starting with about 8.5 x 11. After that, it gets
pricey unless the OP's printer is a wide-carriage which are rare,
as they would have to job out the print job to get 11 x 14 or
poster size. Again, if the printer isn't already owned, one might
try to take images to a store but chances are they won't allow test
prints as printers are typically not even hooked up to anything.
But, if the printer(s) being evaluated have memory card slots as
most photo printers do, then one could take their test images in,
plug in the card, and bring a stack of whatever paper they want to
use. Isn't this a more reliable method of determining what any real
person's opinion is of "good" or "bad" rather than esoteric and
technical debates about lens resolving power, camera sensor
quality, camera firmware, post-picture processing, and the entire
gamut of printer technical specs? Frankly, I've lost track, but I'm
not sure the OP ever rang in again which may be that they posted
troll-like to engender a vicious debate or they're so snowed that
they no longer care.
Alan Browne
> Andrew Koenig wrote:
>> "P§³" <xvzex...@verizon.net> wrote in message
>> news:uktkk.554$Ht4.171@trnddc01...
>>
>>> I've been out of photography for many years so this digital stuff is
>>> new to me so please forgive the dumb question. I'm confused about the
>>> maximum image size one can get out of a camera like the Nikon d700.
>>> What's the largest quality PRINT one can make with this beast.
>>
>>
>> The basic image size is 2832 x 4256 pixels, which is 9.4 x 14 inches
>> at 300 dots/inch or 14x21 inches at 200 dots/inch. For bigger prints
>> than that, you would probably want to use some kind of
>> dithering/upscaling tool such as Genuine Fractals.
>
> sizes.
>
> Also, printers made in the last few years never print as low as 300 dpi;
> you were speaking of pixels per inch, not dots. 300 ppi is a fairly
> standard hi-rez number many print at, but the printers will be
> outputting in excess of 720 dpi in most cases.
A file set for 300 dpi should always print 300 image pixels per inch
regardless of the physical printer pitch.
--
-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
-- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
-- [SI] gallery & rulz: http://www.pbase.com/shootin
-- e-meil: Remove FreeLunch.
John McWilliams
> John McWilliams wrote:
>> Andrew Koenig wrote:
>>> "P§³" <xvzex...@verizon.net> wrote in message
>>> news:uktkk.554$Ht4.171@trnddc01...
>>>
>>>> I've been out of photography for many years so this digital stuff is
>>>> new to me so please forgive the dumb question. I'm confused about
>>>> the maximum image size one can get out of a camera like the Nikon
>>>> d700. What's the largest quality PRINT one can make with this beast.
>>>
>>> How sharp do you want the print to be?
>>>
>>> The basic image size is 2832 x 4256 pixels, which is 9.4 x 14 inches
>>> at 300 dots/inch or 14x21 inches at 200 dots/inch. For bigger prints
>>> than that, you would probably want to use some kind of
>>> dithering/upscaling tool such as Genuine Fractals.
>>
>> Depends on the subject matter; GF may well not be necessary at larger
>> sizes.
>>
>> Also, printers made in the last few years never print as low as 300
>> dpi; you were speaking of pixels per inch, not dots. 300 ppi is a
>> fairly standard hi-rez number many print at, but the printers will be
>> outputting in excess of 720 dpi in most cases.
>
>
> A file set for 300 dpi should always print 300 image pixels per inch
> regardless of the physical printer pitch.
Huh? How do you "set" a file for 300 dpi? Most inkjet printers won't go
that low. What is an "image pixel" printed on paper?
--
john mcwilliams
Scott W
wrote:
> John McWilliams wrote:
> > Andrew Koenig wrote:
> >>news:uktkk.554$Ht4.171@trnddc01...
>
> >>> I've been out of photography for many years so this digital stuff is
> >>> new to me so please forgive the dumb question. I'm confused about the
> >>> maximum image size one can get out of a camera like the Nikon d700.
> >>> What's the largest quality PRINT one can make with this beast.
>
> >> How sharp do you want the print to be?
>
> >> The basic image size is 2832 x 4256 pixels, which is 9.4 x 14 inches
> >> at 300 dots/inch or 14x21 inches at 200 dots/inch. For bigger prints
> >> than that, you would probably want to use some kind of
> >> dithering/upscaling tool such as Genuine Fractals.
>
> > Depends on the subject matter; GF may well not be necessary at larger
> > sizes.
>
> > Also, printers made in the last few years never print as low as 300 dpi;
> > you were speaking of pixels per inch, not dots. 300 ppi is a fairly
> > standard hi-rez number many print at, but the printers will be
> > outputting in excess of 720 dpi in most cases.
>
> A file set for 300 dpi should always print 300 image pixels per inch
> regardless of the physical printer pitch.
>
This depends on what you are using to do the printing. A program like
Photoshop defults to giving you 300 ppi, note Photoshop (elements as
least) uses PPI not DPI.
But if you send the photo to someplace like Costco.com they are going
to print to fill the paper.
And programs other then Photoshop will often try to fill the paper as
well, and this is an option on Photoshop as well.
Scott
Alan Browne
The printing program (such as PS) sets the print size.
I have an Epson printer with 720 ppi resolution, but I always edit to
300 and print it as 300 and for (eg) a 3000 x 2000~ I get a 10x~6.7 print.
John McWilliams
That doesn't begin address the question. You can only resample to ppi,
not dpi.
The driver converts the input expressed as ppi, and while many feel the
printer "needs" to have "native" input resolution of this or that, often
expressed as a percentage, such as half or quarter of its "native
resolution" (usually 720 or 360 ppi for many Epsons, and 300 or 600 for
Canon printers), this is pretty much a marketing ploy by a software
company that uprezzes images to "get that perfect input".
In any event, you can't set a file for 300 dpi. 300 ppi, yes. You can in
some of the higher end printers choose what dpi you want the printer
to spray at.
--
john mcwilliams
fr...@home.com
<david-...@blueyonder.neither-this-bit.nor-this-bit.co.uk> wrote:
>fr...@home.com wrote:
>[]
>> Lastly, pixel stands for "picture element', and even if you could use
>> it for printers, it just isn't done.
>
>From your discussion, it's clear that it's pixels-per-inch which matter
>rather than dots-per-inch, though.
Then you don't really understand... You can interchange 'pixels' with 'dots' if
you want... it's just that dots refer to ink spots and pixels refer to
electronic displays. "Dots" is more applicable to printers because the ink can
migrate in the paper and form dots, whereas the pixels on your display don't
'run'.
You do realize that a red object would have 1/3 the pixels of a gray object, but
would look the same in resolution? If you print 300 DPI you get 300 DPI in
either case...
>Put in another way, it's the MTF of the printer and paper combination
>which matters - how that might degrade the ultimate image quality, and the
>other defects which printers introduce (non-linear transfer
>characteristic, banding, smudges etc.).
The major difference between home style inkjet's and commercial presses is the
arrangement of the dots. Your inkjet spews ink all over the place, the ink can
overlap and puddle together, and the spots are of unknown size due to the fact
their size depends on the absorbs ion of the paper. An offset press on the other
hand has guarantied size ink dots that are in specific places, not overlapped.
