Dan Margulis Applied Color Theory - Field Test of 8-bit vs. 16-bit

Date: Mon, 13 May 2002 20:50:56 EDT
From: DMargulis@aol.com
Subject: 8- and 16-bit correction

Folks,

Shortly before leaving for Europe, I completed several days of work comparing the results of correcting 16-bit files versus performing the same edits on 8-bit files. I'll have a full discussion of what this showed in my book later this year and will include the files on the CD so that others can verify what went on. However, since there was one finding that came as a surprise and would cause me to change workflow in some cases, I thought I would let people know here first.

As many of you know, there has been this ongoing debate as to whether there's merit in leaving files in the cumbersome 16-bit mode as long as possible on the theory that corrections will be more accurate. There are those who have strenuously argued that this is the only way to go, that there is a night-and-day difference in the results, and that people who do things any other way are rubes. None of these experts has ever shown anything other than a histogram to demonstrate that this is so, and my own tests have shown up to now no substantial difference.

I therefore issued a challenge for those who could come up with any color photographic image where any reasonable course of events might conceivably show an advantage for working in 16-bit as opposed to just converting immediately to 8-bit, and said if I could verify it I would admit it and print the results.

Naturally, none of the people who are so vehemently in favor of 16-bit correction had any such files to show, but some list members did, and sent them to me. Particularly, Todd Flashner allowed me the use of his scanner to produce several high-bit images that were disastrously flat and would presumably magnify any advantage that 16-bit might have. Ric Cohn sent over 2 gigs of images and an ungodly number of Epson proofs of images that he felt did demonstrate that working in 8-bit caused banding and other problems.

The bottom line of all my tests was, with one important caveat that I'll get to in a moment, there is no 16-bit advantage. I blasted these files with a series of corrections far beyond anything real-world; I worked at gammas ranging from 1.0 to 2.5 and in all four of the standard RGBs, I worked with negs, positives, LAB, CMYK, RGB, Hue/Saturation, what have you. While the results weren't identical there were scarcely any cases where there would be detectable differences and in those one would be as likely to prefer the 8-bit version as the 16. So, I have no reservation in saying that there's no particular point in retaining files in 16-bit, although it doesn't hurt either.

I'll show all these results later, but the surprise was in the files that Ric sent, which appeared to show just the sort of damage that 8-bit editing is supposed to cause, in an image with a dark rich blue gradient, a worst-case scenario in conjunction with the very dark original scan, which in itself was an attempt to give an advantage to 16-bit editing.

Ric provided both original 8-bit and 16-bit versions of these files. Granted that the necessary corrections were very severe, they still showed that what he said was true: the 8-bit version banded rather badly and the 16-bit did not. I tried several different ways of trying to get around the disadvantage and could not do so without excessive effort.

Ric's 8-bit original, however, was generated from the 16-bit scan not by Photoshop but rather within his own scanner software. Therefore, I tried further tests where I applied the same extreme corrections to the image, but this time not to Ric's 8-bit image but rather a direct Photoshop conversion of Ric's 16-bit image to 8-bit. Shockingly, this completely eliminated the problem. There was no reason to prefer the version corrected entirely in 16-bit.

When Photoshop converts from 16-bit to 8-bit it applies very fine noise to try to control subsequent problems. Most scanners don't. I would have expected this to make a difference but not to the point that the scanner 8-bit file would completely suck and the Photoshop 8-bit file would be just as good as the 16-bit version. I don't know whether this is all a function of Photoshop's superior algorithm or whether the scanner is doing something bad. Furthermore, I don't care. One way or another, the 8-bit scanner file is bad and the 8-bit Photoshop file is good.

I also don't know whether other scanners would have the same problem that Ric's appears to have, but suspect that they might.

The whole thing suggests to me that if one's scanner is capable of generating a 16-bit file, one should probably take it. Thereafter, whether one converts it to 8-bit in Photoshop early or late, it doesn't seem to make a difference.

Dan Margulis


Date: Tue, 14 May 2002 01:20:34 -0700
From: Mike Russell
Subject: Re: 8- and 16-bit correction

[re Dan's 16 and 8 bit based color correction results]

Very interesting indeed, and I look forward to reading about Dan's results
in more detail, as well as seeing the all-important images.

> The whole thing suggests to me that if one's scanner is capable of generating
> a 16-bit file, one should probably take it. Thereafter, whether one converts
> it to 8-bit in Photoshop early or late, it doesn't seem to make a difference.

Aside from scanners, certain >8 bit cameras may have a similar problem
generating sloppy 8 bit images that compare poorly with Photoshop's.

Thanks, Dan for bringing the hi-falutin hi-bit argument down to earth, and reality. Let's hope it either stays there, and that we start "phrasing" our digital imaging arguments more in terms of images instead of mere words.

I have a similar16 bit challenge, modeled after Dan's posted on Usenet, where I'm offering a $100 reward for any scan that benefits from being manipulated in >8 bits per channel. It would seem, now, that if anyone wins my challenge, it will be due to a lack of technique on my part and not the underlying technology.

http://geigy.2y.net/DigPhoto/16bit/challenge.htm


Mike Russell


Date: Mon, 13 May 2002 21:28:50 -0600
From: Andrew Rodney
Subject: Re: 8- and 16-bit correction

on 5/13/02 6:50 PM, Dan Margulis wrote:
>
> As many of you know, there has been this ongoing debate as to whether there's
> merit in leaving files in the cumbersome 16-bit mode as long as possible on
> the theory that corrections will be more accurate.

Who said that? You1re putting words in at least a few people1s mouths.

16 bit isn1t any more accurate. 16 bit files provide a level of comfort for
users who may have no idea what edits or output a file may go through in the
future. It1s about keeping your options open.

> I therefore issued a challenge for those who could come up with any color
> photographic image where any reasonable course of events might conceivably
> show an advantage for working in 16-bit as opposed to just converting
> immediately to 8-bit, and said if I could verify it I would admit it and
> print the results.

Define 3reasonable course of events.2

> Naturally, none of the people who are so vehemently in favor of 16-bit
> correction had any such files to show, but some list members did, and sent
> them to me. Particularly, Todd Flashner allowed me the use of his scanner to
> produce several high-bit images that were disastrously flat and would
> presumably magnify any advantage that 16-bit might have.

I didn1t see this file but there are plenty of reasons where scanning flat in high bit makes a lot of sense! You may have no idea where you want to set clipping values because the data may need many different alterations of the tonal scale. Or you may have a scanner operator that needs to scan as fast as possible so this tonal adjustment can be done over a network to multiple users, setting the highlight and shadow clipping using Photoshop. It greatly speeds up scanning when you can offload these decisions to other users. It adds flexibility when you have to target one scan to multiple devices or uses.

