eeColor processor - ArgyllCMS - Page 11 - AVS Forum
Forum Jump: 
 8Likes
 
Thread Tools
Old 10-30-2014, 05:00 AM - Thread Starter
AVS Special Member
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,099
Mentioned: 49 Post(s)
Tagged: 0 Thread(s)
Quoted: 494 Post(s)
Liked: 464
I hadn't realized @flhoech implemented the CIECAM02 switches in dispcalGUI but it would probably be better if he gave an example since I've only used the command line version. I'm interested in what the smoothing checkbox is doing.

Just a heads up though if you do use this model is that your measurements will no longer line up with Rec709 standards because the model is attempting to match display appearance based on perceptual changes due to environmental conditions.
zoyd is online now  
Sponsored Links
Advertisement
 
Old 10-30-2014, 07:50 AM
Senior Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 295
Mentioned: 12 Post(s)
Tagged: 0 Thread(s)
Quoted: 78 Post(s)
Liked: 53
Quote:
Originally Posted by sillysally View Post
Any tips on how to setup advanced Gamut mapping for my 65VT60 Plasma.
Disable "Enhance effective resolution..." and both "Gamut mapping for...", then enable "Low quality PCS-to-device tables" (this is universally applicable when the profile should be used to create a 3D LUT). Set profile type to "XYZ LUT + swapped matrix".
fhoech is offline  
Old 10-30-2014, 08:14 AM
Senior Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 295
Mentioned: 12 Post(s)
Tagged: 0 Thread(s)
Quoted: 78 Post(s)
Liked: 53
Quote:
Originally Posted by zoyd View Post
I hadn't realized @fhoech implemented the CIECAM02 switches in dispcalGUI but it would probably be better if he gave an example since I've only used the command line version.
The CIECAM02 options are currently only implemented for "normal" display profiles (where the PCS-to-device table is actually used), it won't do anything when the profile is used to create a 3D LUT because the way I'm using collink in dispcalGUI ignores the PCS-to-device tables and inverts the device-to-PCS tables (collink -G). You could specify CIECAM02 options for a 3D LUT via "Set additional commandline arguments" in the "Options" menu though.

Quote:
Originally Posted by zoyd View Post
I'm interested in what the smoothing checkbox is doing.
It applies a very basic linear interpolation filter over the generated colorimetric PCS-to-device table values.

Last edited by fhoech; 10-30-2014 at 08:17 AM.
fhoech is offline  
Old 10-30-2014, 09:03 AM
Advanced Member
 
-Hitman-'s Avatar
 
Join Date: Nov 2004
Location: Manchester, UK
Posts: 699
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 48 Post(s)
Liked: 66
Quote:
Originally Posted by fhoech View Post
Set profile type to "XYZ LUT + swapped matrix".


Could you explain the difference between this ^^^ and the standard XYZ + Matrix, I presume you are recommending swapped matrix because ss is profiling a Plasma display?


Has anyone played with the DR Emphasis as the default is 20%, is this more geared toward LCD/displays with a higher Mll, if using plasma with a very low mll would it be wise to try 10% or even 0% when targeting a Plaw gamma.
-Hitman- is online now  
Old 10-30-2014, 09:12 AM
Senior Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 295
Mentioned: 12 Post(s)
Tagged: 0 Thread(s)
Quoted: 78 Post(s)
Liked: 53
Quote:
Originally Posted by -Hitman- View Post
Could you explain the difference between this ^^^ and the standard XYZ + Matrix, I presume you are recommending swapped matrix because ss is profiling a Plasma display?
There's no point computing an accurate matrix (+ curves) when the profile is (only) going to be used to create a 3D LUT. The only profile element that's relevant in the 3D LUT case is the device-to-PCS table.
If you're going to use the profile in other ways (e.g. installing it so it can be used by ICC color managed applications), then yes, it's important that accurate PCS-to-device tables exist, and may make sense that accurate matrix + curves exist aswell so that applications with flaky ICC color management support can atleast fall back to that in case they can't use the tables.
fhoech is offline  
Old 10-30-2014, 09:22 AM
Advanced Member
 
-Hitman-'s Avatar
 
Join Date: Nov 2004
Location: Manchester, UK
Posts: 699
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 48 Post(s)
Liked: 66
Thanks F, makes sense.


I've been using XYZ +accurate matrix as per the original workflow (Post 2) but will now use swapped matrix as per your suggestion, thanks!
-Hitman- is online now  
Old 11-18-2014, 12:18 PM
AVS Special Member
 
sillysally's Avatar
 
Join Date: Jun 2006
Location: Chicago
Posts: 3,788
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 125 Post(s)
Liked: 272
Is there a new snapshot beta of Argyll CMS for dispcalGUI 2.6????

