Pages: 1 ... 7 8 [9] 10 11 ... 27   Go Down

Author Topic: sRGB vs. Adobe RGB: New color management stand up comic  (Read 124410 times)

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 #160 on: August 24, 2014, 07:17:33 pm »

I didn't come here asking questions about the test.  I just want to know why you think it's going to be flawed when I do the commercial application of a very simple test.
You've been told more than once! Is this again the reading comprehension issue? The target is all wrong. The mode of capture is less than ideal.
Quote
Can people tell the difference, and if not, is it worth venturing out of the sRGB color space?  That's all I want to determine.
Look, if you want the test to produce a predetermined outcome, just tell us and we'll produce something that tells you exactly what you want to hear. Then we can dismiss the test. If you really want to know what the differences in output looks like using sRGB vs. Adobe RGB (1998) I provided a very, very simple test with an image that exists and all you have to do now is correctly treat the data upon output to the correct kind of device (a pro inkjet as you point out but don't define).
Quote
If AdobeRGB is so great, then let the results show that it's nearly unanimous
You've been provided numerous and well thought out methods for doing this and dismissed or ignored them. You're just wasting everyone’s time here but since you say your video will link to these discussions, it does at least allow those who wish to understand the level of your B.S. factor to be gauged by reading your posts.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

MarkM

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 428
    • Alaska Photographer Mark Meyer
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #161 on: August 24, 2014, 07:18:19 pm »

I didn't come here asking questions about the test.  I just want to know why you think it's going to be flawed when I do the commercial application of a very simple test.  A photo will be taken one after another of a color image, and kept in that workflow, to see if people can tell the difference.  This is because I have never seen such a enormous discussion about why it is so important to do exactly a refined workflow.

Can people tell the difference, and if not, is it worth venturing out of the sRGB color space?  That's all I want to determine.  If AdobeRGB is so great, then let the results show that it's nearly unanimous - people can really tell the AdobeRGB is better.  If not, then well, you are arguing over subtle differences that most people can't even pick up.

Gary, you've been told why it's flawed and are choosing to ignore it. The largest problem is that, with one exception, the color checker patches all fit within the sRGB gamut. If you use a proper workflow the prints will be identical. Any difference between these prints will the results of errors in your workflow, not differences of the colorspaces.
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 #162 on: August 24, 2014, 07:18:32 pm »

I do not defend it, at least not in its entirety. You will remember that I was among the first to say he got the message right, but explanation wrong (especially the rainbow part).
Good, then we agree and can move on. Maybe you can help Gary with this new testing protocols, the few of us who've tried are getting nowhere.
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 #163 on: August 24, 2014, 07:22:39 pm »

Good, then we agree and can move on. Maybe you can help Gary with this new testing protocols, the few of us who've tried are getting nowhere.

Same, I enjoyed talking to you guys as newcomer on this forum.
But the various attempts at helping Gary didn't seem to produce much results so far, so that's enough time spend on that.

I look forward discussing with you on other stuff :)
Logged

mouse

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 260
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #164 on: August 24, 2014, 07:24:48 pm »

Mark,

Many thanks for this very lucid explanation.


Here's what's going on:
If you give me an sRGB color like [0, 255, 0] — that's the brightest, most-saturated green possible in sRGB. If I just take those numbers and send them to display system with no color management whatsoever, the display is going to output its brightest, most saturated green. If that display is a device like an iPhone with a gamut very similar to sRGB, these two versions of green will be almost the same. But if that display is a wide gamut display its brightest most saturated green is going to be a substantially different color.  

Going the other direction, suppose I have AdobeRGB's visual equivalent of sRGB (0, 255, 0). That color is about (144, 255, 60) in AdobeRGB. In a color managed environment sRGB (0, 255, 0) and AdobeRGB (144, 255, 60) will match visually. In sRGB this color represents 100% saturation, but in AdobeRGB it represents only about 77% saturation. So what happens if I send the AdobeRGB numbers to sRGB-like display? It will send the numbers (144, 255, 60) directly to output, so rather than getting 100% of the display's saturation I only get about 77% resulting in a duller-than-expected color. If I send those numbers to a wide-gamut display with characteristics similar to AdobeRGB, 77% of it's most saturated green will closer to what we were aiming for.

Of course all of these scenarios can be avoided in a color managed environment where the system would convert colors to the output profile before displaying them.

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 #165 on: August 24, 2014, 07:27:09 pm »

... Learning about color science is a long process I believe, and countless times I realized that what I thought was correct was either wrong or a gross approximation, no doubt it will happen again...

Exactly.

