eeColor processor - ArgyllCMS - Page 22 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 73Likes
Reply
 
Thread Tools
post #631 of 669 Old 04-18-2017, 01:45 PM
AVS Forum Special Member
 
rickardl's Avatar
 
Join Date: Nov 2003
Location: Stockholm, Sweden
Posts: 2,144
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 123 Post(s)
Liked: 95
Quote:
Originally Posted by RIppolito View Post
Why do you want to use madtpg vs the internal pattern generator? Do you think you'll get better results?
Ron
For my specific Samsung Plasma, I got recommendations to use APL patterns, not sure how to do that with
the internal one, and MadTPG seems capable to do it and it was also highly recommended to me because of it's dithering.


Edit: looking at the documentation, it looks like if I set "Display" to "madVR" instead of my HDMI-connected plasma, it would use MadTPG instead of the internal generator, is that correct?
RIppolito likes this.

Last edited by rickardl; 04-18-2017 at 01:59 PM.
rickardl is offline  
Sponsored Links
Advertisement
 
post #632 of 669 Old 04-18-2017, 02:17 PM
Senior Member
 
DrFaxe's Avatar
 
Join Date: Nov 2012
Posts: 220
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 24 Post(s)
Liked: 91
Yes,thats correct. Choose madvr as display if you want it as a pattern generator. But I would not recommend apl pattern for Plasmas, just use window pattern with 10-12% size on black background. I prefer 11% and verify the result with TEDs disc. He uses also 11%.
DrFaxe is offline  
post #633 of 669 Old 04-25-2017, 12:02 AM
AVS Forum Special Member
 
rickardl's Avatar
 
Join Date: Nov 2003
Location: Stockholm, Sweden
Posts: 2,144
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 123 Post(s)
Liked: 95
got the eecolor box yesterday and I have made some
attempts but it looks kind of crazy... overblown whites
which look reddish for LUT no 1 which I have uploaded.
Note: The other pre-installed LUTs looks ok, as does ee Off.
I have put the box between my oppo bdp-103D and plasma
and I have connected my laptop to the HDMI input of the oppo.
Oppo identifies both the input from PC where DisplayCal is running on my extended screen
and the output to the box as 1080p@60Hz 4:4:4 24b.
The pic shows Apply Calibration (vcgt) ticked but tried with it off as well.
Similar results with MadTPG and the internal generator.
The verification in DisplayCal shows reasonable values for ee Off
so the colorimeter seems to be reading ok.
Any ideas?
Attached Thumbnails
Click image for larger version

Name:	dp_set1.jpg
Views:	46
Size:	229.0 KB
ID:	2101345   Click image for larger version

Name:	dp_set2.jpg
Views:	37
Size:	198.3 KB
ID:	2101353   Click image for larger version

Name:	dp_set3.jpg
Views:	37
Size:	265.9 KB
ID:	2101361   Click image for larger version

Name:	dp_set4.jpg
Views:	42
Size:	183.5 KB
ID:	2101369  
Attached Files
File Type: txt profile information.txt (5.3 KB, 9 views)
rickardl is offline  
 
post #634 of 669 Old 04-25-2017, 01:46 AM
Advanced Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 866
Mentioned: 39 Post(s)
Tagged: 0 Thread(s)
Quoted: 432 Post(s)
Liked: 283
You've probably set the wrong encoding for the 3D LUT. It should be "TV RGB 16-235".
rickardl and ConnecTEDDD like this.

DisplayCAL - Graphical front-end for Argyll CMS display calibration and characterization
Current stable version 3.3.4 released 2017-09-09
Previous development snapshot (OUTDATED): 3.2.9 Beta (Windows/0install) released 2017-05-23 | Standalone | Changelog
DisplayCAL on Facebook
fhoech is offline  
post #635 of 669 Old 04-25-2017, 04:17 AM
AVS Forum Special Member
 
rickardl's Avatar
 
Join Date: Nov 2003
Location: Stockholm, Sweden
Posts: 2,144
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 123 Post(s)
Liked: 95
Quote:
Originally Posted by fhoech View Post
You've probably set the wrong encoding for the 3D LUT. It should be "TV RGB 16-235".
Yes, I had...! Now it looks better! No more discoloring.
But the lower levels seems raised vs the standard LUTs and Off-mode.
What about Apply Calibration (vcgt), should that be ticked or unticked?
rickardl is offline  
post #636 of 669 Old 04-25-2017, 05:54 AM
AVS Forum Special Member
 
rickardl's Avatar
 
Join Date: Nov 2003
Location: Stockholm, Sweden
Posts: 2,144
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 123 Post(s)
Liked: 95
Did another quick one, now with Rec 1886, the black level is indeed raised, colors are ok but a bit washed out.
I guess these tone response curves also says the lower levels are raised?


