Pages: 1 ... 19 20 [21] 22 23 ... 27   Go Down

Author Topic: sRGB vs. Adobe RGB: New color management stand up comic  (Read 124388 times)

supercurio

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 58
  • Hi! New here
    • Google+
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #400 on: August 27, 2014, 04:24:25 pm »



Wow Bill that's a great example!

Flowers are some of the best gamut stress-test, nowadays when I go out and observe them I tend to think: Oh wow this one is way out of sRGB  ;D
(And snap a picture in RAW for quality control purposes)

I suppose some flowers can be not only out of sRGB or Adobe RGB coverage but also out of some camera sensors gamuts.
But even in the later case, increasing saturation in editing (like in Lighroom) can produce colors going very far.

Theoretical images, generated can do the trick just as well but flowers are rather more convincing as they're right here in nature and part of many pics.

By the way two questions:

1/
Bill, you are using your scanner to evaluate the print: How wide is your scanner's color space?

2/
Same thing for camera: Did someone already published the measured gamut of today's digital camera sensors?
I have this little project idea for a few months, I wonder if anyone would be interested seeing that.
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #401 on: August 27, 2014, 04:30:19 pm »

Same thing for camera: Did someone already published the measured gamut of today's digital camera sensors?
I have this little project idea for a few months, I wonder if anyone would be interested seeing that.
It terms of gamut discussed over these pages, that's going to be really difficult to pin down.
Check out Wikipedia's article on "CIE 1931 color space"

 <http://en.wikipedia.org/wiki/CIE_1931_color_space <http://en.wikipedia.org/wiki/CIE_1931_color_space> > and also the "color space" article: <http://en.wikipedia.org/wiki/Color_space >

Digital cameras don't have a gamut, but rather a color mixing function. Basically, a color mixing function is a mathematical representation of a measured color as a function of the three standard monochromatic RGB primaries needed to duplicate a monochromatic observed color at its measured wavelength. Therefore, the measured pixel values don't even *get* a gamut until they're mapped into a particular RGB space. Before then, *all* colors are (by definition) possible. This has been discussed in these parts in the past.

Update: http://www.luminous-landscape.com/forum/index.php?topic=22471.200
« Last Edit: August 27, 2014, 04:33:45 pm by digitaldog »
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

supercurio

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 58
  • Hi! New here
    • Google+
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #402 on: August 27, 2014, 04:37:36 pm »

It terms of gamut discussed over these pages, that's going to be really difficult to pin down.
Check out Wikipedia's article on "CIE 1931 color space"

 <http://en.wikipedia.org/wiki/CIE_1931_color_space <http://en.wikipedia.org/wiki/CIE_1931_color_space> > and also the "color space" article: <http://en.wikipedia.org/wiki/Color_space >

Digital cameras don't have a gamut, but rather a color mixing function. Basically, a color mixing function is a mathematical representation of a measured color as a function of the three standard monochromatic RGB primaries needed to duplicate a monochromatic observed color at its measured wavelength. Therefore, the measured pixel values don't even *get* a gamut until they're mapped into a particular RGB space. Before then, *all* colors are (by definition) possible. This has been discussed in these parts in the past.


Recently I've been writing routines (matrix solvers) that generate 3x3 matrices for DNG ColorMatrix and ForwardMatrix under two illuminants (usually A and D65)
Both (or their inverse) then allow to convert from sensor RGB to XYZ.
DNG ColorMatrix does the white balance at the same time, and ForwardMatrix works on already white-balanced RGB sensor values.

I'll only be sure once/if I succeed, but the idea seems likely to be possible at first thought.
If you're telling me it's not possible I'm gonna have to try  ;)

Edit: thanks Andrew for the link, ha! all the cool stuff has already been discussed in this forum.
« Last Edit: August 27, 2014, 04:42:02 pm by supercurio »
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #403 on: August 27, 2014, 04:42:23 pm »

Both (or their inverse) then allow to convert from sensor RGB to XYZ.
If you're telling me it's not possible I'm gonna have to try  ;)
I'm not telling you, I'm reporting what a number of other's such as Thomas Knoll, Eric Walowit have said and the important item expressed was: Therefore, the measured pixel values don't even *get* a gamut until they're mapped into a particular RGB space, and hasn't that happened above?