However, we are, but Garry's audience is not the one willing to embark on that long journey, full of trials and errors, reading books on the subject, white papers, four-hour videos, downloading beta software, searching web, buy wide-gamut monitors, profiling devices, etc., etc. just so they can answer the question "shall I use sRGB or Adobe RGB so that I can post it to Facebook or send it to lab for printing."

Tony Jay

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2965
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #166 on: August 24, 2014, 07:27:58 pm »

I didn't come here asking questions about the test.  I just want to know why you think it's going to be flawed when I do the commercial application of a very simple test.  A photo will be taken one after another of a color image, and kept in that workflow, to see if people can tell the difference.  This is because I have never seen such a enormous discussion about why it is so important to do exactly a refined workflow.

Can people tell the difference, and if not, is it worth venturing out of the sRGB color space?  That's all I want to determine.  If AdobeRGB is so great, then let the results show that it's nearly unanimous - people can really tell the AdobeRGB is better.  If not, then well, you are arguing over subtle differences that most people can't even pick up.
Gary for your test to have any scientific credentials it needs to control for all variables except the one that is being tested.
Your proposal has so many uncontrolled variables that it would be impossible to draw any useful conclusions since whatever was found it would not be possible to work out why there were any differences or similarities in the test results.
This has already been pointed out complete with details! Yes.

There are plenty of PhD's and engineers who frequent this site who can help you plan a series of tests that actually pass scientific muster.
You stating that "I didn't come here asking questions about the test" is again not constructive.
Currently it matters little what results you get - they will have absolutely no credibility.

My previous suggestion to actually learn about colour management from scratch still stands.

Tony Jay
Logged

Czornyj

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1948
    • zarzadzaniebarwa.pl
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #167 on: August 24, 2014, 07:38:37 pm »

However, we are, but Garry's audience is not the one willing to embark on that long journey, full of trials and errors, reading books on the subject, white papers, four-hour videos, downloading beta software, searching web, buy wide-gamut monitors, profiling devices, etc., etc. just so they can answer the question "shall I use sRGB or Adobe RGB so that I can post it to Facebook or send it to lab for printing."

That's still not the reason to spread BS about it.
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 #168 on: August 24, 2014, 07:39:19 pm »

However, we are, but Garry's audience is not the one willing to embark on that long journey, full of trials and errors, reading books on the subject, white papers, four-hour videos, downloading beta software, searching web, buy wide-gamut monitors, profiling devices, etc., etc. just so they can answer the question "shall I use sRGB or Adobe RGB so that I can post it to Facebook or send it to lab for printing."
Many of us were at the same place as Gary's audience and moved on to understand some of these concepts. As I told Gary in one of my first posts to him, (aside from not taking the suggestions personally) is that none of us are born with a knowledge of color, color management or imaging. And yes, we have to decide to work on that understanding, something Gary refuses to do based on his numerous posts here and on his two video sites. Ultimately, whatever misunderstanding these beginners receive that could result in poor results or a desire to ignore the topic are largely Gary's fault. He's apparently beyond help. I'd hope that some of his audience isn't and will read the feedback from those who called him out and maybe come here for useful information.

Ultimately this hasn’t been at all a waste of time. We gained a few new intelligent posters (with one notable exception). There is a paper trail of Gary's writings that he says he'll link to and if not, others will. We made a valid attempt to try to reason with him. It's all here in B&W. So there was something to be gained.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Alan Goldhammer

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4344
    • A Goldhammer Photography
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #169 on: August 24, 2014, 08:48:15 pm »


Also, IMHO both are so limited they're completely outdated and it's very much the time to upgrade to wider gamuts for shooting, edition and display.
Incoming standards are Rec.2020 and also HDR displays, and instead most people are looking in the past.
Today we have dozen of millions of AMOLED displays on consumer products with gamut much wider than anything discussed here so far. It's not even about tomorrow: Its real since the Galaxy S, 5 years ago.

About your test, the methodology is indeed invalid.

You can test one variable at a time.
Either:
- Calibrated camera in sRGB against Adobe RGB -> calibrated printer -> eyes
Or:
- Camera in calibrated RAW -> Output in sRGB against Adobe RGB -> calibrated printer -> eyes

In your experiment, you have too many variables.
- Uncablibrated camera is a variable
- sRGB or Adobe RGB is a variable
- Uncalibrated printer is a variable

The logic says: you cannot test three variables in a single test.

And yes that's correct, digital cameras are not calibrated.
Well, to be more precise: All of them are calibrated (with more or less accuracy), then a custom look is applied.
This is the reason why often one can identify an untouched picture from a Sony camera (tends to have a cold tone), Panasonic camera (typically: muted yellows turning greens), Olympus (vibrant pleasing colors most people like out of the box). Canon look is pretty significant too.