The last 1.7 beta I have for ArgyllCMS is from May 21st.

Thanks.

ss
sillysally is offline  
Old 11-18-2014, 02:45 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 691
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 105 Post(s)
Liked: 93
Quote:
Originally Posted by sillysally View Post
Is there a new snapshot beta of Argyll CMS for dispcalGUI 2.6????

The last 1.7 beta I have for ArgyllCMS is from May 21st.
Nothing of significance. With the help of some others, I've been working on a new feature, but it isn't finished yet. When it's done, I'll do a V1.7 release.
sillysally likes this.
gwgill is offline  
Old 12-09-2014, 07:33 PM
AVS Special Member
 
sillysally's Avatar
 
Join Date: Jun 2006
Location: Chicago
Posts: 3,788
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 125 Post(s)
Liked: 272
Quote:
Originally Posted by gwgill View Post
Nothing of significance. With the help of some others, I've been working on a new feature, but it isn't finished yet. When it's done, I'll do a V1.7 release.
You posted on Freelist a V1.7 update dated 12/10
Is this a copy and paste update like the beta from May.?
Also a dev_src, Zip. Should this also be installed. ??

Thanks,

ss
sillysally is offline  
Old 12-09-2014, 08:11 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 691
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 105 Post(s)
Liked: 93
Quote:
Originally Posted by sillysally View Post
You posted on Freelist a V1.7 update dated 12/10
Is this a copy and paste update like the beta from May.?
It's an update to address a serious problem with illumread. Unless you are profiling printers using FWA (Fluorescent Whitener Additive) compensation, I doubt it's of any interest.
Quote:
Also a dev_src, Zip. Should this also be installed. ??
Only if you normally compile from the source, and wish to get the fixed illumread.
gwgill is offline  
Old 12-14-2014, 07:03 PM
Advanced Member
 
Iron Mike's Avatar
 
Join Date: Oct 2006
Posts: 799
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 25 Post(s)
Liked: 362
Is is currently possible to import a custom patch set into Argyll/DispCal, like a .csv file etc ?

If so, can you elaborate on the requirements of the syntax and file structure ?


Thanks.

- M

calibration & profiling solutions: Lightspace, Spaceman, Calman, Argyll, ColorNavigator, basICColor
profiling & calibration workflow tools: Display Calibration Tools
meter: CR-250, Klein K-10 A, i1Pro, i1D3
AVS thread: Lightspace & Custom Color Patch Set & Gamma Calibration on Panasonic 65VT60
Iron Mike is offline  
Old 12-17-2014, 01:04 PM
Senior Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 295
Mentioned: 12 Post(s)
Tagged: 0 Thread(s)
Quoted: 78 Post(s)
Liked: 53
The testchart editor of dispcalGUI will have CSV im- and export capability in the next version (drag & drop a CSV file to import, click "Export" and choose CSV format to export). This is already in the latest development snapshot (currently 2.6.0.3), available via 0install or standalone.
DrFaxe likes this.
fhoech is offline  
Old 12-17-2014, 01:58 PM
Advanced Member
 
Iron Mike's Avatar
 
Join Date: Oct 2006
Posts: 799
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 25 Post(s)
Liked: 362
Quote:
Originally Posted by fhoech View Post
The testchart editor of dispcalGUI will have CSV im- and export capability in the next version (drag & drop a CSV file to import, click "Export" and choose CSV format to export). This is already in the latest development snapshot (currently 2.6.0.3), available via 0install or standalone.
thanks Flo, I'll check it out

calibration & profiling solutions: Lightspace, Spaceman, Calman, Argyll, ColorNavigator, basICColor
profiling & calibration workflow tools: Display Calibration Tools
meter: CR-250, Klein K-10 A, i1Pro, i1D3
AVS thread: Lightspace & Custom Color Patch Set & Gamma Calibration on Panasonic 65VT60
Iron Mike is offline  
Old 01-12-2015, 07:12 AM
AVS Addicted Member
 
Join Date: Sep 2002
Location: Wetumpka, AL
Posts: 15,614
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 253 Post(s)
Liked: 228
Does this method of 3D LUTs creation require the use of computer's graphics to output your patterns or can you use an external signal generator such as the Accupel 5000 or Videoforge HDMI?

