Re: 16bit vs 8 bit
Re: 16bit vs 8 bit
- Subject: Re: 16bit vs 8 bit
- From: Ric Cohn <email@hidden>
- Date: Thu, 07 Feb 2002 16:34:26 -0500
I9m normally lurking in the background, However, since I have personally
done a good deal of experimenting with 8 bit vs 16 bit images I9m voicing my
opinions here. First, I am a great admirer of both Dan Margulis and Bruce
Fraser as well as many of the other contributors on both lists, but if I
wanted unquestioning belief of anyone or anything I9d join the appropriate
religion (g). I am a commercial photographer mostly doing photography for
advertising. Most of the scanning and retouching I do is on images for my
portfolio, hence my point of view is different than either a fine art
photographer or a production person. I want to avoid both banding and the
blocking of highlights and shadows in printed images, and I want to leave
some 3headroom2 for later adjustments if I decide to vary greatly from the
original artwork. I believe there are some benefits to using 16 bit files if
the image needs large tonal and color adjustments. I generally make my
initial corrections in 16 bits and convert to 8 bits when the image is
reasonably close to what I want. I think this is much like the way good
scans used to be done where you would have the scan operator redo scans
(using the scanners internal 16 bits) until the scan looked close to
correct. However, I believe that many of the reported advantages to editing
16 bit files are really from either poor working methods or looking at
things like the histogram rather than output.
Bruce Lindstom and Lee Varis both make some excellent points. In my own
tests, I have found that the closer I look at real images the more the
differences I see are introduced by either my hardware, by Photoshop, or by
my (poor) work methods. For example, using my scanner and software (a
Microtek ArtixScan 1100 with Scanwizard Pro) I can produce scans using
several different settings. I have found that I can produce 2 different
scans which look and print identically if not adjusted further in Photoshop.
However, when you look at the individual channels there are big differences.
One of these scans will fall apart if edited in 8 bits, but if scanned in 16
bits this same file will hold together with the same edits. However, by
using scanner settings which give me a cleaner scan I can see much subtler
differences between identically corrected 8 bit and 16 bit scans. In fact,
the corrected 3good2 8 bit scan can look better in many ways than a
corrected 3bad2 16 bit scan. As had been pointed out before, what9s
important is the quality of the bits not the quantity.
In my own work I frequently apply masks and I agree 100% with Lee Varis that
these 8 bit B&W channels are a much more frequent degrader of images than 8
bit corrections. How can we get Adobe to bring 12 or more bits to these
channels? Chris are you listening?
Ric Cohn
email@hidden
http://www.riccohn.com
212.924.4450
_______________________________________________
colorsync-users mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/colorsync-users
Do not post admin requests to the list. They will be ignored.