I don't need to alter them their colors are not accurate, it's intentional and most of the time a visual trademark.

Well if you want me to talk in a video why not, that could be fun.
Know that I will have no problem identifying and describing methodologies and misconception in a test tho, I don't care if you use my name, people know me for my work and results already.
Absolutely right about all the handheld display devices.  I just pulled up my website on my Samsung S5 phone and the colors are amazing, actually look a little better than on my wide gamut NEC monitor.  I didn't do anything special other than using the LR web gallery maker which produces sRGB JPGs.  I'm going to have to do some experimenting to see if I go to a wider gamut assignment what the results look like.  It would be nice if LR gives us the option of assigning a different profile in the web gallery section.  I think they do in the normal export tool in the Library.

I agree with you and others about all the variables in this "test."  It's really like the listening tests of high end audio equipment that were also poorly conceived.  I'm pretty biased in this regard having spent my working career in the biopharmaceutical industry where months are spent designing clinical trials to show safety and efficacy of new drugs.  IMO the current test as outlined by Gary is a fool's errand.
Logged

supercurio

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 58
  • Hi! New here
    • Google+
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #170 on: August 24, 2014, 09:10:58 pm »

Absolutely right about all the handheld display devices.  I just pulled up my website on my Samsung S5 phone and the colors are amazing, actually look a little better than on my wide gamut NEC monitor.  I didn't do anything special other than using the LR web gallery maker which produces sRGB JPGs.  I'm going to have to do some experimenting to see if I go to a wider gamut assignment what the results look like.  It would be nice if LR gives us the option of assigning a different profile in the web gallery section.  I think they do in the normal export tool in the Library.

Super AMOLED displays primaries correspond to no existing standard, and the gamma response is between 2.35 and 2.5 average depending on models and samples so you won't find an ICC profile to get color accuracy unless you make it yourself ^^

It's not Adobe RGB at all despite recent claims from Samsung marketing, but in case there's interest here I shall develop that in another thread.

That's a worthy experiment even-though no tool existing currently is able to do that correctly (because generating full screen patterns)
Logged

Schewe

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6229
    • http:www.schewephoto.com
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #171 on: August 24, 2014, 09:18:25 pm »

Can people tell the difference, and if not, is it worth venturing out of the sRGB color space?  That's all I want to determine.  If AdobeRGB is so great, then let the results show that it's nearly unanimous - people can really tell the AdobeRGB is better.  If not, then well, you are arguing over subtle differences that most people can't even pick up.

Again YOUR test is a non-starter because you are shooting a ColorChecker chart that has only one color out of gamut for sRGB. You need to be shooting colors outside of sRGB but within ARGB for your test to make any sense doing...otherwise you are spinning your wheels, proving nothing. But hey, it's your time if you want to waste it. You do your test and publish and we'll all come and blow it up and point out all the errors (which is what we are trying to keep you from doing now).
Logged

Eyeball

  • Full Member
  • ***
  • Offline Offline
  • Posts: 150
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #172 on: August 24, 2014, 09:33:13 pm »

Eyeball, from what I've seen, colors profiles made by Adobe from RAW data (matrices + dcp) are not really accurate, sometimes even really grossly inaccurate.
A concrete example I'm playing with right now is the Panasonic GH4.
Adobe color profile (ForwardMatrix + DCP) is way off, they messed up really bad.

Based on my own measurements, I'm afraid using a camera as input is not applicable for the test Gary is trying to do.

Whether the raw developer used does a perfect job of representing "accurate" color just doesn't seem that important to me for the experiment I outlined.  The key comparison is between two versions of the same image - one output in sRGB and one in AdobeRGB.  My assumption would be that any inaccuracies in the raw development would impact both version equally or nearly so.
Logged

supercurio

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 58
  • Hi! New here
    • Google+
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #173 on: August 24, 2014, 09:37:48 pm »

Whether the raw developer used does a perfect job of representing "accurate" color just doesn't seem that important to me for the experiment I outlined.  The key comparison is between two versions of the same image - one output in sRGB and one in AdobeRGB.  My assumption would be that any inaccuracies in the raw development would impact both version equally or nearly so.

Oh I said that after assuming Gary would ask people to compare prints with the physical colorchecker target itself.
But that was speculation, he didn't go as far when describing his test.
Logged

Alan Goldhammer

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4344
    • A Goldhammer Photography
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #174 on: August 24, 2014, 09:40:05 pm »

Super AMOLED displays primaries correspond to no existing standard, and the gamma response is between 2.35 and 2.5 average depending on models and samples so you won't find an ICC profile to get color accuracy unless you make it yourself ^^

It's not Adobe RGB at all despite recent claims from Samsung marketing, but in case there's interest here I shall develop that in another thread.