Update. After that long set of posts about 'does raw have a color space', here's my copy and paste based on what I think I understand from Thomas, Eric, Chris Murray and Jack Holm:

Quote
Here’s my copy and paste with respect to cameras based on a number of conversations with Eric Walowit and Jack Holms, both on the ICC digital photo group and Thomas Knoll.
Digital cameras don't have a gamut, but rather a color mixing function. Basically, a color mixing function is a mathematical representation of a measured color as a function of the three standard monochromatic RGB primaries needed to duplicate a monochromatic observed color at its measured wavelength. Cameras don’t have primaries, they have spectral sensitivities, and the difference is important because a camera can capture all sorts of different primaries. Two different primaries may be captured as the same values by a camera, and the same primary may be captured as two different values by a camera (if the spectral power distributions of the primaries are different). A camera has colors it can capture and encode as unique values compared to others, that are imaginary (not visible) to us. There are colors we can see, but the camera can't capture that are imaginary to it. Most of the colors the camera can "see" we can see as well. Yet some cameras can “see colors“ outside the spectral locus however every attempt is usually made to filter those out. Most important is the fact that cameras “see colors“ inside the spectral locus differently than humans. No shipping camera that I know of meets the Luther-Ives condition. This means that cameras exhibit significant observer metamerism with respect to humans. The camera color space differs from a more common working color space in that it does not have a unique one to one transform to and from CIE XYZ. This is because the camera has different color filters than the human eye, and thus "sees" colors differently. Any translation from camera color space to CIE XYZ space is therefore an approximation.

The point is that if you think of camera primaries you can come to many incorrect conclusions because cameras capture spectrally. On the other hand, displays create colors using primaries. Primaries are defined colorimetrically so any color space defined using primaries is colorimetric. Native (raw) camera color spaces are almost never colorimetric, and therefore cannot be defined using primaries. Therefore, the measured pixel values don't even produce a gamut until they're mapped into a particular RGB space. Before then, *all* colors are (by definition) possible.

Raw image data is in some native camera color space, but it is not a colorimetric color space, and has no single “correct” relationship to colorimetry. The same thing could be said about a color film negative. Someone has to make a choice of how to convert values in non-colorimetric color spaces to colorimetric ones. There are better and worse choices, but no single correct conversion (unless the “scene” you are photographing has only three independent colorants, like when we scan film).

Do raw files have a color space? Fundamentally, they do, but we or those handling this data in a converter may not know what that color space is. The image was recorded through a set of camera spectral sensitivities which defines the intrinsic colorimetric characteristics of the image. One simple way to think of this is that the image was recorded through a set of "primaries" and these primaries define the color space of the image.

If we had spectral sensitivities for the camera, that would make the job of mapping to CIE XYZ better and easier, but we'd still have decisions on what to do with the colors the camera encodes, that are imaginary to us.
« Last Edit: August 27, 2014, 04:44:46 pm by digitaldog »
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

supercurio

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 58
  • Hi! New here
    • Google+
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #404 on: August 27, 2014, 04:49:41 pm »

I'm not telling you, I'm reporting what a number of other's such as Thomas Knoll, Eric Walowit have said and the important item expressed was: Therefore, the measured pixel values don't even *get* a gamut until they're mapped into a particular RGB space, and hasn't that happened above?

Yes I was being silly.
Okay so if it hasn't been done and published before, I'll see if it can be done: it would be nice to be able to compare camera's sensor gamut.

Of course it will mostly be a curiosity as 3x3 matrices are not enough to map sensor data to exact color anyway, despite they're sufficient to convert RGB to XYZ and vice-versa.
(we would need cameras capturing the visible spectrum instead of RGB channel based on an defined spectral response for something closer to real colors under any illuminant)
Logged

Czornyj

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1948
    • zarzadzaniebarwa.pl
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #405 on: August 27, 2014, 04:50:31 pm »

Bill,

I think this image is not a good candidate to test sRGB vs >sRGB colour spaces on c-prints, there's to little blue-turquise-phtalogreen colors.

