Color management myths and misinformation video

[Czornyj]

If we define colour as unique RGB value

Unique RGB value is not a colour, so it would be wrong and misleading definition.

[mrenters]

Unique RGB value is not a colour, so it would be wrong and misleading definition.

I was talking about unique RGB values within Bill's flower 16 bit image in the sRGB space. Are you saying those aren't colours?

Martin

[Czornyj]

I was talking about unique RGB values within Bill's flower 16 bit image in the sRGB space. Are you saying those aren't colours?

Martin

No, those are only abstract numbers.

[digitaldog]

I also think after reading through this thread that Andrew is going to have his work cut out for him.

Coming to many new posts after my nap  . I do want to address a few items in the order I've read them (up to this post).
1. For those who have made comments about my original question regarding a video, I AM reading and digesting them, I'm not ignoring them. This topic evolved into a lengthy sidebar but that happens here. Remember the "does raw have a color space" set of posts? At one point, I and a few others were believing as we had read "raw has no color space, it's essentially grayscale data" which I believe is an exact quote. That was written for an audience that differs here, for it's audience it worked. My position and understanding has evolved considerably thanks to those posts and from people like Thomas Knoll Eric Walowit and Jack Holms. When GWGill wrote (on my longer post about this subject: I agree that your second post on the topic sounds much more reasonable though.), the long tread and my effort to decipher it was worthwhile! GWGill is in the same respected camp on this subject as Thomas, Jack and Eric so that validation on my understanding was pretty neat. 
2. As the OP, it appears I can update the subject and I'm thinking of doing this. I'm interested in the "does Adobe RGB have more colors than sRGB" discussion because I've learned a lot in one day and I think we're getting somewhere. So maybe the subject should be: Color management myths and misinformation video: Larger gamut, more colors?
3. I think due to the evolution of this thread, trying to rebut the other video's is a waste of time (fools errand) so I think all of you for putting me on task.

I have many more posts to read from last night. That will take some time. Keep em coming.

At this point, the discussion in terms of "does Adobe RGB (1998) have more colors than sRGB" the answer seems to be it could. I also want to take some images in both spaces though ColorThink and extract unique colors to see what happens. Maybe someone's done this already? It is important thus far in my understanding from others to point out some key points.

1. These color space triangles are containers and don't have any color per se until we introduce images. And images is what I believe most if not all of us are interested in here.
2. The image itself is critical to the answer (and a yes or no answer probably isn't going to fly). As expressed, a white document in sRGB and Adobe RGB have the same number of colors.
3. It is important to define what a color is, that a number expressing a color value we can't see isn't a color.

That's all I have so far, and I haven't had a lick of coffee yet. But waking up and seeing all these threads, I wanted to interrupt this show to thank everyone for their time and comments and let you know I have a lot of catching up to do. Hopefully many of you are fast asleep as I write this on the other side of the planet. That will give me some time too. Thanks again!

[bjanes]

I think if I were asked by a neophyte photographer “Does sRGB have more colours than Adobe RGB ?”, my answer would be “No, but Adobe RGB can contain a wider range or gamut of colours than sRGB.  It is not about the ‘number’ of colours but about the range that can be represented [here a picture of the two colour spaces would depict this very well and aid understanding].”.   I would go on to emphasise that in terms of colour spaces it is not the ‘number’ of colours than can be represented but the range or gamut that is important.  I think this is all someone who asks this question would need to know.  If they want to know more there are plenty of excellent books out there – authors Andrew Rodney and Jeff Schewe to name but two.

If it were me I would forget the thermometer analogy and stick to a picture of the two colour spaces (2D or 3D – doesn’t matter).  These show exactly what we are talking about (and no one need die from overheating !).  

Going back to my flower images may shed some light on this topic. For analysis with Colorthink, I resized the same image in ProPhotoRGB and the image converted to sRGB with relative colorimetric rendering to 300 x 200 pixels. If I use Colorthink to obtain a color list of both images, they both have 60,000 values, one value for each pixel in the file.