> The bottom line of all my tests was, with one important caveat that I'll get
> to in a moment, there is no 16-bit advantage. I blasted these files with a
> series of corrections far beyond anything real-world; I worked at gammas
> ranging from 1.0 to 2.5 and in all four of the standard RGBs, I worked with
> negs, positives, LAB, CMYK, RGB, Hue/Saturation, what have you. While the
> results weren't identical there were scarcely any cases where there would be
> detectable differences and in those one would be as likely to prefer the
> 8-bit version as the 16.

Assuming you were done with the editing for the rest of the file1s life. It1s possible you need to composite this file into another in the future. It1s possible that file needs to have several rounds of this kind of editing multiple times for multiple uses. In such a case, having high bit would keep a users options open.

> I'll show all these results later, but the surprise was in the files that Ric
> sent, which appeared to show just the sort of damage that 8-bit editing is
> supposed to cause, in an image with a dark rich blue gradient, a worst-case
> scenario in conjunction with the very dark original scan, which in itself was
> an attempt to give an advantage to 16-bit editing.

You output these files to what devices?????

> Ric provided both original 8-bit and 16-bit versions of these files. Granted
> that the necessary corrections were very severe, they still showed that what
> he said was true: the 8-bit version banded rather badly and the 16-bit did
> not. I tried several different ways of trying to get around the disadvantage
> and could not do so without excessive effort.

That says a mouth full!

> Ric's 8-bit original, however, was generated from the 16-bit scan not by
> Photoshop but rather within his own scanner software.

> The whole thing suggests to me that if one's scanner is capable of generating
> a 16-bit file, one should probably take it. Thereafter, whether one converts
> it to 8-bit in Photoshop early or late, it doesn't seem to make a difference.

So what? Most of today1s scanners produce high bit data. Photoshop doesn1t produce data (unless you start with a blank canvas).

Andrew Rodney


Date: Tue, 14 May 2002 14:13:02 -0000
From: "marshyswamp71"
Subject: Re: 8- and 16-bit correction

> Who said that? You1re putting words in at least a few people1s mouths.

Remember the LAB rounding error debate? - loss, and perhaps accurate can mean different things to different people.

"Ian Lyons -

I'm sure, like me, you have read the various recommendations that Photoshop users should endeavour to edit their images in high-bit mode. The argument has been that by doing so we can minimise the data loss that is inherent in any image editing operation."

http://www.rgbnet.co.uk/ilyons/16bit/16bit_1.htm

Ian does not say either way how he feels or how he sees the 'facts' [clever move], this is taken from a tutorial on high bit workflows for users who do use this mode but suffer from the lack of flexibility. However, this does sum up one of the common arguments which is true - high bits do indeed help prevent or minimize data loss during mode transforms or other edits.

If there is less rounding error in a file, would it not be considered more 'accurate' than one that did have rounding errors? But I think most of us know or suspect that Dan means _visible_ accuracy - as in one image looking superior (or is that one that is inferiror) to account for the time spent in high bits.

A huge factor in this is the intended workflow and how that user feels about the data/image - a photographer making an archive often has a different attitutde to a prepress user in conditions where the image is a 'throw away' one off publication.

"Jeff Schewe -

The ability to do multiple edits without worry of level loss and banding...I suppose the difference could be characterised as professional vs recreational use of Photoshop."

http://www.schewephoto.com/workshop/pdfs/16bit_Editing.pdf

http://www.schewephoto.com/workshop/index.html

Jeff has a PDF which is secured, so I can't copy the text so there is only a short truncated quote.

"Bruce Fraser -

The practical implication of this is simple: If you make your major global edits for tone and color on high-bit files, you stand much less chance of running into visible posterization and banding than you will if you apply the same edits to a file that started out with 8-bit channels."

http://www.creativepro.com/story/feature/7627.html

Finally, the magic word...visible - which is what concerns my particular workflow.

My tests with flatbed scans and some limited CG renders did not make me switch to doing everything in high bit, but if the device or render offers it and I have the luxury in time/memory/space or workflow, then I will use high bits - editing is a different story.

High bit alpha channels in a regular bit file have been wished for before on this list, but I don't know how you can implement mixed bit depth in the same file...but any move which would help with the integrity of alpha masks would be very welcome by me...smart noise is a good band aid, but no banding or contouring in gradated masks as an integral part of the Photoshop engine would be nice

Stephen Marsh.


Date: Tue, 14 May 2002 10:48:29 -0400
From: Ric Cohn
Subject: 8- and 16-bit correction

>>Ric provided both original 8-bit and 16-bit versions of these files. Granted that the necessary corrections were very severe, they still showed that what he said was true: the 8-bit version banded rather badly and the 16-bit did not. I tried several different ways of trying to get around the disadvantage and could not do so without excessive effort.

Ric's 8-bit original, however, was generated from the 16-bit scan not by Photoshop but rather within his own scanner software. Therefore, I tried further tests where I applied the same extreme corrections to the image, but this time not to Ric's 8-bit image but rather a direct Photoshop conversion of Ric's 16-bit image to 8-bit. Shockingly, this completely eliminated the problem. There was no reason to prefer the version corrected entirely in 16-bit.

When Photoshop converts from 16-bit to 8-bit it applies very fine noise to try to control subsequent problems. Most scanners don't. I would have expected this to make a difference but not to the point that the scanner 8-bit file would completely suck and the Photoshop 8-bit file would be just as good as the 16-bit version. I don't know whether this is all a function of Photoshop's superior algorithm or whether the scanner is doing something bad. Furthermore, I don't care. One way or another, the 8-bit scanner file is bad and the 8-bit Photoshop file is good.

I also don't know whether other scanners would have the same problem that Ric's appears to have, but suspect that they might.>>

It's satisfying that Dan has come to (mostly) the same conclusions I came to and I thank him for taking so much time with the images I sent him. His knowledge is truly amazing.

Initially I thought I saw a clear and obvious 16-bit advantage. However, on further experimentation I too found that this difference was mostly due to a difference in the quality of the scanner's data. I have actually figured out how to produce a better 8-bit or 16-bit scan from my scanner. I don't know if this is common or not, but the scanner is capable of giving very clean data when set one way and gives garbage when set another way and this garbage data is a much bigger problem when generated at 8-bits. Significantly, the difference is only obvious when looking at individual channels before correction, but becomes very obvious after applying curves, etc. Unfortunately, the "garbage" way includes the path for profiling the scanner.

My final take is that I now scan at 16 bits in the way I have found gives the best data from my scanner even though this gives a very flat scan. However, I do believe I leave the most "headroom" for later corrections by adjusting the raw 16-bit scan up to the level of what I would normally consider a decently corrected scan before going down to 8-bits. I can't see any difference from even extreme corrections after this point.

By the way, I am using a Microtek ArtixScan 1100 and scan mostly 4x5 and 8x10 transparencies and negatives.