An offset press with 85 DPI can equal the quality of an inkjet with twice that
'resolution'. Remember that those people use specific papers and test the run
first. They also match inks using Pantone and other technologies. I'm not a
pressman but I know one and we've talked about printing for years. I find it
quite interesting.
In the early days of home printing, I had an Amiga computer and an HP inkjet.
There were no drivers available, but I got some hacked ones from Germany, and we
had to do our own 'translation' from display pixels to printer dots. We had to
come up with our own gamma transfer curves. It was a DISASTER!! After using up
a few sets of cartridges and a ton of paper, we gave up and bought IBM style
PCs, if only for printing.
>Cheers,
>David
>
fr...@home.com
>fr...@home.com wrote:
>> On Sat, 02 Aug 2008 13:38:48 GMT, "HEMI-Powered" <no...@none.en> wrote:
>>
>>> added these comments in the current discussion du jour ...
>>>
>>>> On Fri, 01 Aug 2008 13:03:08 GMT, "HEMI-Powered"
>>>> <no...@none.en> wrote:
>>>>
>>>>> Andrew Koenig added these comments in the current discussion
>>>>> du jour ...
>>>>>
>>>>>> "P§ł" <xvzex...@verizon.net> wrote in message
He attacked the OP, saying he was wrong, when in fact he wasn't. Only an expert
can do that, not a tourist with a cheap Canon.
>When he says only ppi matters, that's true insofar as what most of us
>photographers deal with (except for setting the printer driver to x y or
>z # of dpi).
Even you just said "printer DPI".
fr...@home.com
The info is all around.
Presses make use of halftone masks to create set patterns of dots of specific
sizes, on paper of specific quality. Use a magnifier on a magazine and you can
see the pattern of dot dispersal.
Inkjet's spew ink of unknown size which migrates thru paper of unknown
absorption. In an unprofessional setting such as a home, it is impossible to
know precisely what you will get, quality is something you strive for, but it's
luck as to what you end up with. Most inkjet's today will give you a great
photo, but compare them with a good magazine, then compare DPI...
You're happy with your 2400 DPI inkjet? Compare with a glossy magazine with 135
DPI...
fr...@home.com
Yes it was, it was used to describe assembled images from space... If other
people used it before, it wasn't in common usage. For most people today, it is a
very young word, becoming common only in this century.
>> Home style printers such as inkjet's use a more or less random spray pattern of
>> various size dots of ink, which is why their quality is not as good as a
>> commercial printer, which uses diamond or other set patterns of dots.
>A home printer is not all all random, at the very least it uses error
>diffusion dithering, which is a pretty need way to get good high
>resolution detail but also smooth colors for larger areas.
When I said random I wasn't necessarily referring to the dither method, but
rather to the lack of set pattern as in halftone. Any dither method used appears
to be a random sequence, even if it isn't. Halftones can be measured with a
ruler, other types cannot.
>You might want to look at this
>http://en.wikipedia.org/wiki/Dither
Too bad they used different DPI for the examples! Very lame...
>You will see that a random dither is total crap and no one would use
>it.
>No one really uses fixed dither patterns anymore, they just don't work
>well.
>There are also advances in halftones for printing, but most prints
>done do halftones so this is of little interest.
>
>Scott
I don't get all you said...
Me
> On Sun, 03 Aug 2008 15:40:54 +1200, Me <us...@domain.invalid> wrote:
>
>> fr...@home.com wrote:
>>
>>> Home style printers such as inkjet's use a more or less random spray pattern of
>>> various size dots of ink, which is why their quality is not as good as a
>>> commercial printer, which uses diamond or other set patterns of dots.
>>>
>>>
>> Really?
>> Do you have some information on this?
>
> The info is all around.
>
> Presses make use of halftone masks to create set patterns of dots of specific
> sizes, on paper of specific quality. Use a magnifier on a magazine and you can
> see the pattern of dot dispersal.
> Inkjet's spew ink of unknown size which migrates thru paper of unknown
> absorption.
technique to produce variable droplet size - same nozzle but different
waveform and amplitude (Epson), different nozzles (HP), same nozzles,
different heater element (Canon).
> In an unprofessional setting such as a home, it is impossible to
> know precisely what you will get, quality is something you strive for, but it's
> luck as to what you end up with.
supplied by printer/paper maker (let alone custom profiles), colour
managed workflow, the prints are much more accurate than typical lab
prints - where they do "best guess" colour adjustment.
I know /precisely/ what I'll get - fantastic soft proof on screen to
print matches.
> Most inkjet's today will give you a great
> photo, but compare them with a good magazine, then compare DPI...
>
> You're happy with your 2400 DPI inkjet? Compare with a glossy magazine with 135
> DPI...
>
YMMV - try looking at output from a photo inkjet made this century.
Even at the end of last century they were starting to get quite good.
HEMI-Powered
I didn't attack anyone, but I did point out that the correct term
is PPI because all prints regardless of camera and regardless of
printer all start with pixels. It's as simple as that.
John McWilliams
Yes, indeed I did, as that's the correct use; my printer drivers have
settings where I can choose the dpi.
You're just a little too eager to argue!
--
john mcwilliams
David J Taylor
> On Sun, 03 Aug 2008 05:50:35 GMT, "David J Taylor"
> <david-...@blueyonder.neither-this-bit.nor-this-bit.co.uk> wrote:
>
>> fr...@home.com wrote:
>> []
>>> Lastly, pixel stands for "picture element', and even if you could
>>> use it for printers, it just isn't done.
>>
>> From your discussion, it's clear that it's pixels-per-inch which
>> matter rather than dots-per-inch, though.
>
> Then you don't really understand... You can interchange 'pixels' with
> 'dots' if you want... it's just that dots refer to ink spots and
> pixels refer to electronic displays. "Dots" is more applicable to
> printers because the ink can migrate in the paper and form dots,
> whereas the pixels on your display don't 'run'.
Actually pixels can "run" - you are more likely to see that on a CRT
display, though, or on an analogue video feed.
> You do realize that a red object would have 1/3 the pixels of a gray
> object, but would look the same in resolution? If you print 300 DPI
> you get 300 DPI in either case...
It depends on where the dots are printed on the paper. LCD displays can
use some greyscale information to provide a greater spation resolution -
for example ClearType in Microsoft Windows.
>> Put in another way, it's the MTF of the printer and paper combination
>> which matters - how that might degrade the ultimate image quality,
>> and the other defects which printers introduce (non-linear transfer
>> characteristic, banding, smudges etc.).
>
> The major difference between home style inkjet's and commercial
> presses is the arrangement of the dots. Your inkjet spews ink all
> over the place, the ink can overlap and puddle together, and the
> spots are of unknown size due to the fact their size depends on the
> absorbs ion of the paper. An offset press on the other hand has
> guarantied size ink dots that are in specific places, not overlapped.
> An offset press with 85 DPI can equal the quality of an inkjet with
> twice that 'resolution'. Remember that those people use specific
> papers and test the run first. They also match inks using Pantone and
> other technologies. I'm not a pressman but I know one and we've
> talked about printing for years. I find it quite interesting.