However, I can have Colorthink extract unique color values (expressed in L*a*b) as shown here:



Here are the results:






The ProPhoto image has more unique colors, 23,398 to 22,354. This difference is not really significant. However, this does not address the range of the colors, and this can be shown in 3D plots of the images. Here, the ProPhoto image is shown in color and the sRGB in white. The range of color is greater for the ProPhoto image with the horse shoe gamut of vision shown on the bottom.



The reason that the sRGB image has fewer unique colors is that with relative colorimetric rendering, some unique colors in the ProPhoto image that are out of sRGB gamut are clipped to the same value, reducing the number of colors. Had I used perceptual rendering that did not clip values, the number of color values would be same in both images. However, current perceptual rendering does not look at the gamut of the colors actually in the image and proceed accordingly with compression. It merely compresses the values by an arbitrary amount, whether or not they are out of gamut. If the compression is insufficient, clipping can and often does occur. Whether or not all of these values are perceptibly different is another matter.

Bill

[digitaldog]

If I can make a suggestion - the best success that I've had in talking to people for whom "gamut" wouldn't be appropriate is to talk about a "range of colors". So e.g., Adobe RGB can represent a greater range of colors than sRGB.

That's exactly the language I use Sandy. Range not number. In fact I was asked to do a 15 lecture to business folks who deal with selling a product that produces color (can't go into details). They were being told that product A produces billions of colors (due to bit depth) while product B only millions of colors. The competition was also showing gamut maps in 2D. I used exactly the same language by saying Gamut and number of colors specified are completely different. Gamut is range of color not number of colors. Bit depth can specify number of colors but not their saturation within the gamut maps (maybe I should have said Chroma).

Talking about "number of colors" generally results in total confusion, because it confuses number of bits in a particular file format with color space.

It is similar in some respect to the confusion between dynamic range and bit depth going back to the old and I believe useful staircase analogy (height of staircase is a very different attribute than the number of steps).

[digitaldog]

Saying "wider gamut" just shifts the burden to defining then "gamut." Besides, "more vivid colors" simply means more colors.

No it doesn't. We're back to the human vs. buffalo analogy. How does more vivid equate to larger number?
Make two documents in Photoshop with only two pixels. One pixel is 128/128/128, the other is 255/0/0. In the second make both pixels 128/128/128.
One document has a more vivid color in one pixel than the other document. Does it have more colors?

[digitaldog]

The reason that the sRGB image has fewer unique colors is that with relative colorimetric rendering, some unique colors in the ProPhoto image that are out of sRGB gamut are clipped to the same value, reducing the number of colors. Had I used perceptual rendering that did not clip values, the number of color values would be same in both images. However, current perceptual rendering does not look at the gamut of the colors actually in the image and proceed accordingly with compression. It merely compresses the values by an arbitrary amount, whether or not they are out of gamut. If the compression is insufficient, clipping can and often does occur. Whether or not all of these values are perceptibly different is another matter.

Thanks Bill, I'm seeing a similar report too. I just did another test using an image that doesn't have much saturation (one of my dogs at White Sands).

The differences are pretty small (4129 vs. 4237) but Adobe RGB does have more unique colors which had me confused as to whether the statement (Adobe has more colors than sRGB) was true. You raise a very interesting point about the rendering intent which makes this all the more confusing! 

[Slobodan Blagojevic]

...RAW files not optimised by ETTR will show excessive noise in the shadows...

Ah, Tony, yes, but everybody knows that ETTR is simply a sleazy Canon's plot to hide just how shitty their sensors are If you shoot Nikon, Sony or Pentax, nothing to worry about.

[MHMG]

An important point to get across, IMO, is that color management isn't nearly as difficult as many try to make it. The problems start when color management breaks down and stops.

Having been practicing fully color managed workflows for nearly twenty years now, I respectfully disagree that color managed workflows are easier today.  A fully calibrated and fully ICC profiled printing pipeline is nearly as hard to maintain properly today as it was in the mid 1990s if not more so, IMHO. The reasons have changed, but not the overall complexity. Back then we had more difficulties making profiles (e.g., limited hardware and software choices), but today we have more new OS and printer driver updates routinely busting some part of the color managed chain, often doing so in an effort to "simplify it" for the average consumer.  