Ric Cohn

ric@riccohn.com
http://www.riccohn.com
212.924.4450


Date: Tue, 14 May 2002 09:21:53 -0600
From: Andrew Rodney
Subject: Re: Re: 8- and 16-bit correction

on 5/14/02 8:13 AM, Stephen Marsh wrote:

> I'm sure, like me, you have read the various recommendations that
> Photoshop users should endeavour to edit their images in high-bit
> mode. The argument has been that by doing so we can minimise the data
> loss that is inherent in any image editing operation."

That1s true. The more you edit an 8 bit file globally, the more loss of data. Not so with high bit files. If you routinely work on files where there is a single output need or you don1t have the need to archive the data for some unforeseen use, 8 bit should be just fine.

> If there is less rounding error in a file, would it not be considered > more 'accurate' than one that did have rounding errors?

The rounding error is due to conversions of files. I don1t consider edits on an 8 bit file to lose data due to 3rounding errors2 but that1s semantics and not worth getting into.

> But I think > most of us know or suspect that Dan means _visible_ accuracy - as in
> one image looking superior (or is that one that is inferiror) to
> account for the time spent in high bits.

Dan has yet to clarify how he1s deciding to determine whether there was data loss that was noticeable or not. Ink on paper is a lot more forgiving than output at Rez 80 to an 8x10 piece of film or a 40 inch Lightjet print.

> A huge factor in this is the intended workflow and how that user
> feels about the data/image - a photographer making an archive often
> has a different attitutde to a prepress user in conditions where the
> image is a 'throw away' one off publication.

I totally agree. High bit workflows are useful for people that need maximum flexibility. It1s probably not the best solution for those that need to work in production environments where the final output is ink on paper.

> "Jeff Schewe -
>
> The ability to do multiple edits without worry of level loss and
> banding...I suppose the difference could be characterised as
> professional vs recreational use of Photoshop."

Yes, Jeff likes to call users that don1t work in high bit 3recreational users2 which is fine from his background and workflow I guess. I1m a bit more lenient. If you work in production and you have a single output need, I1m fine with an 8 bit workflow from start to finish. I think the workflow is key! If you have a zillon images to output and time is money, and 98% of people viewing the output couldn1t see the data loss, who cares. If you need to scan very large files for multitudes of output, you need to do a lot of compositing, you need to keep your options open and storage isn1t an issue, I think high bit workflows are the way to go. Unlike some people, rather than put down one technique verses the other, I think we should examine our own workflows and use the most appropriate way to handle the files. If you like to hedge your bets about how a file will survive editing, high bit is safest.

> Finally, the magic word...visible - which is what concerns my
> particular workflow.

Yes and where do we decide this visible posterization appears?

> High bit alpha channels in a regular bit file have been wished for
> before on this list, but I don't know how you can implement mixed bit
> depth in the same file...but any move which would help with the
> integrity of alpha masks would be very welcome by me...smart noise is
> a good band aid, but no banding or contouring in gradated masks as an
> integral part of the Photoshop engine would be nice

The use of Alpha channels in high bit, at least for selection work is rather simple! Duplicate the high bit file and dump it to 8 bit. Do all the work you want with the alpha channel. You can shift drag any selection from the 8 bit file to the high bit file in perfect registration as long as the two files remain the identical pixel dimensions.

Dan and Jeff are obviously at different ends of the spectrum here. At last one hopes to 3Prove2 that high bit editing is a waste of time. I don1t know why. If your workflow is such that the time and energy (and lack of some PS features) make it a huge hassle, don1t do it. If you have no idea how much editing a file might need now or anytime in the future, high bit is simply insurance. We should be looking at the workflow and user, plus the eventual use of a file rather than spend time trying to prove that one method over the other is worthless. It1s pretty obvious that input manufactures and those that make image editing software provide high bit for a reason. If you don1t use those capabilities, fine. It doesn1t mean that others can1t take advantage of a high bit workflow.

Andrew Rodney


Date: Tue, 14 May 2002 09:47:03 -0600
From: Ross Fenmore
Subject: Re: 8- and 16-bit correction

>"The ability to do multiple edits without worry of level loss and
>banding...I suppose the difference could be characterised as
>professional vs recreational use of Photoshop."

That kind of statement is what causes a lot of the debate. Anybody can understand that you get less loss of number values, and there may be issues with images that are re-edited many times, but many of us work with 8-bit images professionally without sacrificing visual quality. Dan is one of the few professionals writing today who doesn't seem to have a problem with making that side of the argument. AFAIK, nobody has ever published a book and CD that tackles the issue with a demonstration of actual images.

Ross Date: Tue, 14 May 2002 12:15:32 -0400


From: Scott Olswold
Subject: RE: Re: 8- and 16-bit correction

Stephen,

I follow in your footsteps. My scanner is a 40-bit. So I scan at 40 bit and let Photoshop sort it out. More often than not, I end up converting directly into 8-bit and editing there. Maybe its because I'm more comfortable in 8-bit...I don't know (I don't know because I'm comfortable in RGB mode, but I like having access to ALL of my tools and workflow considerations, not just a subset).

I will capture as much information as possible, but how I work, that's another story.

Scott Olswold


Date: Wed, 15 May 2002 01:59:14 +1000
From: "Stephen Marsh"
Subject: Re: Re: 8- and 16-bit correction

> High bit alpha channels in a regular bit file have been wished for
> before on this list, but I don't know how you can implement mixed bit
> depth in the same file...but any move which would help with the
> integrity of alpha masks would be very welcome by me...smart noise is
> a good band aid, but no banding or contouring in gradated masks as an
> integral part of the Photoshop engine would be nice

>>The use of Alpha channels in high bit, at least for selection work is rather simple! Duplicate the high bit file and dump it to 8 bit. Do all the work you want with the alpha channel. You can shift drag any selection from the 8 bit file to the high bit file in perfect registration as long as the two files remain the identical pixel dimensions.<<

Sorry Andrew, I thought I was clear here...an eight bit mask loaded into a high bit file still has the same basic problems as the regular bit original. Creating soft grads and feathers which are problem free can be a problem - there is an old archive from this list at Dans site which I have read more than once which goes into all this. I understand the workarounds using saved selections from tools using 8 bit dupes and their adjustment layers by saving the curve presets or whatever, my issue is the masking and sigle/near single ink painting artifact issues that can happen with feathers/blurs/soft brushes etc.

You can make a high bit mask feathered by feathering a hard selection (loaded off a reg bit file) in high bit and then pull a linear curve in the mask to make it white...but I have not tested if this high bit created mask is better than one in 8 bit alpha. Feathers and grads can be a problem, even with noise/smart noise.