>
> In the early days of home printing, I had an Amiga computer and an HP
> inkjet. There were no drivers available, but I got some hacked ones
> from Germany, and we had to do our own 'translation' from display
> pixels to printer dots. We had to come up with our own gamma transfer
> curves. It was a DISASTER!! After using up a few sets of cartridges
> and a ton of paper, we gave up and bought IBM style PCs, if only for
> printing.
I would have to take your word for that, although I note that others
disagree. I am really comparing a printer and a display, not different
printing techniques. I accept that a printer "pixel" will be made up of
"dots", but the pixel will be approximate as the number of dots, the dot
size, the dot position, and the ink in the dot will all differ. The fact
that it produces anything which pleases the eye is remarkable! I tend to
prefer an LCD or CRT display for presentation.
Cheers,
David
Alan Browne
> On Sun, 03 Aug 2008 16:38:30 -0400, in rec.photo.digital.slr-systems Alan
> Browne <alan....@Freelunchvideotron.ca> wrote:
>
>
>> I have an Epson printer with 720 ppi resolution, but I always edit to
>> 300 and print it as 300 and for (eg) a 3000 x 2000~ I get a 10x~6.7 print.
>
> why you print at 300pii so the resampling of the driver is not an integer
> multiple of 720. Some of us choose to use Qimage to print from which will
> automatically output at the printer's native resolution.
720 is oversampling 300 by more than 2:1 so there is no discernible
effect where the print is concerned. I make no prints where the viewer
looks at it with a 10x loupe to discern interpolation defects.
Further, I neglected to add, most of my prints are made by a photo store
and their spec is 300. That is what I edit to so that they don't
re-sample the image.
John McWilliams
> m...@mine.net wrote:
>> On Sun, 03 Aug 2008 16:38:30 -0400, in rec.photo.digital.slr-systems Alan
>> Browne <alan....@Freelunchvideotron.ca> wrote:
>>
>>
>>> I have an Epson printer with 720 ppi resolution, but I always edit to
>>> 300 and print it as 300 and for (eg) a 3000 x 2000~ I get a 10x~6.7
>>> print.
>>
>> As you know the printer driver will resample to 720 ppi, I don't
>> understand
>> why you print at 300pii so the resampling of the driver is not an integer
>> multiple of 720. Some of us choose to use Qimage to print from which will
>> automatically output at the printer's native resolution.
>
> 720 is oversampling 300 by more than 2:1 so there is no discernible
> effect where the print is concerned. I make no prints where the viewer
> looks at it with a 10x loupe to discern interpolation defects.
This is the correct procedure. There's a fairly widespread conviction
that the highest resolution will occur if an Epson is fed a 720 ppi
image. It's not true today, if in fact it ever was so.
Qimage is responsible for some of this misdirection, which is not to say
their software isn't just fine.
--
john mcwilliams
Scott W
this some time ago, re-sampling a image to both 300 ppi and 600 ppi,
the 600ppi image produced a print with far more resolution.
http://www.pbase.com/konascott/image/69348129
Did the extra resolution make the print look shaper? if viewed closely
it did, but you had to look very close indeed.
When I have a print done at Costco I re-size to 300ppi, and even then
they make a pretty soft print compared to what even a cheap inkjet can
do, here is a comparison. The Costco print has less noise in it, but
is clearly softer with out looking all that closely.
If I am printing to an inkjet I am not going to down sample to
anything less then 600 ppi, why throwaway sharpness for no reason.
Scott
John McWilliams
What was the original size of the image used to produce that crop?
>
> Did the extra resolution make the print look shaper? if viewed closely
> it did, but you had to look very close indeed.
>
> When I have a print done at Costco I re-size to 300ppi, and even then
> they make a pretty soft print compared to what even a cheap inkjet can
> do, here is a comparison. The Costco print has less noise in it, but
> is clearly softer with out looking all that closely.
>
> If I am printing to an inkjet I am not going to down sample to
> anything less then 600 ppi, why throwaway sharpness for no reason.
I'm not suggesting you downsample! Particularly if you can see a
difference. I am suggesting that upsampling a, say, 300 ppi image to 720
ppi in order to send it to a print driver that is said to want same is
a waste of time, and may even be counterproductive to a sharp final
print. May. And once you've reached a certain size in Megabytes, the
computer and printdriver will become very slow.
--
john mcwilliams
Scott W
Ah, sorry I misunderstood, I think it depends in part on what up-
sampling software you have and how good the printer drivers are.
Scott
.
D-Mac
> m...@mine.net wrote:
>> On Sun, 03 Aug 2008 16:38:30 -0400, in rec.photo.digital.slr-systems Alan
>> Browne <alan....@Freelunchvideotron.ca> wrote:
>>
>>
>>> I have an Epson printer with 720 ppi resolution, but I always edit to
>>> 300 and print it as 300 and for (eg) a 3000 x 2000~ I get a 10x~6.7
>>> print.
>>
>> As you know the printer driver will resample to 720 ppi, I don't
>> understand
>> why you print at 300pii so the resampling of the driver is not an integer
>> multiple of 720. Some of us choose to use Qimage to print from which will
>> automatically output at the printer's native resolution.
>
> 720 is oversampling 300 by more than 2:1 so there is no discernible
> effect where the print is concerned. I make no prints where the viewer
> looks at it with a 10x loupe to discern interpolation defects.
>
> Further, I neglected to add, most of my prints are made by a photo store
> and their spec is 300. That is what I edit to so that they don't
> re-sample the image.
>
>
The developers of "Qimage" might disagree with you Alan. That 24 x 36
print from a 10D which I sent you a few years ago was sent to the
printer (an HP designjet 130, I think from memory) at 180 dpi.
My experience with half a dozen different wide format printers since
then (currently a designjet "Z" series proves that enlargement by
software interpolation is only half the story. How the print driver
process the information is equally important.
That lab you send your photos to... If the have a Fuji frontier system,
the actual print resolution is more like 160 DPI than 300. In the "good
old days" of LED printers etching the image on photo paper before
development, 300 dpi was actually arrived at by the LED diameter and
couldn't be changed. Many, many labs today that use advanced lasers
(like the frontier) still specify 300 dpi.
Why don't you send them a 100 dpi image? Up sample it in Photoshop to
300 dpi so they won't know. The results might surprise you.
--
visit www.D-Mac.info
to relieve the tension...
Usenet is after all Usenet!
John McWilliams
Douglas:
You make the same mistake others do re ppi and dpi! You cannot send a
100 dpi image; it'd be 100 ppi.
--
john mcwilliams
fr...@home.com
>fr...@home.com wrote:
>> On Sun, 03 Aug 2008 15:40:54 +1200, Me <us...@domain.invalid> wrote:
>>
>>> fr...@home.com wrote:
>>>
>>>> Home style printers such as inkjet's use a more or less random spray pattern of
>>>> various size dots of ink, which is why their quality is not as good as a
>>>> commercial printer, which uses diamond or other set patterns of dots.
>>>>
>>>>
>>> Really?
>>> Do you have some information on this?
>>
>> The info is all around.
>>
>> Presses make use of halftone masks to create set patterns of dots of specific
>> sizes, on paper of specific quality. Use a magnifier on a magazine and you can
>> see the pattern of dot dispersal.