Samsung 64F8500, Panasonic 65VT50, Oppo 95, Tivo Roamio for OTA, Dish VIP722, Marantz AV8801 preamp, Rotel Amps, Atlantic Tech 8200 speakers, Seaton Submersive HP, Calman 5, Chromapure, Accupel DVG-5000, VideoForge HDMI II, i1Display3pro, i1pro2, eecolor colorbox.
JimP is online now  
Old 01-12-2015, 12:28 PM
AVS Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 2,514
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 165 Post(s)
Liked: 562
Quote:
Originally Posted by JimP View Post
Does this method of 3D LUTs creation require the use of computer's graphics to output your patterns or can you use an external signal generator such as the Accupel 5000 or Videoforge HDMI?
Hi Jim, currently the only option is to use the HDMI Output or your Desktop/Notebook, no hardware pattern generator is supported.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS + CalMAN ColorChecker
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, CalPC, ControlCAL
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is online now  
Old 01-14-2015, 08:05 AM
Member
 
DrFaxe's Avatar
 
Join Date: Nov 2012
Posts: 199
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 76
Quote:
Originally Posted by fhoech View Post
The testchart editor of dispcalGUI will have CSV im- and export capability in the next version (drag & drop a CSV file to import, click "Export" and choose CSV format to export). This is already in the latest development snapshot (currently 2.6.0.3), available via 0install or standalone.
Hi Florian,

I have a feature request for DispcalUI.
The software shows a nice summary of the gamut coverage and gamut volume for sRGB and Adobe RGB after a profile run.
It would be great to have this for Rec.709 as Your software is also heavily used for TV calibration.

And second, a verification tool for a meter correction matrix.
When performing WRGB readings with a spectro and a corrected colorimeter it would be nice to have an overview that shows the x,y,Y of each reading for both meters and dE2000.

Something like the attached excel chart.

Please consider and if if's not bothering too much, implement it. You will make many users very happy

Thanks!!!
Attached Thumbnails
Click image for larger version

Name:	meter.jpg
Views:	16
Size:	98.6 KB
ID:	477954  
DrFaxe is offline  
Old 01-14-2015, 12:30 PM
Senior Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 295
Mentioned: 12 Post(s)
Tagged: 0 Thread(s)
Quoted: 78 Post(s)
Liked: 53
Quote:
Originally Posted by DrFaxe View Post
I have a feature request for DispcalUI.
The software shows a nice summary of the gamut coverage and gamut volume for sRGB and Adobe RGB after a profile run.
It would be great to have this for Rec.709 as Your software is also heavily used for TV calibration.
Gamut volume and coverage of Rec. 709 and sRGB are... exactly the same sRGB uses the Rec. 709 primaries, and the tone response curve doesn't affect the coverage/volume.

Quote:
Originally Posted by DrFaxe View Post
And second, a verification tool for a meter correction matrix.
When performing WRGB readings with a spectro and a corrected colorimeter it would be nice to have an overview that shows the x,y,Y of each reading for both meters and dE2000.

Something like the attached excel chart.
I can do that.
fhoech is offline  
Old 01-15-2015, 03:30 AM
Member
 
DrFaxe's Avatar
 
Join Date: Nov 2012
Posts: 199
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 76
Thanks a lot for the clarification about sRGB <--> Rec709, for taking care about the profile validation and providing such a great software tool like DispcalUI.
DrFaxe is offline  
Old Yesterday, 08:55 AM
Member
 
DrFaxe's Avatar
 
Join Date: Nov 2012
Posts: 199
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 76
Hello,

I have a question about the rendering intent.

What is the difference between "relative colorimetric" (-ir) and "absolut colorimetric with white point scaling" (-ia -w)?

As the documentation of collink says:
The r intent is like Absolute Appearance mode, but maps the white point from source to destination precisely.
and
The -w flag forces the white points to be mapped from source to destination, irrespective of the intent chosen.

So finally both intents look equal to me.
DrFaxe is offline  
Old Yesterday, 11:10 AM
Senior Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 295
Mentioned: 12 Post(s)
Tagged: 0 Thread(s)
Quoted: 78 Post(s)
Liked: 53
Quote:
Originally Posted by DrFaxe View Post
What is the difference between "relative colorimetric" (-ir) and "absolut colorimetric with white point scaling" (-ia -w)?
I think this is best illustrated by an example.

Scenario 1: Let's say your native display whitepoint has a CCT of around 6900K, and you choose not to adjust it and let the 3D LUT all the work. When using relative colorimetric, the end result will be Rec. 709 with the display whitepoint of 6900K. If you use absolute colorimetric with white point scaling on the other hand, you'll get Rec. 709 with the nominal whitepoint of 6500K.