That's a worthy experiment even-though no tool existing currently is able to do that correctly (because generating full screen patterns)
If only I could get my i1 Pro to read ArgyllCMS patches from my Android phone I would be able to prepare a profile! ;D
Logged

supercurio

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 58
  • Hi! New here
    • Google+
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #175 on: August 24, 2014, 09:43:45 pm »

If only I could get my i1 Pro to read ArgyllCMS patches from my Android phone I would be able to prepare a profile! ;D

Sure you can! Use the Web target :) You can do that easier with the help of DispcalGUI.
But big warning: measurements will be off in terms of gamma curves (and as a result saturation) as patterns will likely be full screen, tripping Samsung's panel's dynamic contrast algorithm that cannot be disabled.
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 #176 on: August 24, 2014, 09:44:14 pm »

Whether the raw developer used does a perfect job of representing "accurate" color just doesn't seem that important to me for the experiment I outlined.  The key comparison is between two versions of the same image - one output in sRGB and one in AdobeRGB.  My assumption would be that any inaccuracies in the raw development would impact both version equally or nearly so.
Agreed and why I suggested Gary could take my printer test file. Or the Roman 16's, or Bill Atkinsion's Tango scanned images (Lab) and convert to the two color spaces in question and print them. That he is going to use the wrong target and shoot two variations isn't necessary and will skew the results hugely. But that's fine. He'll come up with some ridiculous conclusions about what people tell him about the prints. In a nutshell, if someone were to propose the worst possible way to test the two color spaces for a print, Gary has come up with it.

And again, what's the point of all this nonsense anyway? If X number of people prefer one color space over the other, assuming Gary even prints them correctly which is questionable, it will in no way defend the silly flat earth color theories in the two video's he refuses to fix or delete. So now we'll get a 3rd idiotic video from Gary. He's just trying to gather attention for himself from the unsuspecting 'students' he hopes will sign up and pay for more idiotic videos. He's learning from Rockwell. In his video, towards the end, there's a pitch for his "premium channel".

Best thing to do is just let him do his video and get a good laugh out of it.
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 #177 on: August 25, 2014, 04:45:15 am »

It's not Adobe RGB at all despite recent claims from Samsung marketing, but in case there's interest here I shall develop that in another thread.

It should potentially have more gamut in saturated shadows than AdobeRGB, which is desirable for those who are making prints, as that's usually the part that LCD is lacking. I suppose all LCD desktop, laptop, TV and mobile displays will be replaced by OLED in near future (and conventional projectors with laser projectors).

As for the srgb-centeric wet mini lab printers, all newer Noritsu since 37HD series can automatically recognise and print correctly AdobeRGB rendered images, large format printers like Chromira etc. have classic ICC colour management module, and all Frontiers have easily accessible mode where you're not limited to sRGB rendered content. Wet mini labs will be replaced by dry (dye based inkjet) in future, with more gamut than Endura/DP2 c-prints and ability to print from any colour space.

Oh, and BTW any 100$ desktop inkjet printer can "handle" AdobeRGB (and any other colour space) renderded images, and usually have quite huge gamut, far beyond sRGB.

So much for sRGB-like devices - sooner or later they'll gonna share same fate as ColorMatch RGB Apple Trinitron CRT displays.
« Last Edit: August 25, 2014, 04:52:11 am by Czornyj »
Logged
Marcin Kałuża | [URL=http://zarzadzaniebarwa

ripgriffith

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 373
    • ripsart.com
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #178 on: August 25, 2014, 05:01:21 am »

"Myself when young did eagerly frequent
 Doctor and Saint, and heard great Argument
 About it and about: but evermore
 Came out by the same Door as in I went."

The Rubiyat of Omar Khayyam, Verse XXVII
Logged

Torbjörn Tapani

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 319
Re: sRGB vs. Adobe RGB: New color management stand up comic
« Reply #179 on: August 25, 2014, 06:25:27 am »

Have I got this right? Gary is going to conduct a test where he shoots the same target with in camera JPEGs set to sRGB and AdobeRGB. Then send them to a pro lab requesting only sRGB images. Knowing they will treat any image as if it were sRGB. Basically assigning it a colorspace much like changing the display as in the video.

So Gary is going to "prove" that if you don't know what you are doing you should just stay with sRGB. That was never contested. But Gary can say that he was right, just look at the dull colors. No one here is going to change their mind, Gary hopes it all blows over and we all just wasted our time.
Logged
Pages: 1 ... 7 8 [9] 10 11 ... 27   Go Up