>> Dan and Jeff are obviously at different ends of the spectrum here. At last one hopes to 3Prove2 that high bit editing is a waste of time. I don1t know why. If your workflow is such that the time and energy (and lack of some PS features) make it a huge hassle, don1t do it. If you have no idea how much editing a file might need now or anytime in the future, high bit is simply insurance. We should be looking at the workflow and user, plus the eventual use of a file rather than spend time trying to prove that one method over the other is worthless. It1s pretty obvious that input manufactures and those that make image editing software provide high bit for a reason. If you don1t use those capabilities, fine. It doesn1t mean that others can1t take advantage of a high bit workflow.<<

Agreed. I think that it is helpful to know what is going on under the hood, so that when/if a particular workflow is used people know what is going on. The time spent 'proving' the benefits of either method is helpful to formulate both specific and general workflows - even users who may not have found a need for high bits may see a need in a certain case, to me this is better than spending most of your time in a workflow that does not have any benefit - just because you think it might just have a benefit. Quantify, qualify etc. Knowledge is power and more power to the user.

I do not like the lack of features available to high bits in Photoshop, plus there is the speed issue too. I would hate the see the RAM requirements for some common layered files if they could exist in high bit. In the future this and more may change and my current practical concerns will not be an issue, until then I am waiting to implement a specific niche high bit workflow when the need arises - but have yet to find it. Even CG and natural images in grayscale mode did not offer the expected benefits of the higher bits.

Stephen Marsh.


Date: Wed, 15 May 2002 02:09:38 +1000
From: "Stephen Marsh"
Subject: Re: Re: 8- and 16-bit correction

>> I am waiting to implement a specific niche high bit workflow when the need arises - but have yet to find it.<<

Correction, I forgot to mention that my tests did indicate that mode transforms to LAB and other profiles/modes did perform better when the 8 bit original file was moved into 16 bit - even though Photoshop is meant to use internal high bit data for mode transforms on 8 bit data, moving the file into high bits does affect the transfrom for the 'better'. Although not visible, the integrity of the data was 'superior' in the high bit transform from RGB to LAB.

Stephen Marsh.


From: "Dave King"
Subject: Re: [colortheory] 8- and 16-bit correction
Date: Tue, 14 May 2002 17:09:49 -0400

> When Photoshop converts from 16-bit to 8-bit it applies very fine noise to
> try to control subsequent problems. Most scanners don't. I would have
> expected this to make a difference but not to the point that the scanner
> 8-bit file would completely suck and the Photoshop 8-bit file would be just
> as good as the 16-bit version. I don't know whether this is all a function
> of Photoshop's superior algorithm or whether the scanner is doing something
> bad. Furthermore, I don't care. One way or another, the 8-bit scanner file
> is bad and the 8-bit Photoshop file is good.

> I also don't know whether other scanners would have the same problem that
> Ric's appears to have, but suspect that they might.

> The whole thing suggests to me that if one's scanner is capable of generating
> a 16-bit file, one should probably take it. Thereafter, whether one converts
> it to 8-bit in Photoshop early or late, it doesn't seem to make a difference.

I thought I would be able to use an 8-bit workflow with a new Nikon LS-8000,
but 8-bit output has posterized and blocked shadow gradients. Here too the
solution is downsampling from 16 bit output in PS.

Dave


Date: Wed, 15 May 2002 14:34:05 -0700
From: Dennis Dunbar
Subject: Re: 8- and 16-bit correction

This debate is indeed an interesting one. It reminds me of the quote from Arthur Clarke, "For every expert there is an equal and opposite expert".

In my own experience I will say that there have been times when correcting in 16 bit has led to a smoother result. One image comes to mind where I had to deepen the orange yellow of an egg yolk for an HBO ad. Whatever I tried in PS led to banding where it dropped off to shadows, but when I went into Live Picture (where I am told the color corrections are in 16bit) it was easy. A lot of times the designs for the movie posters I work on call for really whacking the color around. And the clients are really hyper about avoiding any banding, justifiably so.

Most importantly there are times when I need to be able to paint in 16 bit. Unfortunately PS just doesn't give us this ability. But this is where I feel 16 bit editing really shines: painting and creating gradations. These examples are easy to find and recreate, but I see little discussion of those points in these debates. (Perhaps it's because PS doesn't yet let us do this so we have to find our "solutions" elsewhere.)

Dennis Dunbar


Date: Wed, 15 May 2002 16:09:29 -0700
From: Lee Varis
Subject: Re: Re: 8- and 16-bit correction

Hi all,

Dennis Dunbar wrote:

> In my own experience I will say that there have been times when
> correcting in 16 bit has led to a smoother result. One image comes to
> mind where I had to deepen the orange yellow of an egg yolk for an HBO
> ad. Whatever I tried in PS led to banding where it dropped off to
> shadows, but when I went into Live Picture (where I am told the color
> corrections are in 16bit) it was easy....

This is very similar to my own experiences. I think, however, that the banding you observed corrected in Live Picture was due to the fact the Live Picture uses a 12 bit mask for gradations and painting effects - in fact all compositing (including color correction "layers") in Live Picture is done with 12 bit "mask" objects which are part of its unique "Fits" file structure.

> Most importantly there are times when I need to be able to paint in 16
> bit. Unfortunately PS just doesn't give us this ability. But this is
> where I feel 16 bit editing really shines: painting and creating
> gradations.

Again, here it is not 16 bit editing that matters but the fact that, even in 16 bit mode, Photoshop's selections are 8 bit alpha channels. Cross fades from one color to another are also handled by 8 bit (256 step) alphas which are routinely "noised" by checking the "dither" checkbox in the gradient tools options. I've been asking for high-bit alpha channels in Photoshop for a while now as I feel that this is much more important than 16 bit editing tools, though, I suspect, full support for 16 bit editing would imply support for 16 bit alphas. The bottom line is that this is not really a color correction issue and you can observe the same "banding" in gradations in a 16 bit RGB image regardless of the type of corrections applied if the "banding" is due to a badly stepped mask (and always, in my experience, from being artificially created through Painting or filters).

regards,

Lee Varis
varis@varis.com
http://www.varis.com
888-964-0024


Date: Wed, 15 May 2002 22:37:56 -0500
From: Jeff Schewe
Subject: Re: 8- and 16-bit correction

On 5/15/2002 3:01 PM, Andrew Rodney wrote:

> Dan and Jeff are obviously at different ends of the spectrum here. At last
> one hopes to 3Prove2 that high bit editing is a waste of time. I don1t know
> why.

Andrew. . .

I suspect that is correct [BG].

I watched, with amusement, when Dan challenged anybody to come up with a 16 bit image that could prove the benefits of editing in 16 bit over 8 bit. I now understand there's a bounty out there of $100 if somebody can prove that edits in 16 bit vs edits in 8 bit is superior. . .

That's a waste of time (and believe me, $100 is no incentive to me).

Fact is there is no image that can prove that one to several edits in 16 bit vs 8 bits is better. I've never advocated editing in 16 bit merely as a method of improving a few tone/color corrections. That's silly. The edits done in the beginning of a tone/color correction does not cause banding. It does indeed "spend bits" and I advocate spending your bits wisely. Where the _REAL_ difference between 16 & 8 bit editing comes is well down the road. Well after your original edits.