Scenario 2: You have adjusted the display whitepoint so it's 6500K. In that case there'll be no difference (or only very little due to variations in the measurements etc.) between absolute colorimetric and relative colorimetric, because source and destination whitepoint already match.

Personally I would opt for scenario 2 (adjust whitepoint using display controls, then use relative colorimetric).
DrFaxe likes this.
fhoech is offline  
Old Yesterday, 03:13 PM
Member
 
DrFaxe's Avatar
 
Join Date: Nov 2012
Posts: 199
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 76
Thumbs up

Thanks a lot Florian!

assuming that my TV's whitepoint is set to CCT6500 and destination is Rec709 D65, and now I refer to the command line switches.
Were is the difference between this two mapping modes:

1) absolut colorimetric + white point mapping (-ia -w)
2) relative colorimetric (-ir)

Am I right that both will show the same results, but 1) could have a more precise D65 white point as possible meassurement erros will be ironed out by the "-w" switch.
DrFaxe is offline  
Old Yesterday, 03:19 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 691
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 105 Post(s)
Liked: 93
Quote:
Originally Posted by DrFaxe View Post
I have a question about the rendering intent.

What is the difference between "relative colorimetric" (-ir) and "absolute colorimetric with white point scaling" (-ia -w)?
They can be quite different. Relative will shift the whole gamut of the input to match the output white points. Absolute does not attempt to match the white points, while the -w forces a local match of the white points. The latter will locally distort the reproduction to achieve this (that's why it's called a "hack"). [ "locally" as in near the color of white only. ]

If your display white point is close to the input space white point, you won't see much if any difference. If you set your display to (say) 9000K, then you would notice that -ia -w doesn't look right.
DrFaxe likes this.
gwgill is offline  
Old Yesterday, 03:27 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 691
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 105 Post(s)
Liked: 93
Quote:
Originally Posted by DrFaxe View Post
assuming that my TV's whitepoint is set to CCT6500 and destination is Rec709 D65, and now I refer to the command line switches.
Were is the difference between this two mapping modes:

1) absolut colorimetric + white point mapping (-ia -w)
2) relative colorimetric (-ir)

Am I right that both will show the same results, but 1) could have a more precise D65 white point as possible meassurement errors will be ironed out by the "-w" switch.
There is no difference between these two option in regard to what happens with the white point - both assumed that you want input 255,255,255 mapped to 255,255,255, and that (somehow) your display white point is accurate D65. The difference is in what happens with the rest of the gamut. If you use -ir then absolute accuracy of the rest of the gamut will also depend on how accurately you have manual matched your TV white to D65. Using "-ia -w" the overall absolute accuracy will not depend on your TV white to D65 matching (it depends on the accuracy of the profile of course).

In any event the absolute accuracy is not that important - we strongly adapt to the white point, whatever it is. It is the relative (to white) accuracy of the other colors that counts far more.

Note that -w always risks some distortion near white, depending on how much the white point if shifted. If the shift is small, then the distortion will be invisible.
DrFaxe likes this.

Last edited by gwgill; Yesterday at 03:45 PM.
gwgill is offline  
Old Yesterday, 03:28 PM
Senior Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 295
Mentioned: 12 Post(s)
Tagged: 0 Thread(s)
Quoted: 78 Post(s)
Liked: 53
Quote:
Originally Posted by DrFaxe View Post
(-ia -w)
Oops, I read -ia -w as -iaw. What I said above applies to the -iaw case only. Graeme's answer is of course the correct one
fhoech is offline  
Old Yesterday, 03:48 PM
Member
 
DrFaxe's Avatar
 
Join Date: Nov 2012
Posts: 199
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 76
Thanks to all of You. Now I understand everything.

Personally, after reading Your explanation, relative colorimetric seems to be the best option and I'm wondering why all the guides here recommend to use absolut with white point scaling.
Actually I use the -ia -w option and it doesn't look wrong to me nor I see any distortions near white, the profile CCT is 6537K .
Anyway, I have the best looking picture on my TV I can imagine and I'm really very happy You provide such a great software and support, thanks a lot for this.

@Florian
I have to admit that my first question was formulated wrong because I thought -ia -w command line is "absolute with white point scaling" in DispCalUI, which is obviously not.
Sorry for the confusion I have caused.

Last edited by DrFaxe; Today at 02:56 AM.
DrFaxe is offline  
 
Thread Tools


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off