For example, I recently discovered that the Apple Colorsync feature used with typical consumer level software for those who don't want to pay the Adobe price of admission hands off to the typical printer driver using relcol with no BPC, and Apple has essentially taken away rendering intent choice from the consumer. So, if you are printing to PDF or to glossy photo papers with high color gamut and dmax this lack of rendering intent control is arguably OK, but one will get decidedly plugged up shadows for any paper that doesn't print with a high Dmax value.  I can understand hiding rendering intent options from the typical consumer to avoid novice confusion, but Relcol without BPC rather than perceptual as the unchangeable default?  What were apple software engineers thinking?  Other than using Adobe software, the only viable way to print to a wide range of media on a Mac these days is to open the document in the Colorsync Utility and print from there. What a PITA yet it's the only way on a new Mac to invoke  a perceptual rendering intent, one of the cornerstones in any color managed workflows. In other words, using Colorsync as a pathway to the printer driver with widely used non color managed apps (e.g., WORD, Pages, Pixelmator, etc.etc) won't let you choose the perceptual rendering intent or any other intent for that matter other than RelCol.  That's but one example to illustrate where modern color management seems to have taken a serious step backward every time we think it's moving one step easier

cheers,
Mark
http://www.aardenburg-imaging.com

[smthopr]

I think trying to talk about the “number of colours” in a given colour space is a complete red herring and will confuse the target audience of Andrew’s proposed video.

Trying to keep it simple…

As I understand it (and I am ready to be corrected on this) a colour space defines a gamut or ‘range’ of colours.  It is, I believe, misleading to talk about a gamut or range as containing a given “number” of colours.  To put it very simply the gamut or range defines the boundaries of what can be represented.

Still keeping it simple my understanding is that bit depth defines the number of differences between colours that can be represented and would, therefore, be colour space agnostic.

I think if I were asked by a neophyte photographer “Does sRGB have more colours than Adobe RGB ?”, my answer would be “No, but Adobe RGB can contain a wider range or gamut of colours than sRGB.  It is not about the ‘number’ of colours but about the range that can be represented [here a picture of the two colour spaces would depict this very well and aid understanding].”.   I would go on to emphasise that in terms of colour spaces it is not the ‘number’ of colours than can be represented but the range or gamut that is important.  I think this is all someone who asks this question would need to know.  If they want to know more there are plenty of excellent books out there – authors Andrew Rodney and Jeff Schewe to name but two.

If it were me I would forget the thermometer analogy and stick to a picture of the two colour spaces (2D or 3D – doesn’t matter).  These show exactly what we are talking about (and no one need die from overheating !). 

This approach seems reasonable...

[Rhossydd]

but today we have more new OS and printer driver updates routinely busting some part of the color managed chain, often doing so in an effort to "simplify it" for the average consumer.

As far as I'm aware this is only effects Mac users. The Windows CM system hasn't really changed in the last 15 years. Almost all issues that have changed with respect to colour management on software on Windows have been as a result of trying to deal with Mac changes (eg loosing the NCM print option in Photoshop).

[Simon Garrett]

As far as I'm aware this is only effects Mac users. The Windows CM system hasn't really changed in the last 15 years. Almost all issues that have changed with respect to colour management on software on Windows have been as a result of trying to deal with Mac changes (eg loosing the NCM print option in Photoshop).

Not quite true!  Microsoft introduced a new colour management system called "Windows Color System" with Vista in around 2008. 

However, as with Mac OS it's not automatic.  That is, programs have to choose to use it, and use its APIs, in order to get colour management.  Not all Microsoft programs do colour management even now - e.g. Internet Explorer is not properly colour managed. 

[Jim Kasson]

The Windows CM system hasn't really changed in the last 15 years.

There was a change with Vista, which caused some problems -- all fixed now, as far as I can tell.

http://www.microsoft.com/en-us/news/press/2005/sep05/09-13vistacanonpr.aspx

Jim

[digitaldog]

I didn't even worry about color space until the recent Fong blowup and after perusing various websites, I think I have come to the conclusion that Gary Fong got the right answer using the wrong math.

He didn't even get to math, that would have been even more hilarious.