Fuji Crystal Archive (which is a very shitty paper) is not a good choice either. So or so the profile form Costco looks a little bit weird, which raises a question how it was made - my profile of Fuji CA from Noritsu QSS3701HD has 455k dE^3, and profiles of DP2 and Endura Supra are rather in >500k dE^3 territory.

Sad truth is that small format wet dry printer controllers are dirty little bastards, so without knowing what the operator is really doing such tests are not really useful...
« Last Edit: August 27, 2014, 04:53:51 pm by Czornyj »
Logged
Marcin Kałuża | [URL=http://zarzadzaniebarwa

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #406 on: August 27, 2014, 04:53:45 pm »

So or so the profile form Costco looks a little bit weird, which raises a question how it was made - my profile of Fuji CA from QSS3701HD has 455k dE^3, and profiles of DP2 and Endura Supra are rather in >500k dE^3 territory.
I'll let Bill answer the specifics but if it came from here: http://www.drycreekphoto.com, Ethan knows what he's doing IMHO.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Czornyj

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1948
    • zarzadzaniebarwa.pl
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #407 on: August 27, 2014, 04:56:27 pm »

I'll let Bill answer the specifics but if it came from here: http://www.drycreekphoto.com, Ethan knows what he's doing IMHO.

I know very well that Ethan knows what he's doing, but maybe he couldn't do what he would like to do in this case ;) (just a wild guess)
« Last Edit: August 27, 2014, 04:58:10 pm by Czornyj »
Logged
Marcin Kałuża | [URL=http://zarzadzaniebarwa

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #408 on: August 27, 2014, 05:01:05 pm »

I know very well that Ethan knows what he's doing, but maybe he couldn't do what he would like to do in this case ;) (just a wild guess)
It's possible and I agree, this probably isn't the best, consistent and potentially well run device to do such tests. And considering Gary said something about a 'pro lab with an inkjet printer', I'd be more comfortable using such a device for this kind of test. Bet the color gamut is a lot larger too.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #409 on: August 27, 2014, 06:03:33 pm »

Update, I now see the images Bill.
Curious what rendering intent was used for the print output color space? Did you even have a choice?

For the matrix based conversions, relative colorimetric was used since that is the only choice. I also used RC with the custom profile. The colors were so far out of gamut that perceptual did not make much difference. BTW, I modified my original post to add the scanned photos, which were inadvertently omitted in the original post.

Bill
Logged

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #410 on: August 27, 2014, 06:11:04 pm »

Wow Bill that's a great example!

Flowers are some of the best gamut stress-test, nowadays when I go out and observe them I tend to think: Oh wow this one is way out of sRGB  ;D
(And snap a picture in RAW for quality control purposes)

I suppose some flowers can be not only out of sRGB or Adobe RGB coverage but also out of some camera sensors gamuts.
But even in the later case, increasing saturation in editing (like in Lighroom) can produce colors going very far.

Theoretical images, generated can do the trick just as well but flowers are rather more convincing as they're right here in nature and part of many pics.

By the way two questions:

1/
Bill, you are using your scanner to evaluate the print: How wide is your scanner's color sp

2/
Same thing for camera: Did someone already published the measured gamut of today's digital camera sensors?
I have this little project idea for a few months, I wonder if anyone would be interested seeing that.

I don't really know the gamut of the scanner. Strictly speaking it does not have a gamut, since it will output something for whatever is placed on it. The same is true for the camera. It  is a low end Epson and the scans look a bit dull compared to the prints, which I evaluated visually. I added the scanned image to my original post. It did not make it into the original post.

The camera was the Nikon D800e with the raw rendered by ACR into ProPhoto with no clipping. The raw file has a very wide color range.

Bill
Logged

Czornyj

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1948
    • zarzadzaniebarwa.pl
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #411 on: August 27, 2014, 06:11:59 pm »

For the matrix based conversions, relative colorimetric was used since that is the only choice. I also used RC with the custom profile. The colors were so far out of gamut that perceptual did not make much difference. BTW, I modified my original post to add the scanned photos, which were inadvertently omitted in the original post.

Bill

There's no choice when you're converting to matrix profile, but when you're converting from matrix profile to LUT profile, then of course you have a choice.
« Last Edit: August 27, 2014, 06:14:50 pm by Czornyj »
Logged
Marcin Kałuża | [URL=http://zarzadzaniebarwa

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #412 on: August 27, 2014, 06:24:35 pm »

Bill,

I think this image is not a good candidate to test sRGB vs >sRGB colour spaces on c-prints, there's to little blue-turquise-phtalogreen colors.