Photoshop is pure math. . .everything is numbers. Everything done in Photoshop is the result of an algorithm. And believe me, Photoshop's precision is not infinite. Where the benefits of original 16 bit editing shows is down the road after doing corrections and applying a filter or two (or dozen). Add a dozen layers with various opacities and blending modes and maybe a layer effect or 2. Do a gradated adjustment layer. Do any series of extended retouching or manipulations that many image artists do every day. Then you'll find out that at some point in the process, you've rounded enough data and spent enough bits that the gradation between point "A" and "B" no longer has enough bits to produce a smooth gradation of tone or color. Guess what, you've got banding. And, guess what else, you've just proved that _STARTING_ with your initial tone/color edits in 16 bits and conserving your data bits to spend later in the editing process allows you to avoid the dreaded banding.

If you scan a chrome and do a slight to moderate curve correction and transform from RGB > CMYK, should you scan in 16 bits? Nope. If you do slight to moderate to even substantial tone/color edits and then plan on assembling 1/2 dozen composited images with filters, layer blending and effects and maybe 20 hours in image manipulations, will you benefit a _LOT_ from doing your original edits in 16 bits? You bet your ass you will.

I started scanning in 16 bits because I ran into constant problems of ending up spending hours working on an image only to find that at some point, a tone gradation in an image banded. Once banded, there's nothing you can do to recover the lost spent data.

My point is that if you need maximum flexibility and edit ability and you don't want to worry about banding, start your creative retouching in a "perfect 8 bits/channel". That only comes after doing initial tone/color corrections in 16 bit.

Sure, you can use the scanner software to do tone/color corrections in the scanner's high bit depth and thus benefit from 10, 12, 14 or 16 bits of data. The big problem I have with that workflow is this. . .there ain't a scanner on earth who's software allows you to do "local" corrections of tone/color. Scanners don't come with the ability to put a lasso around an area to adjust just that area. The other weak point of scanner (or camera software for that matter) is that a scanner preview just purely isn't as accurate as Photoshop to show exactly what the pixels look like. Scanners give "previews" but you don't see the full resolution till _AFTER_ you've applied the tone/color corrections during the scan. At that point if you touch it again you won't have a full 8 bit file. . .you're spending the bits on a wasted edit. I also think, and perhaps I'm biased, that Photoshop is the best "pixel viewing" application on the planet. I know of no other app that is as accurate in showing on screen, exactly how a set of pixels will look when you output. . .either photo or halftone reproduction. So, if I want the most accurate on screen representation of just exactly how those pixels will look reproduced, I'm just not interested in using a scanner software or camera capture software's interpretation of those pixels.

Nope. . .Dan and the rest of you are welcome to continue scanning in 8 bits and doing whatever you want to do to your images. . .but if you want absolute total control over tone and color without the risk of breaking the image somewhere down the road. . .you better learn to edit in 16 bits.

Yes, the tools are more limited and yes the files are 2x the size. . .so? Ram is cheap and so are hard drives these days. I've learned to edit in 16 bit to the point where even the Photoshop engineers couldn't believe how far one can go if you're determined. You can paint (using history), you can copy/paste (using clone between 2 16 bit documents), you can use adjustment layers (in an 8 bit duplicate and save out as a setting), you can use color range in 8 bit and transfer the 8 bit selection into 16 bits for application. You can clone and heal and run enough filters to do just about anything you need to do to start off your imaging in the "perfect 8 bits".

And yes, I'll stand by the line "recreational" if you squander and waste your data bits just getting an image tone/color corrected in 8 bit. . .cause if you do that, you're working with considerably less than 8 bits/channel and deserve the banding you are likely to incur.

Dan and others who claim 16 bit editing isn't better than 8 bit editing for routine scanning and corrections are correct. But, I'm a photographer and image manipulator. I choose to spend my data bits wisely because because what I do to images isn't routine. I manipulate the heck out of images. I choose to do all initial tone/color correction in 16 bit to avoid banding down the road. And, personally I feel no compulsion to prove anything to anybody except my clients that pay me a lot of money.

Regards,
Jeff Schewe
16 bit advocate and proud of it. . .


Date: Wed, 15 May 2002 21:36:17 -0700
From: Kevin Connery
Subject: Re: Re: 8- and 16-bit correction

I agree in theory with Jeff Schewe's statements about multiple, long-term manipulation potentially causing problems.

I disagree with the term "recreational", as there are many (many, many) professional users who simply do not, and never will, perform as many manipulations on their images as it takes to cause a visible rounding. I've blended a few dozen images from multiple sources, in different colorspaces, with local tweaking on each one, and never ran into that problem. That's more tweaking than all the photographers I've met (excluding Jeff, at least).

Prepress folks who are doing conversions from various colorspaces to go to various printers/presses, by the descriptions given, don't need to use 16-bit; are they "recreational" users? Portrait photographers doing image retouching prior to sending images to the lab for their customers are "recreational" users? Commercial photographers doing image cleanup prior to sending images to a client for approval are "recreational users"?

It may be a quibble, but it's a derogatory label on an entire class of users that doesn't benefit from being used, AND paints a great many full-time professionals with the same label. At best, it reduces communication. At worst...

--kdc
(FWIW, I scan in 16-bit, do my primary tone adjustments there, then
convert to 8-bit. Only the 8-bit files are archived.)


Date: Thu, 16 May 2002 09:26:11 EDT
From: Dan Margulis
Subject: Re: 8- and 16-bit

Jeff Schewe writes,

>>Nope. . .Dan and the rest of you are welcome to continue scanning in 8 bits and doing whatever you want to do to your images. . .but if you want absolute total control over tone and color without the risk of breaking the image somewhere down the road. . .you better learn to edit in 16 bits.>>

Rather than continuing to post the same defensive bluster to every group credulous enough to listen, it must be a better use of your time to produce even one image that demonstrates the point. After all, this is supposed to be critical, night and day, the difference between professional and recreational imaging. If an image exists that shows such a dramatic difference, why not show it, rather than just make claims?

You are of course welcome to advocate editing while drinking water out of the wrong side of the glass, or only after offering prayers to the goddess of the volcano Popocatapetl, or only after anointing the mouse with magic oil. In any of these cases, you are the one advocating the inconvenient workflow, so it's up to you to make the case for it.

Dan Margulis


Date: Thu, 16 May 2002 08:25:21 -0600
From: Ron Kelly
Subject: Re: Re: 8- and 16-bit correction

Jeff Schewe wrote:
>I advocate spending your bits wisely. And, guess what else, you've just
> proved that _STARTING_ with your initial tone/color edits in 16 bits and
> conserving your data bits to spend later in the editing process allows you
> to avoid the dreaded banding.

Seems to me that spending your bits wisely is somehow at odds with a workflow that carries a whole lot of extra freight on every image every time "just in case it's needed". If that's the best argument for 16 bit editing then we might also do everything at the maximum resolution that our scanners and systems can handle "just in case it's needed" down the road. I agree with the idea of spending wisely. I have a new computer system with lots of RAM and disk space but I don't intend to carry unneccessary freight. Those who would advocate such a policy would have figured that the buffalo could never be hunted to near extinction.

