Re: 16 bits = 15 bits in Photoshop?
Re: 16 bits = 15 bits in Photoshop?
- Subject: Re: 16 bits = 15 bits in Photoshop?
- From: email@hidden
- Date: Mon, 18 Apr 2005 13:46:57 EDT
Roger writes,
>>It seems to me from reading all the arguments presented by everyone, that it
is moot to try to establish 8 bit or 16 bit as absolutes. Bruce works in 16
bits whereas Dan works in 8 bits. Jim also like working in 8 bits. Mark like
16 bits. Jon is in favor of 16 bits and so is Ray. I think the *better*
point would be not to adamantly stick to one or the other but open up to a
"pluralistic" approach, some people like 8 bits for all their reasons while
some other prefer for all their reasons.>>
Just to clarify--only one side is speaking in absolutes here. One side is
claiming superiority, often using very strident terms. Nobody is claiming that
8-bit is superior, and nobody is refusing to work in 16-bit. All I am saying is
I am reluctant to double file sizes when there is no indication that doing so
will be of any use. If and when there is such an indication, I'm happy to
change my workflow.
>>But what I'm slowly gathering is that
bit depths should not be discussed as absolutes but as "situation" or
application-specific. I tend to agree with Robert, above, and his sky
example, because I met my share of gradation problems that have not always
ended happily. With restrospect and based on the recent arguments presented
by many, it seems to me when the subject matter is characterized by rapid
transitionning of tones, lots of alternating dark and medium and light
shades, low bit depth is sufficient to accurately represent the scene, but
when the transitionning is more subtle, as in Robert's example of a sky,
then having more bits to discretely map all that smooth transitionning is
more important.>>
That's correct. The considerations and the results are the same as in
determining how much resolution an image needs. Robert's example needs to be
conservative in that regard. The first example, even when output at the same size, can
get by with considerably less resolution.
The difference is that significantly excessive or significantly inadequate
resolution can actually make a difference on output, enough so that almost
everyone will pick out the correctly resolved one as better, especially after
editing. I've published such side by side examples.
Extra resolution causes extra smoothness, which is sometimes good, sometimes
bad. The extra smoothness is also visible in a 16-bit file after it's been
heavily edited--but you have to look quite carefully, ordinarily at a high
magnification. The edits would have to be unthinkably huge to make a visible
difference. However, if somehow that difference could be brought out, then yes,
Robert's image would probably correct better in 16-bit and the other one would be
judged better if it were corrected in 8-bit.
However, the first order of business has to be to demonstrate that an image
like Robert's *can* look better under certain circumstances when corrected in
16-bit, not just as a theory, but as a fact. To date, nobody has established
that.
Dan Margulis
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden