Best resolution to use for 8x10 prints?

[Lunaray]

Hello,

Is there a general resolution/bit depth guideline to use for highest quality
8x10 prints?

I thought I would be real smart and test my new scanner (Nikon LS-8000) and
choose the absolute maximum size & bit depth for my first roll of 6x7 film
and save them as "tiff" files; well needless to say, they turned out pretty
big (356 megs ea. , which got me to thinking, I'm never going to print
anything larger than 8.5" x 11" (unless I replace my current printer) so
what should I be scanning them at to ensure the best possible quality,
without being totally absurd (356 meg files)? Any suggestions?

Thanks!

Ray

 

[David J. Littleboy]

Hello,

Is there a general resolution/bit depth guideline to use for highest quality
8x10 prints?

I thought I would be real smart and test my new scanner (Nikon LS-8000) and
choose the absolute maximum size & bit depth for my first roll of 6x7 film
and save them as "tiff" files; well needless to say, they turned out pretty
big (356 megs ea. , which got me to thinking, I'm never going to print
anything larger than 8.5" x 11" (unless I replace my current printer) so
what should I be scanning them at to ensure the best possible quality,
without being totally absurd (356 meg files)? Any suggestions?

Caution: heresy follows.

Scan at the max resolution in 16 bits. In your favorite 16-bit photo editor,
adjust the black and white points and get the contrast and brightness
roughly right, and save as an 8-bit best quality jpeg. (I actually use 95%
quality, 4:4:4 subsampling (i.e., no subsampling) in Picture Window Pro.)
This is your archive file, and will be around 25MB for 6x7. In this day of
DVD and 120GB disks, not unreasonable.

Then take that jpeg, crop, sharpen lightly, downsample to a sensible dpi,
and sharpen lightly again.

If you downsample to 2400 dpi, you won't lose much detail, but the file will
still be ridiculous overkill for 8.5x11.

I print with Qimage, which automagically resamples optimally for the
printer.

David J. Littleboy
Tokyo, Japan

 

[Douglas MacDonald]

Apart from the chuckle at what you did.
200 DPI get you pretty good inkjet prints
300 DPI gets you photographs from the Lab.
Apply compression to the files as JPGs and you'll get a relatively small
file size which will return superb prints. At a pinch you can 'upsize' your
200 dpi files to 300 dpi and not (usually) see any difference in the photos
you get back.

Douglas

 

[Lunaray]

Thanks David, but what I don't understand is why you would scan at the
highest resolution (dpi/ppi) and maximum bit depth (16 bits per channel) and
then reduce everything before saving it? How does this differ from choosing
the resolution and bit depth before scanning? I've also heard that for
printing, it's best to use an uncompressed file type i.e. bmp, or tif; any
thoughts on this?

 

[Lunaray]

Hey, I thought I was doing the right thing! ;-)

Thanks for the info, I'll try it!

Ray

 

[David J. Littleboy]

Thanks David, but what I don't understand is why you would scan at the
highest resolution (dpi/ppi) and maximum bit depth (16 bits per channel) and
then reduce everything before saving it?

The idea is to get an archive file that's as good as possible.

How does this differ from choosing
the resolution and bit depth before scanning?

Probably not a lot*. But the idea is to get as much information as possible
from the scanner, do early tonal and color manipulations in 16-bit so that
the 8-bit form is as accurate as possible.

*: I've found that scanning at 4000 dpi and then sharpening lightly and
downsampling to 2000 dpi makes images that look better than scanning at 2000
dpi. This is in my tests with my 8000. YMMV, but I think this is right.

I've also heard that for
printing, it's best to use an uncompressed file type i.e. bmp, or tif; any
thoughts on this?

That's where the heresy lies. My opinion is that scanned files are actually
pretty poor: noisy and low resolution relative to the number of samples. So,
in my opinion, using an uncompressed format for scans is insane overkill.

People who think that every bit that comes out of the scanner is god-given
perfection object to using lossy compression.

David J. Littleboy
Tokyo, Japan

 

[Preston Earle]

"Lunaray" asked: "what I don't understand is why you would scan at the
highest resolution (dpi/ppi) and maximum bit depth (16 bits per channel)
and then reduce everything before saving it?" and "David J. Littleboy"
answered: "The idea is to get an archive file that's as good as
possible." and "But the idea is to get as much information as possible
from the scanner, do early tonal and color manipulations in 16-bit so
that the 8-bit form is as accurate as possible." and "My opinion is that
scanned files are actually pretty poor: noisy and low resolution
relative to the number of samples. So, in my opinion, using an
uncompressed format for scans is insane overkill. People who think that
every bit that comes out of the scanner is god-given perfection object
to using lossy compression."
--------------------------------

I generally agree with what DJL is saying, except I am on the other side
of the 8-bit vs. 16-bit argument. (See:
http://www.ledet.com/margulis/ACT_postings/ColorCorrection/ACT-16-bit-2002.htm
[or http://tinyurl.com/br19]. Only under very specific circumstances
will it make any difference in the appearance in the final scans whether
you save 8-bit or 16-bit files from the scanner, and the benefit seems
to be in using Photoshop to convert to 8-bit from 16-bit rather than the
scanner software. Color correcting 16-bit files vs. 8-bit files once in
Photoshop doesn't seem to make any difference. There have been long
discussions on this subject before, and I suspect everyone's mind is
made up--one way or the other.) In this mostly-free world, you are
welcome to do either.

I do take exception to the concept of archiving a huge file on general
principles. I can think of no reasons to scan larger than
300ppi-print-output-size. For 8x10 prints from 6x7 negatives, that's
what, about 1000-scanner-ppi?

I think the most likely significant improvement in quality in the future
is going to come from improved scanner technology, so we are going to be
rescanning all the important images we want to be "the best". There is
no reason IMHO to worry too much about archive file quality today when
the quality of future scanners will make the archives obsolete anyway.
I'd save 2400x3000 pixel files as high-quality jpegs at about 2MB. YMMV.

Preston Earle
(e-mail address removed)