For workflow purposes, why put yourself through extra steps by shooting aRGB?

For superior output? Bill illustrated that in the locked post, and Gary admitted it too.

You wind up having to convert it to sRGB a lot of times for sharing online/on tablets/etc. and even for some printing services apparently.

Yes you do if your goal is to optimized the data for the output. If you don't care (or don't know this) then yes, stick to sRGB.

So it seems like for most (not all) people, it'd be more time-efficient to shoot JPG + RAW and have the JPG on sRGB.

Yes and no (maybe). If your goal is quick and dirty (OK, quick and clean), just skip raw. If your goal is to capture all the data, not have some in-camera processor decide what is 'correct color', you want raw. And rendering the image for output is a fundamental part of photography if you have an interest in that process (see: http://www.lumita.com/site_media/work/whitepapers/files/pscs3_rendering_image.pdf). Again, if your goal is to produce 'snap shots', you don't care about how the image is rendered, stick with sRGB and move on.

This won't work for some people but I think it would work for most people.

It is dangerous to make assumptions and speak for anyone but yourself in this regard.

Y'all can argue about color management in an ideal world, but we don't live in such an ideal world yet, so shooting sRGB JPG + RAW seems like the most time-efficient way to deal with current realities, yes?

If that workflow makes sense to you, works for you, then the answer for you is clear. And I don't see too many arguments about color management, at least in the locked thread. There were agreements about technically correct teachings of a technical subject with the person who was writing technically incorrect statements  . I don't recall anyone suggesting not implementing color management.

[Rhossydd]

Y'all can argue about color management in an ideal world, but we don't live in such an ideal world yet, so shooting sRGB JPG + RAW seems like the most time-efficient way to deal with current realities, yes?

That can work, yes.
However if you have any aspirations to improve your photos and do any post processing on them, eg cropping, adjusting colour balance/exposure/generally getting the best from your camera, you're going to have to use some sort of image editor. Once you choose to do that you might as well just use RAW in the first place, then pass everything though a package like Lightroom/Aperture/Capture One and let that handle resizing and colourspace issues for online use and they also do a great job of optimising home printing too. Then you're not stuck with twice as many image files to deal with and get the best image quality.

[Rhossydd]

Not quite true!  Microsoft introduced a new colour management system called "Windows Color System" with Vista in around 2008.  

But it didn't significantly change or break anything!
The only real difference was that it made installing profiles easier as it added an 'install profile' option to the file's properties on the right click menu.
Yes, there are a few more obscure options in the, well buried, system menus for colour management, but the defaults work fine and even the most advanced users needn't worry about changing anything there.

[digitaldog]

Having been practicing fully color managed workflows for nearly twenty years now, I respectfully disagree that color managed workflows are easier today.

I think they are overall (look at Lightroom as an example) but I agree, there are lots of areas that stuff can break and has. Especially for printing which isn't getting the love from Apple and other's it used to.

[PeterAit]

I confess that I do not see what all the fuss is about regarding color management. To me, it is a tool for getting good prints, and the inner technical workings of it are of no interest to me unless this knowledge helps me get better prints and web images. When I was starting out in digital photography, a friend who is an experienced photographer took me to his studio and taught me the following basics:

• Always shoot RAW
• Get a wide-gamut monitor and a calibration system and use the latter at least 1x/month
• In software, always work in ProPhoto
• Make sure you have the right profile for your printer and paper
• Apply the profile in software or at the printer, never both
• Softproofing is your friend

Since the very beginning my prints have matched my monitor and my web images have looked the way I want them to look. So, where's all the complexity?

[digitaldog]

This is course opens up whether we should be capturing and editing in higher bit depths. I would definitely argue that we need to, even if we can't see any difference because it gives us more latitude to manipulate and adjust the image before it starts to posterize.

No question about that! Especially given the high bit data we're starting with from our capture devices. Unlike the 'what is better Adobe RGB or sRGB?' debate some who's name we will not mention use to attract attention to themselves, I've not yet seen the same people argue high bit capture and editing is a poor idea. But let's not give them any ammo in that respect. Those same people will tell their audience there's no reason to capture raw data!