>Yes - I worked in R&D in the printing industry for many years.
Thats nice... I'm happy for you.
>> Inkjet's spew ink of unknown size which migrates thru paper of unknown
>> absorption.
>Not if you use correct paper. Canon / HP / Epson use a different
>technique to produce variable droplet size - same nozzle but different
>waveform and amplitude (Epson), different nozzles (HP), same nozzles,
>different heater element (Canon).
Most people use the paper that was on sale that week...
>> In an unprofessional setting such as a home, it is impossible to
>> know precisely what you will get, quality is something you strive for, but it's
>> luck as to what you end up with.
>Rubbish. With correctly calibrated display and using ICC profiles
>supplied by printer/paper maker (let alone custom profiles), colour
>managed workflow, the prints are much more accurate than typical lab
>prints - where they do "best guess" colour adjustment.
>I know /precisely/ what I'll get - fantastic soft proof on screen to
>print matches.
You sound like a pro - you are excluded from my comments. I was referring to the
average home user.
>> Most inkjet's today will give you a great
>> photo, but compare them with a good magazine, then compare DPI...
>>
>> You're happy with your 2400 DPI inkjet? Compare with a glossy magazine with 135
>> DPI...
>>
>
>YMMV - try looking at output from a photo inkjet made this century.
>Even at the end of last century they were starting to get quite good.
Yup, quite good... matching a 135 DPI magazine for sure...
fr...@home.com
>>>
>>>That's a bit much, dontcha think? I haven't seen Jerry saying
>>>he was expert at this.
>>
>> He attacked the OP, saying he was wrong, when in fact he
>> wasn't. Only an expert can do that, not a tourist with a cheap
>> Canon.
>>
>>>When he says only ppi matters, that's true insofar as what
>>>most of us photographers deal with (except for setting the
>>>printer driver to x y or z # of dpi).
>>
>> Even you just said "printer DPI".
>>
>>
>I didn't attack anyone, but I did point out that the correct term
>is PPI because all prints regardless of camera and regardless of
>printer all start with pixels. It's as simple as that.
Talking to you is like talking to a brick...
Scott W
I have compared a scan from a 133 DPI magazine and from my cheap
inkjet, the inkjet has far more resolution then the magazine. Both
scans were done at 600 ppi.
http://www.pbase.com/konascott/image/101215825/original
If you have a better example of a 133 DPI magazine photo feel free to
post it.
Scott
D-Mac
So tell me John... How big is a pixel?
Is it bigger or smaller than a dot?
For that matter, what shape is a pixel?
Are all dots round?
Let's assume for a moment a dot is round, presuming the dots plopped out
by an inkjet printer are also round and they overlap to form a seamless
print, how does a square pixel get to fit in to this scenario?
The pixels in Photoshop are natively square How then do these become
dots put on a page?
The presumption you make is that pixels shown on a Computer monitor
(square pixels) are not interchangeable with round dots that make up a
photograph in the description of resolution of a photograph. WRONG!
It's the same stupidity that prevails when measuring the resolution of
film. X number of Lines Per Inch which carries with it the notion
horizontal resolution has no bearing on the clarity of a photo.
Inkjet printers have their resolution described as X horizontal dots by
x vertical dots with the horizontal dots always exceeding the vertical
count.
Monitors have their resolution described as Lines per inch vertically
the same as TV sets do. So in your little dream world you'd describe a
monitor's image as pixels per inch (which is technically incorrect)
before it gets transformed into dots per inch for printing a photograph
(which is a correct description).
And don't try to tell me Adobe is any bastion of traditional
descriptions either. They helped destroy Typefaces, cases and fonts for
describing characters by joining Microsoft in calling the whole family a
font. Why not butcher your traditional descriptions too?
The point is John, dots and pixels are 100% interchangeable as valid
descriptions of resolution. Neither have any measurable bulk, only
quantity which is the same quantity regardless of which word you use.
D-Mac
You an I clashed once before over your lopsided attempt to make a JPEG
image look inferior to a RAW image. As long as you keep using historic,
poorly harvested or even deliberately chosen examples to prove your
point, it's going to keep happening.
The Inkjet sample used error diffusion whilst the magazine sample used
(relatively) low grade screens for the printing.
Offset presses used to produce short (and sometimes long) run magazines
- many of which are old but still highly serviceable frequently use
screen pitches of around 130 dpi for convenience. There is less likely
hood of problems than trying to make old printers produce the quality of
HP's Indigo printing presses.
You really ought to set yourself the goal of unbiased comparisons if you
want to avoid flame wars erupting again.
Do yourself a favour, join up with snapfish and order a cheap "photo
book" They are printed on Indigo presses. Use a photo from one of these
to compare with your inkjet. You will get a substantial difference.
My magazines are printed at "photographic quality" - not on an indigo
and stand up very well to scrutiny because of the diffusion rather than
the screening used for printing the colours.
Technology does not stand still. The printing press' still being paid
off are obsolete when compared to current technology but because of
their cost, will continue to be used for many, many years in the hope of
recovering the investment. Inkjets are not bound by that criteria.
John McWilliams
Your conclusion is flat out wrong; dpi and ppi are not just the same.
--
john mcwilliams
D-Mac
>> The point is John, dots and pixels are 100% interchangeable as valid
>> descriptions of resolution. Neither have any measurable bulk, only
>> quantity which is the same quantity regardless of which word you use.
>
> Your conclusion is flat out wrong; dpi and ppi are not just the same.
>
Well John...
If you have some specification or conversion mathematics set in concrete
that I am not aware of that will prove otherwise, I'd really like to
hear it.
The simple fact is that like air, dots and pixels of any amount can be
jammed into a measurement container. At the end of discussions, there
will be just as many dots as there are pixels in a given resolution.
They are an interchangeable measurement.
Colin.D
An image at 300 PIXELS PER INCH is NOT printed at 300 DOTS PER INCH.
The printer uses multiple dots to create an single image pixel.
My Canon printer prints at 4800* 2400 DPI - dots per inch. That's
ELEVEN MILLION, FIVE HUNDRED AND TWENTY THOUSAND dots per SQUARE INCH of
print. That is more than NINE HUNDRED TWENTY-ONE MILLION dots for a
10*8 print
My image sure as hell wasn't 900 megapixels, at 300 PPI it was about
7.2 megapixels
The printer driver upsamples the image to 600 PPI, native for the Canon,
so the image being printed is now 28.8 megapixels.
So we have 28.8 megapixels at 600 PPI printed with 921.6 megadots,
that's 921/28.8, which is 32 DOTS PER PIXEL.
You wanted math, there it is.
Colin D.
Ray Fischer
>John McWilliams wrote:
>>> The point is John, dots and pixels are 100% interchangeable as valid
>>> descriptions of resolution. Neither have any measurable bulk, only
>>> quantity which is the same quantity regardless of which word you use.
>>
>> Your conclusion is flat out wrong; dpi and ppi are not just the same.
>
>Well John...
>If you have some specification or conversion mathematics set in concrete
>that I am not aware of that will prove otherwise, I'd really like to
>hear it.