Ron Kelly


Date: Thu, 16 May 2002 08:05:42 -0600
From: "Les De Moss"
Subject: Re: Re: 8- and 16-bit

>Dan Writes: "...volcano Popocatapetl."

For those of us (or me, at least) lacking Dan's substantial world knowledge and eloquent use of vocabulary: A dormant volcano in Central Mexico. Thanks Dan, for the constant and unexpected enrichment!

-Les


Date: Thu, 16 May 2002 11:36:58 -0500
From: N9VJG
Subject: Re: Re: 8- and 16-bit correction

At 8:25 AM -0600 5/16/02, Ron Kelly wrote:

>Seems to me that spending your bits wisely is somehow at odds with a
>workflow that carries a whole lot of extra freight on every image every
>time "just in case it's needed". If that's the best argument for 16 bit
>editing then we might also do everything at the maximum resolution that
>our scanners and systems can handle "just in case it's needed" down the
>road.

Quite true. Well, some folk obviously have more time on their hands than others--Especially when their arrogance repels all those clients who like to "pay a lot of money" for their service.

I'm learning that this business is not about beating your chest "my way is the best!" and who can make the most money. It's about strong alliances, selfless sharing with clients/colleagues and consistent practice of the Golden Rule. Money and experience is the by-product. However, this recent barrage of verbal diarrhea from certain folk does nothing but annoy the reader and obfuscate the value of the message.

Mom always said: "Eric, you can't attract bees without honey."

-- Eric Curtis Bond
Photo Grafix
http://www.abetterreality.net
(847) 673-7043


Date: Thu, 16 May 2002 10:25:35 -0700
From: Jan Steinman
Subject: Re: 8- and 16-bit correction

From: Jeff Schewe
>Photoshop's precision is not infinite. Where the benefits of original 16 bit editing
>shows is down the road after doing corrections and applying a filter or two
>(or dozen)...

I think that's a valid theoretical argument, but one thing Photoshop does to ameliorate the problem is stochastic dithering.

The hypothesis is: 1) bits of low significance do not make visible differences, 2) editing amplifies bits of low significance, causing their differences to become visible, therefore 2) when performing operations that "spread bits," fill them in with a pseudo-random pattern.

I don't profess to know in detail anything about what Photoshop is doing with noise, except that Timo seems to think it is a travesty, so there must be some good about it... :-) But I could imagine stochastic dithering being used in such a way that simulates infinite precision math, particularly if the noise is "smart," and takes into account surrounding values.

Are such dithered values "real?" No, none of this stuff is really real, anyway. "Golden Eyes" arguments get nowhere with me -- I'm interested in results. But based on my signal processing background, I can imagine a way that Adobe COULD be allowing multiple 8-bit edits with nearly the VISUAL fidelity of doing the same thing in 16 bits.

This is not meant to refute either Jeff's nor Dan's real-world experience. It IS meant to refute some of the purists-theorists, who make the obvious arguments regarding granularity of data after multiple operations in 16 vs 8 bits. There is more to information theory than that!

-- : Jan Steinman
: Bytesmiths
: 19280 Rydman Court, West Linn, OR 97068, 503.635.3229


Date: Thu, 16 May 2002 11:57:22 -0600
From: Andrew Rodney
Subject: Re: Re: 8- and 16-bit correction

on 5/16/02 11:25 AM, Jan Steinman wrote:

> I don't profess to know in detail anything about what Photoshop is doing with
> noise, except that Timo seems to think it is a travesty, so there must be some
> good about it... :-)

ROTFL (I really was). I guess we both have the same impression of the guy.

> This is not meant to refute either Jeff's nor Dan's real-world experience. It
> IS meant to refute some of the purists-theorists, who make the obvious
> arguments regarding granularity of data after multiple operations in 16 vs 8
> bits. There is more to information theory than that!

Workflow is key here. For the vast majority of the users on this list, 8 bit is going to be the way to go since pushing a lot of files through the workflow is bad enough when they are half the size of a 16 bit file. But when ultimate quality and flexibility is more important than workflow (or you1re lucky enough to charge by the hour [g]), high bit has a certain comfort level and flexibility. I1d certainly like to see more options available to Photoshop users (as I1m sure Jeff would) like high bit adjustment layers or the ability to use the fade command. Lets hope the next version of Photoshop (running perhaps a G5 chip) can do this for those that want to work this way.

Andrew Rodney


Date: Thu, 16 May 2002 13:58:58 -0500
From: Chris Brown
Subject: Re: Re: 8- and 16-bit

Dan,

> Rather than continuing to post the same defensive bluster to every group
> credulous enough to listen,

&

> You are of course welcome to advocate editing while drinking water out of the
> wrong side of the glass, or only after offering prayers to the goddess of the
> volcano Popocatapetl, or only after anointing the mouse with magic oil.

This is why I enjoy this group. There is a highly tuned BS filter which
helps us apply knowledge in lieu of mere pontification of theory.

Chris Brown


Date: Thu, 16 May 2002 15:03:53 -0500
From: Jeff Schewe
Subject: Re: 8- and 16-bit

On 5/16/02 11:18 AM, Dan Margulis wrote:

> Rather than continuing to post the same defensive bluster to every group
> credulous enough to listen, it must be a better use of your time to produce
> even one image that demonstrates the point. After all, this is supposed to be
> critical, night and day, the difference between professional and recreational
> imaging. If an image exists that shows such a dramatic difference, why not
> show it, rather than just make claims?

Pretty much all of my work the last 5 years was scanned in 16 bit for initial tone & color correction. You are welcome to look at my work and see for yourself, no banding. . .even after hours of editing and extreme manipulations. I'll let my work speak for itself.

> You are of course welcome to advocate editing while drinking water out of the
> wrong side of the glass, or only after offering prayers to the goddess of the
> volcano Popocatapetl, or only after anointing the mouse with magic oil. In
> any of these cases, you are the one advocating the inconvenient workflow, so
> it's up to you to make the case for it.

Well, that's a useful statement. . .
I pretty much though my post was making my case.

Regards,
Jeff Schewe


Date: Thu, 16 May 2002 15:12:23 -0500
From: Jeff Schewe
Subject: Re: 8- and 16-bit correction

On 5/16/02 11:18 AM, Ron Kelly wrote:

> Seems to me that spending your bits wisely is somehow at odds with a
> workflow that carries a whole lot of extra freight on every image every
> time "just in case it's needed". If that's the best argument for 16 bit
> editing then we might also do everything at the maximum resolution that
> our scanners and systems can handle "just in case it's needed" down the
> road.

