Re: 16-bit vs 8-bit images
Re: 16-bit vs 8-bit images
- Subject: Re: 16-bit vs 8-bit images
- From: Lee Varis <email@hidden>
- Date: Tue, 12 Feb 2002 14:37:03 -0800
- Organization: Varis PhotoMedia
Re:
>
Now, I scan RAW in highbit mode and archive that. I also duplicate it and
>
globally tone the data in highbit mode to where I'd want the image to come
>
from a scanners driver, IOW, fully toned globally. Then I dupe that and
>
save. The dupe I convert to 8-bit mode and work using that full-range tool
>
set.
This is very sensible. At the very least you have to archive the highbit
(I much prefer this term over 16-bit) RAW file - basically this is what
I do with my digital camera captures. I just don't bother to save the
highbit toned file as I generally do quite a bit of "interpretive" color
editing in 8 bits after I've rendered the general color & tone in
highbit.
>
<The real bitch is that you don't know if you need high-bit until it's too
>
late...>
This is why you want that RAW file - if you change your mind about a
particular 8 bit rendering you'll want to re-render using the highbit
data. However, I generally find that you do tend to know if you're going
to have a problem with a particular image at the highbit toning stage
and at that point its your skill at color editing that's going to make
the most difference between getting a good image or a mediocre one. The
wealth of tools available in 8 bits might suggest a walk down the 8 bit
path despite what the "histogram" might look like.
>
I'm of the mind that a highly skilled editor will be able to produce
>
vastly better quality from 8-bit data than a less skilled editor will from
>
16-bit data.
One strategy I use is to prepare two different tonal renderings from the
highbit data - one that favors shadow detail and another that favors
highlight detail. These can then be converted to 8 bits, layered and
blended using luminance masks or advanced blending modes that allow you
to make more interactive judgments. In the end I usually end up with a
lot of little editing tricks that would not be possible in 16 bits with
Photoshop's current tool set.
>
It's not a religious issue for me the way it is
>
with some people, and ther are certainly no absolutes involved.
I think it really depends on the images you work with and its a case by
case basis for how much highbit editing you're going to want to do. For
me, once I have reasonably toned captures I stay in 8 bits until I have
my interpreted color where I want it & then I'm done - I don't bother
with highbit all the way. If it "looks" good to me I'm happy whether the
histogram has "combs" in it or not. If your workflow includes a lot of
work with color negatives, as Bruce pointed out earlier, it may
certainly be to your advantage to stay in 16 bits as long as you can - I
don't know. (thanks for the heads up on that Bruce!)
I just don't think that there's something invisible in that 16 bit file
that can be magically preserved by staying in 16 bit until the bitter
end and that magical 16 bit essence will somehow make for a brilliantly
improved image at some point in the distant future when we have dichroic
crystal holography that suspends a 3D virtual image in the air in front
of your face. By then the quaint, antique Epson 10000 prints that my
great grandchildren might (hopefully) still have would probably be worth
more despite being rendered from 8-bit files.
my .03
Regards,
Lee Varis
email@hidden
http://www.varis.com
888-964-0024
_______________________________________________
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.