A pixel can be one 2^8 different colors. A dot from an inkjet printer
can be one of 7 different colors.
They're obviously different.
>The simple fact is that like air, dots and pixels of any amount can be
>jammed into a measurement container.
That's the inane quibbling of somebody who hasn't a clue.
> At the end of discussions, there
>will be just as many dots as there are pixels in a given resolution.
Bullshit.
--
Ray Fischer
rfis...@sonic.net
Ray Fischer
>John McWilliams wrote:
>>> The point is John, dots and pixels are 100% interchangeable as valid
>>> descriptions of resolution. Neither have any measurable bulk, only
>>> quantity which is the same quantity regardless of which word you use.
>>
>> Your conclusion is flat out wrong; dpi and ppi are not just the same.
>
>Well John...
>If you have some specification or conversion mathematics set in concrete
>that I am not aware of that will prove otherwise, I'd really like to
>hear it.
A pixel can be one 2^24 different colors. A dot from an inkjet printer
can be one of 7 different colors.
They're obviously different.
>The simple fact is that like air, dots and pixels of any amount can be
>jammed into a measurement container.
That's the inane quibbling of somebody who hasn't a clue.
> At the end of discussions, there
>will be just as many dots as there are pixels in a given resolution.
Bullshit.
--
Ray Fischer
rfis...@sonic.net
John McWilliams
To you I guess they are. But not to many others.
A basic understanding of dots and pixels helps in understanding the
printing process, and how to best send an image to a printer, and how to
set the printer properly as well.
Dots and pixels are not the same, although at one point years ago they
were close to one another.
--
john mcwilliams
Scott W
Fred was the one saying that a fix screening at 135 (I am guessing he
really meant 133) dpi would beat out a inkjet printer, so it is a
fixed screen at 133 that I gave him.
Sure there are a lot better screening, but Fred does not seem to be
aware of this.
So show us a scan of one of your prints, it is pretty hard to beat an
injet for sharpness.
Scott
D-Mac
John... You haven't provided a single thing to support your claims.
Colin tried to confuse interpolation by a printer driver with pixels per
inch and the village idiot (Ray Fisher) is just in for the thrill thing.
Just one thing John, that supports your claim, that's all I asked for
and you try to create a smoke screen with this "basic understanding" crap.
Show me where a pixel is larger or smaller than a dot and you win the
discussion. Don't do it and all you're doing is being stubborn.
D-Mac
I don't have a scanner.
Ray Fischer
Apparently all you can do is be an asshole. YOu certainly can't
defend your idiocy with any facts.
>Show me where a pixel is larger or smaller than a dot and you win the
>discussion.
I already did, dumbshit, but you're an asshole who can't admit to
being wrong. Many people have shown how you're wrong. but you'd
rather be an assole.
--
Ray Fischer
rfis...@sonic.net
David J Taylor
[]
> Show me where a pixel is larger or smaller than a dot and you win the
> discussion. Don't do it and all you're doing is being stubborn.
D-Mac,
Suppose you print a 6MP file which is 3000 x 2000 pixels as an 240 x 160
mm image. Each source pixel is 80 um (micrometre) square - much larger
than the typical printer dot. E.g. my Epson R300 claims 5760 x 1440 dpi -
4.4 x 17.6 um.
David
HEMI-Powered
du jour ...
> You make the same mistake others do re ppi and dpi! You cannot
> send a 100 dpi image; it'd be 100 ppi.
>
I just can't fathom why this very same debate occurs so often. I
can see how some turn PPI into DPI but it helps to avoid the mental
gear change in units if one simply remembers that the fundamental
unit of visual information in the digital world is the pixel, but
it has no meaning whatsoever wrt a print unless and until it is
actually printed by some real device, and then it is all important
how many of them there are applied per linear inch of paper - no
matter the technology used to do the actual printing.
--
HP, aka Jerry
Don't be a fop or a blooter, make only pithy comments on Usenet
HEMI-Powered
>>term is PPI because all prints regardless of camera and
>>regardless of printer all start with pixels. It's as simple as
>>that.
>
>
> Talking to you is like talking to a brick...
ah, yes, and the insults start, so here's one for you: I don't
reason with fools, elitists, or prima donnas and your time is up
HEMI-Powered
...
> On Aug 3, 2:59 pm, f...@home.com wrote:
>> On Sun, 03 Aug 2008 15:40:54 +1200, Me <u...@domain.invalid>
>> wrote:
>> >f...@home.com wrote:
>>
>> >> Home style printers such as inkjet's use a more or less
>> >> random spray p attern of various size dots of ink, which
>> >> is why their quality is not as good as a commercial
>> >> printer, which uses diamond or other set patterns of dots.
>>
>> >Really?
>> >Do you have some information on this?
>>
>> The info is all around.
>>
>> Presses make use of halftone masks to create set patterns of
>> dots of spec ific sizes, on paper of specific quality. Use a
>> magnifier on a magazine and yo u can see the pattern of dot
>> dispersal.
>>
>> Inkjet's spew ink of unknown size which migrates thru paper
>> of unknown absorption. In an unprofessional setting such as a
>> home, it is impossible to know precisely what you will get,
>> quality is something you strive for, bu t it's luck as to
>> what you end up with. Most inkjet's today will give you a
>> grea t photo, but compare them with a good magazine, then
>> compare DPI...
>>
>> You're happy with your 2400 DPI inkjet? Compare with a
>> glossy magazine with 135 DPI...
>
> I have compared a scan from a 133 DPI magazine and from my
> cheap inkjet, the inkjet has far more resolution then the
> magazine. Both scans were done at 600 ppi.
> http://www.pbase.com/konascott/image/101215825/original
> If you have a better example of a 133 DPI magazine photo feel
> free to post it.
>
Magazines are printed using half-tone dots which a scanner pics
up as noise. Hardly a valid comparison to a digital image
produced by a camera or a scan from something that is continuous
tone like a film print. But what any scan has to do with an
inkjet printer is beyond me.
HEMI-Powered
What difference does it make what shape a pixel is or a dot on a
printed page? If one sticks to the fundamentals, a pixel is so
small that it's geometric shape is inconsequential yet the number
of pixels available to lay down on the paper has a direct bearing
on the ultimate quality of the print. That is, assuming that the
original digital image is, in fact, sharp enough in the first
place and doesn't suffer from any common defects such as
pixellation, JPEG artifacts, banding, softness, etc. And, one
must also assume that a quality photo printer is being used and
that it's printer driver works well, not to mention that decent
paper is used for maxium sharpness and detail, which usually
means ultra-glossy, whereas matte is usually chosen for portraits
where absolute sharpness is actually a detriment.
Scott W
My comparison is between a magazine photo and an inkjet, not a digital
image or a film print. The scan shows the difference in the level of
detail that is being created with both printing technologies. The
half-tone is not picked up as noise, the dots are very clear, and it
should be clear that the inkjet is showing far more detail in the
print. This is not to say that detail is the only think to worry
about in a print.
Scott
HEMI-Powered
jour ...
>>If you have some specification or conversion mathematics set
>>in concrete that I am not aware of that will prove otherwise,
>>I'd really like to hear it.
>
> A pixel can be one 2^24 different colors. A dot from an
> inkjet printer can be one of 7 different colors.
>
> They're obviously different.
'Tis true in 24-bit digital that there's only 2^8 "shades" in the
Red, Green, and Blue channels. How many ink colors there are
starts at 4, usually - Cyan, Magenta, Yellow, and Black to
attenuate the darkness. More than 4, such as the Photo Cyan and
Photo Magenta on my Canon Pixma 6600 are really not 2 new colors.
Thus, the transformation from RGB to CMYK has to convert
16,000,000 colors - or more if using 48-bit color space - to only
4 real colors thus some sort of mixing of hundreds of tiny
"dots" and/or some sort of dithering pattern must be used.
>>The simple fact is that like air, dots and pixels of any
>>amount can be jammed into a measurement container.
>
> That's the inane quibbling of somebody who hasn't a clue.
>
>> At the end of discussions, there will be just as many dots
>> as there are pixels in a given resolution.
>
> Bullshit.
>
I think that the number of actual dots, no matter how many
squirts the inkjet printer uses, does correspond directly to the
number of pixels. If it didn't, such obvious defects from too
little pixels - most notably aliasing, pixellation, and banding
wouldn't occur. Or, if the software used by the driver has an
anti-aliasing component, then careful examination of very fine
detail will reveal that it has been mushed in order to prevent
what people used to called "stair stepping" and other derrogatory
terms to describe what is really a print merely made from not
enough visual information, i.e., not enough pixels.
BTW, I've lost track of who the OP is, has anybody heard from
them lately in this esoteric argument?
HEMI-Powered
du jour ...
>>>> The point is John, dots and pixels are 100% interchangeable
>>>> as valid descriptions of resolution. Neither have any
>>>> measurable bulk, only quantity which is the same quantity
>>>> regardless of which word you use.
>>>
>>> Your conclusion is flat out wrong; dpi and ppi are not just
>>> the same.
>> If you have some specification or conversion mathematics set
>> in concrete that I am not aware of that will prove otherwise,
>> I'd really like to hear it.
>>
>> The simple fact is that like air, dots and pixels of any
>> amount can be jammed into a measurement container. At the end
>> of discussions, there will be just as many dots as there are
>> pixels in a given resolution. They are an interchangeable
>> measurement.
>
> To you I guess they are. But not to many others.
>
> A basic understanding of dots and pixels helps in
> understanding the printing process, and how to best send an
> image to a printer, and how to set the printer properly as
> well.
>
> Dots and pixels are not the same, although at one point years
> ago they were close to one another.
>
Dots and pixels aren't the same if one considers only physical
"dots" sprayed by the print heads of a modern photo printer which
can produce the illusion of many thousands of DPI, but in
reality, what is happening is that each pixel in the image being
transformed from RGB space to CMYK space by the printer driver
must take into consideration the particular technology used by a
given printer make and model to produce something visible.
And, I don't think I've seen anyone talk about the difference
between the transmitted light of a PC monitor and the reflected
light of a print and folks have only touched on how the quality
of the camera/lens, it's image processing software, the post-
processing software used, and the way the image is seen on-screen
nor the way that paper and ink type cause absorbtion to vary
widely perhaps either enhancing or ruining whatever inherent
image quality there may be.
But again, what does any of this matter since once someone
selects and buys a camera - of any make and model, P & S, EVF, or
DSLR, and then chooses and buys a printer, the rest of this is
both academic and moot. The combo of camera, user, and printer is
pretty much fixed for any given photograph and exactly how the
transformation from digital image to print becomes irrelevant. It
still seems to me that the only thing that matters in the end is
how the print looks, and the person with the most votes is the
creator, and I can't recall hearing from them.
These sort of "dumb question" posts appear every so often when
newbies buy a brand new max mega pixels camera and somebody's
printer then want to know "how big can I print?" when some simple
tests would provide a more than reasonable answer, far better
than wading their way though long-winded arguments from the
"experts", which is to explain why I am a pragmatist and not a
theorist. I leave the latter to them who write magazine reviews.
John McWilliams
> John McWilliams added these comments in the current discussion
> du jour ...
>
>> You make the same mistake others do re ppi and dpi! You cannot
>> send a 100 dpi image; it'd be 100 ppi.
>>
> I just can't fathom why this very same debate occurs so often. I
> can see how some turn PPI into DPI but it helps to avoid the mental
> gear change in units if one simply remembers that the fundamental
> unit of visual information in the digital world is the pixel, but
> it has no meaning whatsoever wrt a print unless and until it is
> actually printed by some real device, and then it is all important
> how many of them there are applied per linear inch of paper - no
> matter the technology used to do the actual printing.
The end product is what counts; all agreed.
The number of pixels per inch affects that. Agreed.
The number of dpi the printer prints at also affects that. Agreed.
PPI and DPI are not synonymous. All agreed except D=Mac, and....
[there must be some more....]
--
john mcwilliams
HEMI-Powered
...
>> Magazines are printed using half-tone dots which a scanner
>> pics up as noise. Hardly a valid comparison to a digital
>> image produced by a camera or a scan from something that is
>> continuous tone like a film print. But what any scan has to
>> do with an inkjet printer is beyond me.
>
> My comparison is between a magazine photo and an inkjet, not a
> digital image or a film print. The scan shows the difference
> in the level of detail that is being created with both
> printing technologies. The half-tone is not picked up as
> noise, the dots are very clear, and it should be clear that
> the inkjet is showing far more detail in the print. This is
> not to say that detail is the only think to worry about in a
> print.
>
I understood your basic point but what I was commenting on was
that it really doesn't make sense trying to compare resolution in
a book, magazine, or calendar photo to that of a film scan or a
digital image because we don't know anything at all about the
original photo's quality, but we do know that the half-tone dots
obscure much of the fine detail.
I don't know how you can say that the half-tone dots aren't
picked up as geometrically regular noise (vs. random noise
produced by scanning graining film prints or high ISO digitals).
The dots do exist, scanners do pick them up, and you're left
either with thousands and thousands of dots in the scanned image
or some sort of compromise if you've used descreening at scan
time and then software noise reduction when post-processing.
As to the dots being very clear, I'd agree with that if one is
looking at the printed photo with a jewelwer's loupe but not any
scan I've ever seen. But, whether the dots are or aren't easily
visible, the appearance of sharpness in a printed photo certainly
depends to a great extent on the quality of the original, and the
quality of the printing process including the paper used. It is a
rare magazine that can match the photo quality of a fine book
dedicated to quality photos if for no other reason than the dots
in the book are much smaller/finer.
HEMI-Powered
Finally, some agreement! BTW, anyone have an idea of what the OP
thinks of all this stuff?
fr...@home.com
>>>I didn't attack anyone, but I did point out that the correct
>>>term is PPI because all prints regardless of camera and
>>>regardless of printer all start with pixels. It's as simple as
>>>that.
>>
>>
>> Talking to you is like talking to a brick...
>
>ah, yes, and the insults start, so here's one for you: I don't
>reason with fools, elitists, or prima donnas and your time is up
Wow, are you ever stupid!!! Like I give a shit what an un-educated tourist with
a cheap camera thinks!!!