Fuji Crystal Archive (which is a very shitty paper) is not a good choice either. So or so the profile form Costco looks a little bit weird, which raises a question how it was made - my profile of Fuji CA from Noritsu QSS3701HD has 455k dE^3, and profiles of DP2 and Endura Supra are rather in >500k dE^3 territory.

Sad truth is that small format wet dry printer controllers are dirty little bastards, so without knowing what the operator is really doing such tests are not really useful...

Thanks for your comments. I'm not very expert in these matters and merely selected an image with vivid colors. I could try other other images, but I think that it would demonstrate what we already know: sRGB is not the best space in which to send images to this printer. However, one can get passable results with sRGB for noncritical family photos, etc., and the price is quite reasonable. For my more critical work, I print with my Epson 3880 and am happy with the results.

Bill
Logged

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #413 on: August 27, 2014, 06:28:06 pm »

I know very well that Ethan knows what he's doing, but maybe he couldn't do what he would like to do in this case ;) (just a wild guess)

Yes, I downloaded the profile from Drycreek.com

Bill
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #414 on: August 27, 2014, 06:50:47 pm »

Thanks for your comments. I'm not very expert in these matters and merely selected an image with vivid colors. I could try other other images, but I think that it would demonstrate what we already know: sRGB is not the best space in which to send images to this printer.
I think the work you did was a superb demonstration dismissing much of what Will and Gary have been preaching which is the crux of this thread. It illustrates a number of data points that one would hope Will and Gary would examine and comment on, namely:

1. Using sRGB to their so called 'sRGB printer' doesn't produce optimal results. Even Gary admitted this from his prints!
2. Using a real world image with saturated colors is more effective in evaluating the output than using a target. Consider Bill's notes on the detail seen in saturated colors on the print. We'd not see this if the image was a target who's surface is a rather smooth color.
3. Bill used an outside lab and even placed the orders with different parameters, so his analysis is based on real rubber hitting the road.

Hopefully Gary and/or Will can examine just this one test and see that what many have been stating about RGB working space choices for output to print play a role and as I said on his site, sRGB is never the right answer for the best print output (unless again, you are forced by some lab that is rather cluless about color management to send them sRGB).

Further:
Quote
Strictly speaking it does not have a gamut, since it will output something for whatever is placed on it.
Exactly. With a scanner, the target (film or a print) has a gamut, and ideally matches or exceeds what we'll scan. Much more difficult with a digital camera. Over the years, we've seen companies attempt to design differing targets, Gary's SG comes to mind. GretagMacbeth wanted a camera target that had a wider gamut so they updated the original with glossy, saturated patches which wrecked havoc on the profiling software if there was any reflections on a patch. Then they produced a 3rd generation target, the DC with no glossy patches but with a smaller gamut as a result.
« Last Edit: August 27, 2014, 06:55:08 pm by digitaldog »
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Tony Jay

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2965
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #415 on: August 27, 2014, 07:40:03 pm »

One other thing to add to this debate.

Even if one is using an external photo lab to make one's prints there is still NEVER a reason to restrict oneself to sending images tagged with sRGB.
Choice allows one to to choose not to use labs that insist on sRGB-tagged images and that do not use a colour-managed workflow.
There is NEVER a good reason to arbitrarily restrict the potential colour gamut of printer output.

Tony Jay
Logged

MarkM

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 428
    • Alaska Photographer Mark Meyer
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #416 on: August 27, 2014, 08:01:51 pm »


Even if one is using an external photo lab to make one's prints there is still NEVER a reason to restrict oneself to sending images tagged with sRGB.

But there is, and it's Gary's point, which if he avoided his explanation would be pretty sound for a certain percentage of the camera-owning public. The reason to restrict yourself to sending sRGB is that you are my grandmother. She's going to pull images right off her card and send them to her friends — probably as full-res 12MB attachments unless some friendly email client fixes for her. This is the same way she is going to order prints, too. She's not going to buy photoshop, or download dry creek profiles, or process raw images, she's certainly not going to contemplate the difference between a relative and perceptual rendering intent. She is why we have sRGB jpegs and they're perfect for her. When she hands me her camera and asks me to set it up, it's getting set for sRGB jpegs in P mode — at least until you can be reasonably sure that her lab will be color managed.