Ron,

Funny you should say that. . .because as a routine, I scan at the highest optical resolution of my scanner. With 35mm that's 5760ppi, 120mm it's 3200ppi and for 4x5 it's 1600ppi. I scan at the highest resolution, in 16 bit so I can be assured of maximum quality in my work. I work at high resolution so the the imaging and assemblies I do are at the resolution of the film so that the retouching and manipulation is invisible.

> Those who would advocate such a policy would have figured that
> the buffalo could never be hunted to near extinction.

Amusing, but not accurate with regards to the buffalo. The buffalo was hunted to near extinction by a relatively small group of greedy men who saw the opportunity to make a fortune while wiping out the plains Indian's chief source of food and thus making it easier to steal land. You should visit the Cody Museum some time.

Regards,
Jeff Schewe


Date: Thu, 16 May 2002 17:13:15 -0600
From: Ross Fenmore
Subject: Re: Re: 8- and 16-bit

Jeff Schewe writes,

>Pretty much all of my work the last 5 years was scanned in 16 bit for
>initial tone & color correction. You are welcome to look at my work and see
>for yourself, no banding. . .even after hours of editing and extreme
>manipulations. I'll let my work speak for itself.

Thousands of great 8-bit images have been edited over the last decade or so. Maybe we should let the work speak for itself.

>I'll stand by the line "recreational" if you squander and waste
>your data bits just getting an image tone/color corrected in 8 bit. . .cause
>if you do that, you're working with considerably less than 8 bits/channel
>and deserve the banding you are likely to incur.

I've been "recreating" since the late eighties. When exactly should I expect my comeuppance?

>...I scan at the highest optical resolution of my scanner...I scan at the
>highest resolution, in 16 bit so I can be assured of maximum quality in
>my work.

If one scans at the highest resolution needed, they can be reasonably assured that they've captured everything useful. I'm guessing that if 50,000 ppi or 128-bits per channel scans were possible, you would recommend that we not waste those valuable data bits?

I'm sure you do great work, Jeff, but your judgments regarding how other professionals work is uncalled for.

Ross Date: Thu, 16 May 2002 18:02:08 -0700


From: Deborah Davis
Subject: Re: Re: 8- and 16-bit

>I'm sure you do great work, Jeff, but your judgements regarding how
>other professionals work is uncalled for.

Au Contraire! I spent a week in Jeff's class at the Santa Fe Workshops simply because I value his judgments so highly. Jeff's judgment is EXACTLY what's called for by a large number of working professional photographers doing high end jobs. He knows this stuff incredibly well and is a talented, generous, and entertaining teacher to boot.

As others have stated, Jeff's approach isn't for everyone on this list but for those of us doing the kinds of work he is, his experience and generous advice is most appreciated.

Respectfully,
Deborah Davis
Los Angeles


Date: Thu, 16 May 2002 21:55:10 -0500
From: David Riecks
Subject: Re: Re: 8- and 16-bit

At 06:02 PM 05/16/2002 -0700, Deborah Davis wrote:
>
>As others have stated, Jeff's approach isn't for everyone on this
>list but for those of us doing the kinds of work he is, his
>experience and generous advice is most appreciated.

Deborah:

I think this does point out a distinction that I've not seen anyone voice here.

At the risk of pulling a "Gilligan" I think Jeff has a point... and I think Dan does as well.

Jeff is using many different forms of output devices, among them film recorders. Now I'm not exactly certain, but my understanding is that the requirements of a film recorder are much more rigorous than a CMYK RIP. With that in mind, if one of the possible destinations for an image file is a film recorder, I'd be inclined to keep the image in 16 bit mode as long as possible in the imaging workflow.

Now if what Dan is saying is that there are few images that need full 16 bit per channel editing when the destination is a 4 color offset press, then I'm inclined to believe that's the case. Dan is a well known expert when it comes to prepping images for this form of output and I trust his judgement.

But that doesn't necessarily mean that his conclusion can be extended to every and all form of digital output that exists.

So in that case the driving factor is what you intend to do with the file. And that's not something that I believe has been addressed in this forum.

David

David Riecks
http://www.riecks.com, Chicago Midwest ASMP member
http://zillionbucks.com "The Webhost for your Creative Business"


Date: Fri, 17 May 2002 12:55:33 -0000
From: "marshyswamp71"
Subject: Re: 8- and 16-bit

--- In colortheory@y..., David Riecks wrote:

> But that doesn't necessarily mean that his conclusion can be extended to
> every and all form of digital output that exists.
>
> So in that case the driving factor is what you intend to do with the file.
> And that's not something that I believe has been addressed in this forum.

Very good point David, although it may not have been mentioned this time, in the past I have made this very point - and I am sure others have before me.

This is why I think understanding the process in both theory and applied theory is the key, since there are many different users and workflows out there.

Regards,

Stephen Marsh.


Date: Fri, 17 May 2002 11:38:33 -0700
From: Dennis Dunbar
Subject: Re: Re: 8- and 16-bit

I've been enjoying this debate a great deal, and I see value in the points that Jeff makes as well as the points that Dan makes.

David Riecks also makes a point here that we should not forget, Jeff and Dan both come from different areas. Their work involves different issues and so what one learns as a cardinal rule does not necessarily apply to the other's experience.

I've heard many platitudes over the years. Some of these turn out to be true, and some turn out to be just sales talk. What is important is that we look at these issues and think them through ourselves to see what works for us.

If I tell you that a full sized movie poster can be printed just fine from a 40 mb file you may think I'm nuts, or prone to sloppy work. If I tell you it's gotta be 300mb you may think I'm really detail oriented, or getting paid by the minute. The truth is both are true, but for different cases. Experience tells us what we need and when. Listening to debates about how to split infinitives helps build our knowledge base, but we need to know for ourselves when and how to apply what we learn.

Thanks to both Jeff and Dan and to all the others for sharing their experiences here, you're adding to the general knowledge pool and we all benefit from that.

Dennis Dunbar


ate: Fri, 17 May 2002 13:14:11 -0600
From: Ross Fenmore
Subject: Re: Re: 8- and 16-bit

Okay, I didn't realize he had a fan club--so maybe that justifies the condescension. Head bowed, I leave to go work on my smaller, lower-end images. :)

Ross

>Au Contraire! I spent a week in Jeff's class at the Santa Fe
>Workshops simply because I value his judgments so highly. Jeff's
>judgment is EXACTLY what's called for by a large number of working
>professional photographers doing high end jobs. He knows this stuff
>incredibly well and is a talented, generous, and entertaining teacher
>to boot.


Date: Fri, 17 May 2002 20:18:40 -0400
From: Todd Flashner
Subject: Re: Re: 8- and 16-bit

on 5/16/02 10:55 PM, David Riecks wrote:

> Now if what Dan is saying is that there are few images that need full 16 bit
> per channel editing when the destination is a 4 color offset press, then I'm
> inclined to believe that's the case. Dan is a well known expert when it
> comes to prepping images for this form of output and I trust his judgement.
>
> But that doesn't necessarily mean that his conclusion can be extended to
> every and all form of digital output that exists.

