MadVR - ArgyllCMS - Page 61 - AVS Forum
Forum Jump: 
 59Likes
Reply
 
Thread Tools
post #1801 of 2795 Old 05-20-2014, 06:12 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 656
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 90 Post(s)
Liked: 84
Quote:
Originally Posted by VerGreeneyes View Post

Graeme, speaking of which, what are your personal thoughts on compression and expansion as opposed to matching (and clipping)? It seems to me that mapping a smaller gamut to the appropriate subset of a wider gamut would be the most technically accurate, yet you'd be losing the most vividly saturated colors that the display is capable of (whereas with compression, I suppose you avoid losing saturation detail if the target gamut is smaller in some way).
Some of the wide gamut TV vendors have attempted such schemes. Being something of a purist, I tend to regard this idea as a gimmick - good for games perhaps, but nothing to do with reproducing what was intended.

The ArgyllCMS gamut mapping code was developed for a much more difficult situation than slight mismatches between RGB gamuts - matching RGB gamuts to CMYK printing gamuts. Not only are these a different size (RGB often being noticeably bigger in many color areas), but quite different shapes. They are opposites - where RGB is strong, CMY is weak. Where CMY is strong, RGB is weak.

So such technology would be very useful if we were in the fortunate situation of having a lot of wide gamut video sources, and trying to get the best reproduction possible on more limited displays. Instead, the situation is the reverse - we have wide gamut displays available, but little or no source material that exploits it.

This is a standardization and technology issue. In principle the technical solution is to have the source define it's gamut, and for the display system to then automatically (perhaps with some guidance from the creator and the viewer) create a gamut mapping to display it. Theoretically this could be dynamic - different scenes could be mapped optimally.

From a practical point of view there are four obstacles:
  • Creating an agreed industry wide standard capable of providing the needed technical information.
  • Having implementations widely available that fully take advantage of the standard. Not only must practical gamut mapping technology be accessible to the implementers, it has to be accepted as providing a good result for most situations. High quality gamut mapping technology tends to be both propriety and a little slow.
  • Having the production technology and creative understanding of what the standard provides, so as to be able to provide widely available productions that take advantage of the standard.
  • Be backwards compatible with existing production workflows and displays.

There have been some attempts to tackle this very problem in the newer video standards. From a color science and color management point of view, I've got to say that I think these attempts are pretty woeful. I understand why this is so (see the list above).

To be an effective standard, I think someone would have to create both the standard and a usable sample implementation to prove it works, and so that all the TV vendors could simply drop it into place. That would take money, expertise, co-operation and time.
Quote:
I can't honestly tell what I find more pleasant in general, so this is a bit pedantic, but I'm wondering how you'd decide which to use in various situations.
I can't really answer that, since at the end of the day I'd make a subjective judgement based on how it looks. Checking the degree of mismatch by examining the 3D gamut hulls would also be a guide though.
gwgill is offline  
Sponsored Links
Advertisement
 
post #1802 of 2795 Old 05-20-2014, 06:18 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 656
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 90 Post(s)
Liked: 84
Quote:
Originally Posted by Tribalate View Post

If the lighting in my movie room stays the same, is there any advantage to measure the ambient lighting with i1Display Pro and use it with the BT1886 curve?
The best thing is to experiment and see what you think.
Quote:

There were some stuff about doing in the argyllcms online documentation, but would it be possible to also use dispcalGUI for it? The method suggested in the documentation would be, if I understood correctly, to use the same calibration chain except in collink.

For example in the documentation:
you could do it using pure CIECAM02 adjustment and a black point mapping:

collink -v -ctv -dmd -da:1 -G -ila Rec709.icm TV.icm HD.icm # very dark conditions

using both to model a reference video display system that is adapted to your viewing conditions:

collink -v -Ib -c md -dmd -da:5 -G -ila Rec709.icm TV.icm HD.icm # very dark conditions
There are a couple of mechanisms available. One is to change the parameters to the viewing conditions model, if it is being used.
The other is to alter the power used with the BT.1886 curve (collink -I b:x.x or -I B:x.x where x.x is the power), if it is being used. You'd use a larger power for a darker viewing environment.
gwgill is offline  
post #1803 of 2795 Old 05-20-2014, 06:27 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 656
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 90 Post(s)
Liked: 84
Quote:
Originally Posted by JimP View Post

I finally got the programs loaded per the first post in this thread.
Original question was how do you profile a jeti 1211 to a i1d3.
Where do you run this from?
I don't really understand what you're asking. If you are using the command line directly, you run it from the command line using the appropriate option flags and arguments. If you are using DispcalGUI, use "Tools->Create Colorimeter Correction..."
gwgill is offline  
post #1804 of 2795 Old 05-21-2014, 02:26 AM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 656
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 90 Post(s)
Liked: 84
I'm please to make available a beta test version of ArgyllCMS tools that supports the Klein K10-A Colorimeter.

MSWin 32 bit executables are here,
MSWin 64 bit executables are here.

They contain the following:

spotread.exe
dispcal.exe
dispread.exe

which add K10-A support, as well as

ccxxmake.exe

which deprecates the -T option, and now mandates the new -t parameter.

Also included is

targen.exe

which now orders the patches using an updated display timing model.

I'd imagine that these can be used to overwrite the V1.6.3 executables and should then work with DispcalGUI, except for ccxxmake, due to the parameter change.
gwgill is offline  
post #1805 of 2795 Old 05-21-2014, 05:30 AM
Senior Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 234
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 56 Post(s)
Liked: 46
Quote:
Originally Posted by gwgill View Post

I'm please to make available a beta test version of ArgyllCMS tools that supports the Klein K10-A Colorimeter.
Exciting smile.gif
I'm updating dispcalGUI for the ccxxmake changes (about ready) and K10 support. Some questions if I may:
- Does the K10 support CCSS files?
- Looking at the code, the list of display types available is dynamically populated from the 96 programmable memories of the K10, so I'm going to parse the `dispcal -?`output (as I don't have a K10, it would help me if you or someone else could kindly send me the output of dispcal -? with the K10 connected). Is there a base set of display types that is always available? (reason I'm asking is because I'm wondering if the list should always contain LCD and refresh modes, and what the appropriate command line -y options for these would be if that's the case e.g. would "n" and "r" be appropriate?)

Thanks!
fhoech is online now  
post #1806 of 2795 Old 05-21-2014, 05:51 AM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 656
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 90 Post(s)
Liked: 84
Quote:
Originally Posted by fhoech View Post

- Does the K10 support CCSS files?
No, it doesn't have spectral sensitivity information, instead relying on calibration matricies.
Quote:
- Looking at the code, the list of display types available is dynamically populated from the 96 programmable memories of the K10, so I'm going to parse the `dispcal -?`output (as I don't have a K10, it would help me if you or someone else could kindly send me the output of dispcal -? with the K10 connected).
I'll email you this.

[ At some stage I'd like to add an API to add/change/delete calibrations in instruments like the K10-A and colorhug, so that I can then write a utility to make such management accessible. ]
Quote:
Is there a base set of display types that is always available? (reason I'm asking is because I'm wondering if the list should always contain LCD and refresh modes, and what the appropriate command line -y options for these would be if that's the case e.g. would "n" and "r" be appropriate?)
Hmm. Not really. The -y option is populated from a list that is: hard coded in the driver for anything native + any matrices that are stored in the instrument + installed ccss files if the instrument can use them + installed ccmx files if they match that instrument type.

Something like the DTP92 only has a CRT mode, while the Smile only has LCD modes. In others there's not a combined refresh/non-refresh + calibration setting, it's native refresh/non-refresh + calibration on the top of that.
gwgill is offline  
post #1807 of 2795 Old 05-21-2014, 10:16 AM
Senior Member
 
anta1974's Avatar
 
Join Date: Jun 2009
Location: Ukraine
Posts: 310
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 20
N3W813!
Help me please!

What needs to change in your instructions to calibrate gamma=2.35, but BT.1886.
What settings should be output ATI driver- YCbCr 4:2:2 or RGB limited or RGB full?
What settings should be output madvr render and GPU- 16-235 or 0-255 ?
anta1974 is offline  
post #1808 of 2795 Old 05-21-2014, 10:37 AM
Senior Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 234
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 56 Post(s)
Liked: 46
Quote:
Originally Posted by gwgill View Post

No, it doesn't have spectral sensitivity information, instead relying on calibration matricies.
I'll email you this.

Thanks!
Quote:
Originally Posted by gwgill View Post

[ At some stage I'd like to add an API to add/change/delete calibrations in instruments like the K10-A and colorhug, so that I can then write a utility to make such management accessible. ]

Sounds good.
Quote:
Originally Posted by gwgill View Post

Quote:
Originally Posted by fhoech 
Is there a base set of display types that is always available? [...]
Hmm. Not really. The -y option is populated from a list that is: hard coded in the driver for anything native + any matrices that are stored in the instrument + installed ccss files if the instrument can use them + installed ccmx files if they match that instrument type.

Something like the DTP92 only has a CRT mode, while the Smile only has LCD modes. In others there's not a combined refresh/non-refresh + calibration setting, it's native refresh/non-refresh + calibration on the top of that.
Makes sense. So far I've also hardcoded the native type selections per instrument (with the exception of the newly added K10 support).
fhoech is online now  
post #1809 of 2795 Old 05-21-2014, 10:39 AM
Senior Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 234
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 56 Post(s)
Liked: 46
Quote:
Originally Posted by anta1974 View Post

What needs to change in your instructions to calibrate gamma=2.35, but BT.1886.
You need to change two settings: Under calibration, select tone curve Rec. 1886 and then simply change the gamma value to 2.35 (assuming you're using dispcalGUI 2.0 or later). In the 3D LUT creation window, do the equivalent.
anta1974 likes this.
fhoech is online now  
post #1810 of 2795 Old 05-21-2014, 10:42 AM
Senior Member
 
anta1974's Avatar
 
Join Date: Jun 2009
Location: Ukraine
Posts: 310
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 20
But I want to have a linear gamma=2.35 - power formula, but no BT.1886!
anta1974 is offline  
post #1811 of 2795 Old 05-21-2014, 10:42 AM
Senior Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 234
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 56 Post(s)
Liked: 46
I've made available a new development snapshot of dispcalGUI (2.1.0.1) which is compatible with the Argyll CMS beta (Klein K10-A support and ccxxmake parameter change).
anta1974 likes this.
fhoech is online now  
post #1812 of 2795 Old 05-21-2014, 10:50 AM
Senior Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 234
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 56 Post(s)
Liked: 46
Quote:
Originally Posted by anta1974 View Post

But I want to have a linear gamma=2.35 - power formula, but no BT.1886!
Well that's something different than you original question then smile.gif
Under calibration tone curve, select gamma and then set it to 2.35. Enable black point compensation, then use the "Rec709_Gamma235.icm" as source in the 3D LUT creation window (and disable the BT.1886 checkbox). Note that this approach may have some problems.
anta1974 likes this.
fhoech is online now  
post #1813 of 2795 Old 05-21-2014, 10:54 AM
Senior Member
 
anta1974's Avatar
 
Join Date: Jun 2009
Location: Ukraine
Posts: 310
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 20
I do not need compensation black.
Why do I need to enable it?
anta1974 is offline  
post #1814 of 2795 Old 05-21-2014, 11:37 AM
Senior Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 234
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 56 Post(s)
Liked: 46
Quote:
Originally Posted by anta1974 View Post

I do not need compensation black.
Why do I need to enable it?
You're right in that you do not need BPC specifically. What you do need though unless you want to risk crushing your video blacks is a way to map the black point, and as you don't want to use BT.1886 for this, you have basically two options:

- Use black point compensation (BPC)
- Or use a non-colorimetric rendering intent (e.g. luminance matched appearance or perceptual)
anta1974 likes this.
fhoech is online now  
post #1815 of 2795 Old 05-21-2014, 12:04 PM
Advanced Member
 
3ll3d00d's Avatar
 
Join Date: Sep 2007
Location: London, UK
Posts: 572
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 236 Post(s)
Liked: 79

I am attempting to create a 3d lut on a Windows 8.1 machine. I have tried using madTPG from madvr 0.87.9 (as shipped with jriver) and madvr 0.87.10 (downloaded manually) & dispcalGUI 2.0 and 2.1. I've tried running as admin or running normally. In any case, I get the "new_disprd() failed with 'Window Access Failed' error. madTPG is still open & all apps are configured as per the 1st post in this thread.

 

Any ideas on what I'm doing wrong here?

 

--------------------------------------------------------------------------------
Calibrate & profile
--------------------------------------------------------------------------------
Session log: madVR 2014-05-21 0.3127x 0.329y Rec. 1886 S XYZLUT.log

Working directory:
  c:\
   users\
    matt\
     appdata\
      local\
       temp\
        dispcalGUI-nvg5pu\

Command line:
  dispcal.exe
    -v2
    -dmadvr
    -c1
    -yn
    -qh
    -m
    "-w0.3127,0.329"
    -G2.4
    -f0
    -k0
    -A4.0
    "madVR 2014-05-21 0.3127x 0.329y Rec. 1886 S XYZLUT"

dispcalGUI: Starting interaction with subprocess
Setting up the instrument
Product Name:      i1Display3
Serial Number:     I1-13.A-02
Firmware Version:  v1.03
Firmware Date:     05Jun12
dispcal: Error - new_disprd() failed with 'Window Access Failed'

dispcalGUI: Reached EOF (OK)
dispcalGUI: Checking subprocess status
dispcalGUI: Subprocess no longer alive (OK)
...aborted.
Calibration has not been finished.
3ll3d00d is online now  
post #1816 of 2795 Old 05-21-2014, 03:30 PM
Advanced Member
 
3ll3d00d's Avatar
 
Join Date: Sep 2007
Location: London, UK
Posts: 572
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 236 Post(s)
Liked: 79
Quote:
Originally Posted by 3ll3d00d View Post
 

I am attempting to create a 3d lut on a Windows 8.1 machine. I have tried using madTPG from madvr 0.87.9 (as shipped with jriver) and madvr 0.87.10 (downloaded manually) & dispcalGUI 2.0 and 2.1. I've tried running as admin or running normally. In any case, I get the "new_disprd() failed with 'Window Access Failed' error. madTPG is still open & all apps are configured as per the 1st post in this thread.

 

Any ideas on what I'm doing wrong here?

It seems I needed to run the madvr installfilter script, all was well after I ran this. I tested on a win7 box and it worked without running that so not sure what is going on there. 

3ll3d00d is online now  
post #1817 of 2795 Old 05-22-2014, 02:33 AM
Advanced Member
 
3ll3d00d's Avatar
 
Join Date: Sep 2007
Location: London, UK
Posts: 572
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 236 Post(s)
Liked: 79

I have hit another problem. I left it measuring and it failed with the following error. The machine has 8G of RAM and is not loaded.

 

***************************************************************************************************************************************************************************************************************
Doing White point fine tune:
Before fine tune, rel WP = XYZ 0.964250 0.999969 0.825030, Lab 99.998784 0.013821 -0.012611
After fine tune, rel WP = XYZ 0.964200 1.000000 0.824900, Lab 99.999999 0.000004 -0.000002
                 abs WP = XYZ 0.896199 0.995327 1.121212, Lab 99.819018 -11.261840 -21.855425
Creating fast inverse input lookups
White point XYZ = 0.896199 0.995327 1.121212, Lab = 99.819018 -11.261840 -21.855425
Find black point
Black point XYZ = 0.926488 1.019834 1.140341, Lab = 100.761911 -9.889692 -21.482569
Display Luminance = 45.269967
Done A to B table creation
Setting up B to A table lookup
Creating B to A tables
 0%
Rev cache RAM = 1610 Mbytes
C:\Program Files (x86)\Argyll_V1.6.3\bin\colprof.exe: Warning - rev: bwd vertex 0 is not prime or secondary (vflag = 0)(Check that your measurement data is sane!)
C:\Program Files (x86)\Argyll_V1.6.3\bin\colprof.exe: Error - rev_reduce_cache: run out of rev  virtual memory!
dispcalGUI: Reached EOF (OK)
dispcalGUI: Checking subprocess status
dispcalGUI: Subprocess no longer alive (OK)
...aborted.
3ll3d00d is online now  
post #1818 of 2795 Old 05-22-2014, 04:57 AM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 656
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 90 Post(s)
Liked: 84
Quote:
Originally Posted by 3ll3d00d View Post

White point XYZ = 0.896199 0.995327 1.121212, Lab = 99.819018 -11.261840 -21.855425
Black point XYZ = 0.926488 1.019834 1.140341, Lab = 100.761911 -9.889692 -21.482569
Your data isn't sane. You're black is lighter than your white!

Something went wrong with the measurement process.

Try creating and measuring a small chart (say 50 colors), and check that the process looks right (patches get displayed), and that the resulting .ti3 file has reasonable values.

and/or

Use spotread to measure some patches manually (ie. created using dispwin -m), and check that you get reasonable values.
gwgill is offline  
post #1819 of 2795 Old 05-22-2014, 05:22 AM
Advanced Member
 
3ll3d00d's Avatar
 
Join Date: Sep 2007
Location: London, UK
Posts: 572
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 236 Post(s)
Liked: 79
Quote:
Originally Posted by gwgill View Post

Something went wrong with the measurement process.

Try creating and measuring a small chart (say 50 colors), and check that the process looks right (patches get displayed), and that the resulting .ti3 file has reasonable values.

and/or

Use spotread to measure some patches manually (ie. created using dispwin -m), and check that you get reasonable values.
When you say "check the process like right", what do you mean exactly?

When I set it going then I see it saying something about argyllcms patches and then it starts cycling through different colours. I leave the room at that point as I assume it is doing its thing.
3ll3d00d is online now  
post #1820 of 2795 Old 05-22-2014, 06:41 AM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 656
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 90 Post(s)
Liked: 84
Quote:
Originally Posted by 3ll3d00d View Post

When you say "check the process look right", what do you mean exactly?
You need to check that the measured value of the patches seems reasonable, and if not, figure out where in the process this is going wrong.

If you haven't played with a instrument and measured a few display colors manually (ie. white, 50% grey, black) and seen what typical XYZ values are for these, then now is the time.
gwgill is offline  
post #1821 of 2795 Old 05-22-2014, 06:57 AM
Advanced Member
 
3ll3d00d's Avatar
 
Join Date: Sep 2007
Location: London, UK
Posts: 572
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 236 Post(s)
Liked: 79
Quote:
Originally Posted by gwgill View Post


You need to check that the measured value of the patches seems reasonable, and if not, figure out where in the process this is going wrong.

If you haven't played with a instrument and measured a few display colors manually (ie. white, 50% grey, black) and seen what typical XYZ values are for these, then now is the time.

the projector was calibrated by a pro ~18 months ago. I measured it yesterday with an i1pro3 in chromapure and it's drifted a bit; red is particularly low, it is a bit dimmer. I didn't drill into it as nothing was obviously out of the ordinary. I will look into it in more detail though.

 

EDIT: I took the easy road and just reran the process to check the problem is repeatable, it ran to completion without issues :confused: I suppose I'll just see if it recurs in future.

3ll3d00d is online now  
post #1822 of 2795 Old 05-22-2014, 08:56 AM
Senior Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 234
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 56 Post(s)
Liked: 46
dispcalGUI also has a function to check existing and new measurements (in the "Tools" menu) that may help in figuring out what went wrong in the process. It's not perfect meaning it assumes an sRGB-like display and may report false positives, but gross errors like the black measured as white will stand out. It'll list all patches that it considers faulty and should make it relatively easy to spot obvious errors by looking at visual differences between the expected color (left side) and what was measured (right side).
fhoech is online now  
post #1823 of 2795 Old 05-22-2014, 11:27 AM
Senior Member
 
anta1974's Avatar
 
Join Date: Jun 2009
Location: Ukraine
Posts: 310
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 20
why I have no choice madVR in 2.Select 'madVR' for Display device?

there is only
1) Primary
2) Web@localhost
3) Untethered
anta1974 is offline  
post #1824 of 2795 Old 05-22-2014, 12:15 PM
Senior Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 234
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 56 Post(s)
Liked: 46
Quote:
Originally Posted by anta1974 View Post

why I have no choice madVR in 2.Select 'madVR' for Display device?
Which version of Argyll CMS respective dispcalGUI are you using? (madVR was added in Argyll 1.6 and dispcalGUI 1.5.2.5 and later)
fhoech is online now  
post #1825 of 2795 Old 05-22-2014, 12:17 PM
Senior Member
 
anta1974's Avatar
 
Join Date: Jun 2009
Location: Ukraine
Posts: 310
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 20
argyll 1.6.3 and dispcalgui 2.1
anta1974 is offline  
post #1826 of 2795 Old 05-22-2014, 12:20 PM
Senior Member
 
anta1974's Avatar
 
Join Date: Jun 2009
Location: Ukraine
Posts: 310
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 20
THANK YOU VERY MUCH!!!
it turned out that was the old way argyll 1.4.
anta1974 is offline  
post #1827 of 2795 Old 05-22-2014, 08:52 PM
Member
 
umeng2002's Avatar
 
Join Date: Aug 2006
Posts: 69
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 19
When I go to make the 3dLUT, I select the rec709 source profile and uncheck BT.1886 mapping since I used 2.2 with all black level offset for the calibration; but when it goes to generate the LUT, the BT. 1886 box magically checks itself on and applies it to the 3dLUT. What's going on?

Should I have selected the rec709_Gamma22 source profile instead?
umeng2002 is offline  
post #1828 of 2795 Old 05-22-2014, 09:14 PM
Senior Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 234
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 56 Post(s)
Liked: 46
Quote:
Originally Posted by umeng2002 View Post

When I go to make the 3dLUT, I select the rec709 source profile and uncheck BT.1886 mapping since I used 2.2 with all black level offset for the calibration; but when it goes to generate the LUT, the BT. 1886 box magically checks itself on and applies it to the 3dLUT. What's going on?
That's not intended, but might actually be the right thing to do - I don't think anybody would ever want the unaltered Rec. 709 TRC, it will produce a "washed out" result when used with typical video material in dim/dark viewing environment. [ I'll still fix it though, the way it's now is not intended and confusing ]
Quote:
Originally Posted by umeng2002 View Post

Should I have selected the rec709_Gamma22 source profile instead?
Yes, in your case that should get you what you're after.
anta1974 likes this.
fhoech is online now  
post #1829 of 2795 Old 05-23-2014, 02:02 AM
Advanced Member
 
3ll3d00d's Avatar
 
Join Date: Sep 2007
Location: London, UK
Posts: 572
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 236 Post(s)
Liked: 79

I thought I would just post back to say thanks to the authors.

 

After getting past the initial setup problems, I generated a lut, installed it and measured the results (dE down from 7 to 2 on the greyscale, various colour errors typically have dE <= 1). A serious improvement in accuracy on my setup (jvc x3) for nothing more than clicking a few buttons, it's really v slick now so great work :)

 

In terms of tweaking the process to my specific setup, where would one go from here? is there a good summary of how one might finetune this posted somewhere?

3ll3d00d is online now  
post #1830 of 2795 Old 05-23-2014, 05:08 PM
Member
 
VerGreeneyes's Avatar
 
Join Date: May 2013
Posts: 112
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Thanks for the detailed reply, Graeme. I'll admit it wasn't the answer I was hoping for, but it makes sense.
VerGreeneyes is offline  
Reply Display Calibration

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


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