YOU CAN FUCK OFF NOW!!!
Bwahahaha!
PLONK
fr...@home.com
No, you misunderstood... I meant that a 2400dpi inkjet is not 20 times better
than a 133dpi mag or any other similar press work. Sort of tongue in cheek....
I was making the point that the 2400 DPI rating was 'false'. I'm not sure what
press specs would be required to beat an inkjet, but I'm sure you won't need a
2400dpi press! But what would you get from you inkjet if you set it to 133??
>Sure there are a lot better screening, but Fred does not seem to be
>aware of this.
Oh, I'm aware of lots of things, but I'm trying to stick with the 'pixels' vs
'dots' debate and the 'lies' of the printer companies. I used to use a printer
driver that allowed the choice of soft screens. Since my acceptance of the PC
world, I ignore it all now and just scale my pictures to 300 and print away...
m...@home.com
>Scott W added these comments in the current discussion du jour
>...
>
>>> Magazines are printed using half-tone dots which a scanner
>>> pics up as noise. Hardly a valid comparison to a digital
>>> image produced by a camera or a scan from something that is
>>> continuous tone like a film print. But what any scan has to
>>> do with an inkjet printer is beyond me.
>>
>> My comparison is between a magazine photo and an inkjet, not a
>> digital image or a film print. The scan shows the difference
>> in the level of detail that is being created with both
>> printing technologies. The half-tone is not picked up as
>> noise, the dots are very clear, and it should be clear that
>> the inkjet is showing far more detail in the print. This is
>> not to say that detail is the only think to worry about in a
>> print.
>>
>I understood your basic point but what I was commenting on was
>that it really doesn't make sense trying to compare resolution in
>a book, magazine, or calendar photo to that of a film scan or a
>digital image because we don't know anything at all about the
>original photo's quality,
You don't know the specs of your camera??
>but we do know that the half-tone dots
>obscure much of the fine detail.
Depends on mask size...
>I don't know how you can say that the half-tone dots aren't
>picked up as geometrically regular noise (vs. random noise
>produced by scanning graining film prints or high ISO digitals).
You don't know the meaning of the word noise... those dots in the scan are real,
not noise.
>The dots do exist, scanners do pick them up, and you're left
>either with thousands and thousands of dots in the scanned image
>or some sort of compromise if you've used descreening at scan
>time and then software noise reduction when post-processing.
>
>As to the dots being very clear, I'd agree with that if one is
>looking at the printed photo with a jewelwer's loupe but not any
>scan I've ever seen.
You've never seen the dots in a scan of a magazine? Are you Stevie Wonder?
> But, whether the dots are or aren't easily
>visible, the appearance of sharpness in a printed photo certainly
>depends to a great extent on the quality of the original, and the
>quality of the printing process including the paper used. It is a
>rare magazine that can match the photo quality of a fine book
>dedicated to quality photos if for no other reason than the dots
>in the book are much smaller/finer.
I got more news for you, Shylock - books are printed on presses as well,
sometimes halftone.
HEMI-Powered
> un-educated tourist with a cheap camera thinks!!!
>
fingers up your ass and walk on your elbows for all I care. Like I
said, I have no use for fools, elitists, or prima donnas and you
fit all 3.
HEMI-Powered
>>that it really doesn't make sense trying to compare resolution
>>in a book, magazine, or calendar photo to that of a film scan
>>or a digital image because we don't know anything at all about
>>the original photo's quality,
>
> You don't know the specs of your camera??
Of course I do but what does that have to do with the print
quality of a magazine that will be scanned, not photographed?
>>I don't know how you can say that the half-tone dots aren't
>>picked up as geometrically regular noise (vs. random noise
>>produced by scanning graining film prints or high ISO
>>digitals).
>
> You don't know the meaning of the word noise... those dots in
> the scan are real, not noise.
Talk to an engineer, noise is any unwanted signal. It doesn't
have to be random or caused by sensors to be "noise".
I'm done talking to you. I've already discussed all the points
you are disputing. Your retorts are idiotic so don't deserve a
response.
> I got more news for you, Shylock - books are printed on
> presses as well, sometimes halftone.
>
You need to read more and talk less. I already said that books
are half-toned, but they have the luxury of higher quality
printing processes and vastly superior paper.
Chris Malcolm
>> I got more news for you, Shylock - books are printed on
>> presses as well, sometimes halftone.
> You need to read more and talk less. I already said that books
> are half-toned, but they have the luxury of higher quality
> printing processes and vastly superior paper.
With the exception of course of those magazines which are printed on
higher quality printing processes and vastly superior paper, such as
the high quality photographic art magazines.
--
Chris Malcolm c...@infirmatics.ed.ac.uk DoD #205
IPAB, Informatics, JCMB, King's Buildings, Edinburgh, EH9 3JZ, UK
[http://www.dai.ed.ac.uk/homes/cam/]
savvo
>> Wow, are you ever stupid!!! Like I give a shit what an
>> un-educated tourist with a cheap camera thinks!!!
I don't think you did write that, but you've carelessly lost your
attributions.
> Idiot, I don't have a cheap camera. As for you, stick both your
> fingers up your ass and walk on your elbows for all I care.
Still obsessed with the anal insertions? It does say an awful lot about
the workings of your mind.
--
savvo orig. invib. man
HEMI-Powered
jour ...
>> You need to read more and talk less. I already said that
>> books are half-toned, but they have the luxury of higher
>> quality printing processes and vastly superior paper.
>
> With the exception of course of those magazines which are
> printed on higher quality printing processes and vastly
> superior paper, such as the high quality photographic art
> magazines.
>
There's exceptions to every rule, but the common factor is is
significantly smaller dot patterns and much less ink absorbsion
in the papers used in high-quality photo books and some
magazines. However, most mags are printed with rather coarse dot
patterns on very thin paper, so thin that often the text or
photos on the back side bleed through when trying to scan. Now, I
still don't understand what the size of half-tone dots has to do
with printer resolution based on the size of the image, which was
the OP's original question.
It continues to be true that more pixels generally means higher
quality prints mainly from reducing aliasing IF and only if the
larger image actually carries additional image information and is
itself not flawed in some material way. And again, I cite what
should be an easy to understand example: two 8 MP cameras, one a
$100 P & S and the other a entry level to mid range SLR such as a
Nikon D70 or Canon Rebel XT each in the 6-8 MP range. I doubt
anyone would dispute the latter will produce a superior image
with much less noise unless either the camera or the photographer
is broken.
C J Campbell
> Dumb question;
> I've been out of photography for many years so this digital stuff is new to
> me so please forgive the dumb question. I'm confused about the maximum image
> size one can get out of a camera like the Nikon d700. What's the largest
> quality PRINT one can make with this beast.
> Thanks
> Pete
And now, for someone who actually knows what he is talking about:
http://www.bythom.com/printsizes.htm
--
Waddling Eagle
World Famous Flight Instructor
HEMI-Powered
jour ...