The problem, other than Gary's  technical explanation, is that presumably if you're watching a video on youtube about photography, you have a greater interest in the details than my grandmother and you are hoping to move beyond her level of skill. If that's the case then maybe a better, more complete explanation is in order that tells the whole story correctly. Otherwise, Andrew's 30 second explanation is all you need.
Logged

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #417 on: August 27, 2014, 08:26:39 pm »

That's a really good example of the problems and solutions. One image that didn't show up inline in your post that really demonstrates the difficulty is the last one with the four scans: http://bjanes.smugmug.com/Costco-Gamut/i-RjcpcFc

Also, it would be interesting to see the original proPhotoRGB image itself. It would also be interesting to see what happens if you shot it in sRGB as a jpeg — do the camera's internal conversions handle the clipped colors more gracefully than the colorimetric conversions?

Mark,

The original ProPhoto image was obtained from a focus stack of several raw files rendered into 16 bit ProPhotoRGB. There is slight clipping in the red channel even in ProPhotoRGB and the image was admittedly over-processed. However the print on my Epson 3880 gave highly saturated colors with no notable blocking up of the highlights. Here is a link to the original TIFF:

http://adobe.ly/1nFgN4C

And here is a link to one of the NEFs. There is no clipping of the raw channels.

NEF:

http://adobe.ly/1vnZ35T

The setup is long gone and it is not possible to re-shoot it. One could process the NEF in the latest Nikon raw processing software (Capture NX 2), which reportedly duplicates the internal camera JPEG processing. I don't have that software, but anyone who has it can give it a try. Anyone is given permission to use the images for further testing.

Bill
Logged

Tony Jay

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2965
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #418 on: August 27, 2014, 08:30:35 pm »

But there is, and it's Gary's point, which if he avoided his explanation would be pretty sound for a certain percentage of the camera-owning public. The reason to restrict yourself to sending sRGB is that you are my grandmother. She's going to pull images right off her card and send them to her friends — probably as full-res 12MB attachments unless some friendly email client fixes for her. This is the same way she is going to order prints, too. She's not going to buy photoshop, or download dry creek profiles, or process raw images, she's certainly not going to contemplate the difference between a relative and perceptual rendering intent. She is why we have sRGB jpegs and they're perfect for her. When she hands me her camera and asks me to set it up, it's getting set for sRGB jpegs in P mode — at least until you can be reasonably sure that her lab will be color managed.

The problem, other than Gary's  technical explanation, is that presumably if you're watching a video on youtube about photography, you have a greater interest in the details than my grandmother and you are hoping to move beyond her level of skill. If that's the case then maybe a better, more complete explanation is in order that tells the whole story correctly. Otherwise, Andrew's 30 second explanation is all you need.
I fully accept that may happen - by choice.
However, taking Gary Fong's explanation at face value it would appear that one has no choice.
Obviously that is nonsense and the point of my post - and it still stands - is that there is no reason to arbitrarily restrict the gamut of printer output.

Tony Jay
Logged

Slobodan Blagojevic

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 18090
  • When everyone thinks the same, nobody thinks
    • My website
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #419 on: August 27, 2014, 08:31:57 pm »

...Even if one is using an external photo lab to make one's prints there is still NEVER a reason to restrict oneself to sending images tagged with sRGB.
Choice allows one to to choose not to use labs that insist on sRGB-tagged images and that do not use a colour-managed workflow.
There is NEVER a good reason to arbitrarily restrict the potential colour gamut of printer output.

NEVER say never, Tony  ;)

I send my files for 20x30 canvas in sRGB because my print lab says so. If any of you pro-choice guys would be willing to prove your point and bankroll me to switch to a real pro lab, that charges 2-3 times more, hey, I would be eternally grateful.

P.S. And no, I am not Mark's grandmother ;)
« Last Edit: August 27, 2014, 08:34:00 pm by Slobodan Blagojevic »
Logged
Pages: 1 ... 19 20 [21] 22 23 ... 27   Go Up