I would agree on principle, but principle only goes so far. Until someone demonstrates that to be true we will all go round and round on the merry-go-round of principle Vs published evidence.

What we can thank Dan for in this dialog, if nothing else, is his steadfast commitment to backing up his claims with images. At least we can trust that within the parameters of his manipulations (whether we like his parameters or not), and input/output conditions, Dan will include bona fides to support his claims.

If his detractors - or anyone who's experience differs - shared the same commitment to demonstrate of what they speak this whole dialog would shed a lot more light and a little less heat.

Todd Flashner


Date: Sat, 18 May 2002 21:08:39 -0600
From: Ron Kelly
Subject: Re: Re: 8- and 16-bit

Todd Flashner wrote:

>Until someone demonstrates that to be true

Indeed. It's not sufficient to say that you have to take a semi trailer to the supermarket every time; you have to PROVE that it's neccessary before you can say that it is. Just coming home with some nice groceries doesn't do it.


The corrollary is that if you can't show it's necessary, it ain't so.

Ron Kelly


Date: Tue, 21 May 2002 10:38:16 -0300
From: André Vallejo
Subject: 8x16 bits to Dan:What about BW?

Hi Dan.
I've been following the thread about 8 x 16 bits editing,and a question came to my mind? Are the considerations you made valid for BW images, or can we severely degrade the image doing tonalmanipulations in 8 bits/pixel?
Regards,
André


Date: Tue, 21 May 2002 10:32:23 -0400From: "David Chusid"
Subject: 16-bit vs. 8-bit

To those who advocate working in 16-bit:

If Photoshop were to offer 24-bit, would you advocate that? How about 32-bit or 64-bit? At what point would you decide that you had enough data?

More importantly, what criteria would you use to make that decision?

I submit that if one is interested in quality, eventually one has no choice but to use his or her eyes and look at the image. And when one does that one finds that 8-bit works just fine.


Date: Tue, 21 May 2002 11:31:53 -0400
From: Scott Olswold
Subject: RE: 8x16 bits to Dan:What about BW?

André,

I'm not Dan, but when it comes to grayscale, I still prefer working in RGB (even if the photo is B&W) to perform tonal correction and then convert to grayscale using either Channel mixer or the RGB > Lab > Gray route. Then I just tweak the Levels to establish any printer limits and let 'er rip.

Scott


Date: Tue, 21 May 2002 08:47:58 -0600
From: Andrew Rodney
Subject: Re: 16-bit vs. 8-bit

on 5/21/02 8:32 AM, David Chusid wrote:

> If Photoshop were to offer 24-bit, would you advocate that? How about 32-bit
> or 64-bit? At what point would you decide that you had enough data?
>
> More importantly, what criteria would you use to make that decision?

More than 8 bits pre color is all you need and many capture devices deal with 10 bit, 12 bit, 14 bit, 16 bit etc. Photoshop takes all such 3high bit2 data and simply calls it 316 bit2 which is fine. The benefits of high bit editing is taking the best high bits and ending up with 8 good bits per color of data. Often, when devices that only capture 8 bits pre color, you get 1 to 2 bits of noise. So a 10bit capture is just as good in many respects as a 14 bit capture as long as you get more than 8 bits of true data to use.

There is no such device I know of that captures more than 16 bits pre color. I can1t think why anyone would create or need such a device.

When you have high bit data (let1s say 16 bits per color), you have 48 bit color.

I know of no output device that will accept or use more than 8 bits of per color. So at some point in the file1s life, you simply must end up with 8 bits pre color. It1s what you do with the file BEFORE you sample down to 8 bits per color (24 bit color) that counts.

Andrew Rodney


Date: Tue, 21 May 2002 14:10:28 -0400
From: Todd Flashner
Subject: Re: 16-bit vs. 8-bit

on 5/21/02 10:47 AM, Andrew Rodney wrote:

> I know of no output device that will accept or use more than 8 bits of per
> color. So at some point in the file1s life, you simply must end up with 8
> bits pre color. It1s what you do with the file BEFORE you sample down to 8
> bits per color (24 bit color) that counts.

Just for the sake of trivia, Jon Cone says his PiezoBW driver (ver6, and pro
(for the Epson 7000) from inkjetmall.com) will pass 16-bit grayscale PSD
from file to print head.

Todd Flashner


Date: Tue, 21 May 2002 13:53:26 EDT
From: Dan Margulis
Subject: RE: 8x16 bits to Dan:What about BW?

André writes,

>>I've been following the thread about 8 x 16 bits editing,and a question came to my mind? Are the considerations you made valid for BW images, or can we severely degrade the image doing tonalmanipulations in 8 bits/pixel?>>

Technically there is more of a case for 16-bit in B/W than in color images, however it's unlikely to make any real difference.

Bit depth works along the same lines, although it doesn't have nearly as pronounced an effect as, resolution. The higher the resolution of the original capture, the more pixels have to be averaged to achieve final output. This results in more smoothness, more consistency--if that's what you really need. It results in oversoftness, lack of focus, if it's not. Similarly, less resolution provides a crisper, snappier look, if that's what you want, or a harsh and jagged look if it isn't.

One can vary resolution quite a bit without hurting anything but it's definitely possible to scan at a resolution that is so high that it actually harms quality, just as it is to scan at a resolution so low that the result will be disagreeable.

Varying bit depth has a similar effect, but much less of it. That is, pictures corrected in 8-bit will, if the correction was very aggressive, seem very slightly sharper than those done in 16-bit. While the difference is basically inconsequential, if we held a gun to people's head and forced them to choose between two color images, most of the time, if they saw a difference at all, they would choose the one corrected in 8-bit. Some of the time, of course, they'd choose the 16-bit version.

In color, having extra channels softens the image, taking away some of the 16-bit advantage. If one channel is extra harsh it's not such a big deal. In B/W, this effect doesn't exist, so there would be a lot more cases where one might have a slight preference for a 16-bit correction.

Dan Margulis


Date: Tue, 21 May 2002 14:16:41 -0600
From: Ross Fenmore
Subject: Re: 16-bit vs. 8-bit

On Tue, 21 May 2002 10:32:23 -0400, you wrote:

>If Photoshop were to offer 24-bit, would you advocate that? How about 32-bit or 64-bit? At what point would you decide that you had enough data?
>
>More importantly, what criteria would you use to make that decision?

I think many people in business (i.e. professional) look to allocate their resources in a most efficient manner. That might dictate that one use the fewest bits needed, not necessarily the most available. Imaging for scientific applications, or filtering higher-bit image data with more complex algorithms are examples where 8-bits can come up short, but I've been involved in several graphics-related businesses over the years and haven't encountered any real problems that I would attribute to 8-bits per channel editing. The significant question remaining regarding these "8 or 16-bit" threads--when is a demonstration of any such real problems forthcoming?

Ross

Adobe Photoshop training classes are taught in the US by Sterling Ledet & Associates, Inc.