>> Dumb question;
>> I've been out of photography for many years so this digital
>> stuff is new to me so please forgive the dumb question. I'm
>> confused about the maximum image size one can get out of a
>> camera like the Nikon d700. What's the largest quality PRINT
>> one can make with this beast. Thanks
>> Pete
>
> And now, for someone who actually knows what he is talking
> about:
>
> http://www.bythom.com/printsizes.htm
>
A very nice discussion, very technical, very thorough, but also
very unworkable. No real-world photographer can possibly control
all the variables discussed no matter how right the author is nor
do I at all think that over 300 PPI is necessary to create a good
print without discussion other criteria such as subject type,
lighting, viewing distance, and the photographer's own
definitions of "good" and "bad".
I see, though, that the author has again perpetuated the myth of
using DPI in place of PPI. It only becomes DPI at the time of
printing. Prior to printing, the calculation is clearly PPI since
their simply ain't no ink dots in a digital image, only millions
of pixels, each of generally one of 16.7M colors (unless 48 bit
is used).
Other than that, this is a nice read, thanks for posting the
link. I've saved it for future reference.
HEMI-Powered
>>> Wow, are you ever stupid!!! Like I give a shit what an
>>> un-educated tourist with a cheap camera thinks!!!
>
> I don't think you did write that, but you've carelessly lost
> your attributions.
>
>> Idiot, I don't have a cheap camera. As for you, stick both
>> your fingers up your ass and walk on your elbows for all I
>> care.
>
> Still obsessed with the anal insertions? It does say an awful
> lot about the workings of your mind.
>
It is important to live up to, or down to, one's reputation
especially dealing with the many mental midgets that inhabit
Usenet. Are you perhaps looking for another graphic metaphor,
perhaps like mind more of your own Fucking business and less of
mine?
savvo
> savvo added these comments in the current discussion du jour ...
>
>>>> Wow, are you ever stupid!!! Like I give a shit what an
>>>> un-educated tourist with a cheap camera thinks!!!
No, I didn't say that.
>>
>> I don't think you did write that, but you've carelessly lost
>> your attributions.
>>
>>> Idiot, I don't have a cheap camera. As for you, stick both
>>> your fingers up your ass and walk on your elbows for all I
>>> care.
And certainly not that.
>>
>> Still obsessed with the anal insertions? It does say an awful
>> lot about the workings of your mind.
>>
> It is important to live up to, or down to, one's reputation
> especially dealing with the many mental midgets that inhabit
> Usenet. Are you perhaps looking for another graphic metaphor,
> perhaps like mind more of your own Fucking business and less of
> mine?
>
So your 'business' is anal insertion then? If you don't want people
discussing it, don't keep mentioning it in public newsgroups.
m...@home.com
>>>I understood your basic point but what I was commenting on was
>>>that it really doesn't make sense trying to compare resolution
>>>in a book, magazine, or calendar photo to that of a film scan
>>>or a digital image because we don't know anything at all about
>>>the original photo's quality,
>>
>> You don't know the specs of your camera??
>
>Of course I do but what does that have to do with the print
>quality of a magazine that will be scanned, not photographed?
YOU said:
>or a digital image because we don't know anything at all about
>>>the original photo's quality,
So YOU don't know the specs of your own camera!
>>>I don't know how you can say that the half-tone dots aren't
>>>picked up as geometrically regular noise (vs. random noise
>>>produced by scanning graining film prints or high ISO
>>>digitals).
>>
>> You don't know the meaning of the word noise... those dots in
>> the scan are real, not noise.
>
>Talk to an engineer, noise is any unwanted signal. It doesn't
>have to be random or caused by sensors to be "noise".
I AM an engineer you fucking idiot!!! Those dots ARE the scan, not un-wanted
noise!!
FUCK, you are the dumbest shit on newsnet!!!
HEMI-Powered
> On Wed, 06 Aug 2008 14:19:15 GMT, "HEMI-Powered"
hey, dumb ass, I don't "scan" magazines with a camera. and,
anything in a photograph or a scan which is "unwanted" by the
user is, by definition, noise, asshole. glad you don't practice
engineering anywhere near me, miscreant Fuck ass, because it is
quite clear that you took neither electrical, mechanical, or
systems engineering or you wouldn't make such obvious falacious
statements.
and, idiot fuck face, when I speak of print quality in a book,
magazine or calendar, I speak of the entire gamut of which the
dot pattern size, angle, and spacing are just a part of it. but
then, fuck heads - can you be a fuck head if your head is firmly
stuck up your ass as yours so obviously is? - try to scan crap-
ass magazine photos on paper so thin you can see through it with
an $8,000 DSLR.
whadda maroon!
m...@home.com
Hey ASSHOLE, YOU SAID:
>
>I don't know how you can say that the half-tone dots aren't
>picked up as geometrically regular noise
So you are saying that the dots of a halftone are NOISE!!!!
So you don't know the difference between signal and noise!!
So stop lying, you dumb ass, your posts are public record!
Scott W
You are correct, but your language would make most people reading your
post believe you don’t know what you are talking about. The person
who can keep the discussion in a reasonable tone sound much more
believable, IMHO.
Scott
gn_user
<christoph...@hotmail.com> wrote:
<2008080614405811272-christophercampbell@hotmailcom>:
>On 2008-07-31 18:06:34 -0700, "P§ł" <xvzex...@verizon.net> said:
>
>> Dumb question;
>> I've been out of photography for many years so this digital stuff is new to
>> me so please forgive the dumb question. I'm confused about the maximum image
>> size one can get out of a camera like the Nikon d700. What's the largest
>> quality PRINT one can make with this beast.
A. None. The Nikon D700 cannot print.
>
>And now, for someone who actually knows what he is talking about:
>
>http://www.bythom.com/printsizes.htm
Thank you.
And the ones who told him.
www.jpeg.org
www.jpeg.org/public/jfif.pdf
gn_user
<cd371456-33c2-47c2...@p10g2000prf.googlegroups.com>:
<snip>
>You are correct, but your language would make most people reading your
>post believe you don’t know what you are talking about. The person
>who can keep the discussion in a reasonable tone sound much more
>believable, IMHO.
Well you recognized it in spite of the language, didn't you? While I
don't disagree in principle, I think the worst obscenity in a forum
like this is a glib presentation of error.
What's the point of these groups anyway? If you learn something here
and wind up 200 miles from home with the wrong battery, would it help
to recall you were informed in a reasonable and believable fashion?
Not talking about mistakes or misunderstandings, that's not what I
mean at all. I'm referring to counterfeiters, frauds--people that make
it up as they go along without actually knowing anything of the
subject. Why would someone do that?
Hemi is one of these posters yet he continues pounding out screed that
bloats and devaluates these threads. In spite of the fact he has in
fact spent yrs studying these things including one-on-one time with
one of the software developers of Paint Shop Pro, he still doesn't
know fuckall about any of it. Why? Who cares? Do you?
Ask yourself a question, if he's actually learned something about the
issues discussed here, where is it? Why isn't it being repeated
for the benefit of others? Why is he continually in conflict with the
simplest facts?
If you give it a little thought it may come to you. ;-
reg'ds
--
"Ignorance is preferable to error, and he is less remote from the
truth who believes nothing than he who believes what is wrong."
-Thomas Jefferson (Notes on Virginia, 1782)