Edit: I think I got it now... check pic #2 ,
Attached Thumbnails
Click image for larger version

Name:	tone response.jpg
Views:	37
Size:	89.4 KB
ID:	2101529   Click image for larger version

Name:	tone response2.jpg
Views:	34
Size:	128.2 KB
ID:	2101561  

Last edited by rickardl; 04-25-2017 at 07:05 AM.
rickardl is offline  
post #637 of 669 Old 04-26-2017, 03:44 AM
AVS Forum Special Member
 
rickardl's Avatar
 
Join Date: Nov 2003
Location: Stockholm, Sweden
Posts: 2,144
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 123 Post(s)
Liked: 95
Will do longer sessions soon with DisplayCAL but I have some questions:

1. I have a Samsung PS64F8500 plasma and I am using the Native colorspace but will there be any benefit from increasing the color setting from 50 to, let's say, 60 to give DisplayCAL/ArgyllCMS "more" to work with?

2. When using i1 Dsplay pro (i1D3) with HCFR, zoyd suggests an integration time of 0.5 sec as the best option. Is there any point in overriding the min display update delay/settle time multiplier in DisplayCAL for getting more accurate readings or is this just waste of "time"?
rickardl is offline  
post #638 of 669 Old 04-26-2017, 04:05 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,537
Mentioned: 63 Post(s)
Tagged: 1 Thread(s)
Quoted: 1732 Post(s)
Liked: 2131
Quote:
Originally Posted by rickardl View Post
Will do longer sessions soon with DisplayCAL but I have some questions:

1. I have a Samsung PS64F8500 plasma and I am using the Native colorspace but will there be any benefit from increasing the color setting from 50 to, let's say, 60 to give DisplayCAL/ArgyllCMS "more" to work with?

2. When using i1 Dsplay pro (i1D3) with HCFR, zoyd suggests an integration time of 0.5 sec as the best option. Is there any point in overriding the min display update delay/settle time multiplier in DisplayCAL for getting more accurate readings or is this just waste of "time"?
Hi,

1) If you increase so much the color then you will have color luminance clipping which is not good for a display performance before the starting the display characterization.

Here are some images I designed to showcase what you are checking when are you looking to some basic patterns all of us are using for pre/post calibration verification, with an RGB Cube Space presentation.

For example, for pre-calibration we are checking using the Contrast Flashing Bars Pattern:



..but it helps if you check additionally the RGBCMY to prevent clipping there also:



...and after the contrast check, you move to color luminace clipping test, using the 7-Color Clipping Bars Pattern to prevent clipping of each color channel (WRGBCMY), this sometimes can be fixed be removing some additionally clicks from the contrast control also (or color slider), you are checking these areas:



2) About meter intergration time, this requires some testing to see which intergration time in seconds provide more stable/repeatable readings.

Use a low Luminace patches , for example 5% Gray / 5% Red / 5% Green / 5% Blue and take 4-5 measurements in continue mode to see your meter repeatability in xyY. A lot of times low intergration times can provide poor repeatability or failed readings.

Test various time settings to see which one is better to use; 0.5 , 0.75 , 1 sec etc..
rickardl and anta1974 like this.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is offline  
post #639 of 669 Old 04-26-2017, 04:35 AM
Senior Member
 
DrFaxe's Avatar
 
Join Date: Nov 2012
Posts: 220
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 24 Post(s)
Liked: 91
Hi,
besides Ted`s post, I'd like to add some more suggestions.
Looking at your setup, I would not include the Oppo in the chain. Just use Laptop --> eecolor --> TV. Use extend display in your graphic card settings. Open the madLevelsTweaker of madvr and drag the window to the TV. Activate "Force PC" levels and click apply. Then reboot the laptop and check the display settings in the madvr TPG. You will see a device "SII Repeater". Make sure you select "Digital Monitor / TV" and "TV Levels (16-235) in the settings. Now Your levels should be correct.
Native gamut setting for the TV is fine. Just make sure you have enough color luminance but avoid clipping as TED said.

Regarding the meter integration time, argyllcms works with adaptive integration time. This will provide accurate and fast readings, I would not recommend to change this. Otherwise you will loose accurary or speed, depending on the setting.

Regarding the patch settle time, argyll usualy set the optimized time right at the beginning by displaying a short patch sequence. This always worked fine for my Panasonic Plasma, I had something like 270-280ms.
What timing do you see with your TV?

S/W: LightSpace HCC, Calman 5, DisplayCal/ArgyllCMS
H/W: CR-100, I1Pro2, I1D3, eeColor, RB8 Generator
Tools: ControlCal, Ted's LightSpace CMS Calibration Disk
DrFaxe is offline  
post #640 of 669 Old 04-26-2017, 04:36 AM
AVS Forum Special Member
 
rickardl's Avatar
 
Join Date: Nov 2003
Location: Stockholm, Sweden
Posts: 2,144
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 123 Post(s)
Liked: 95
I used 56 for the color setting on the display when I ran DisplayCAL yesterday and when I checked the contrast clipping and color clipping patterns on the Spears & Munsil v2 disk, it showed no clipping with the LUT enabled but showed color clipping with it disabled. So even though it clipped in the intial setting, the resulting LUT was good with no clipping and that made me wonder if I should increase it even more to cover more of the gamut to start with.

DisplayCAL (argyllcms I guess) seems to measure integration time since it writes something similar to this: “Measured display update delay of 51 msec, using delay of 168 msec_0 msec inst reaction” and I am just wondering if it is any point of using a larger minimum value or is it safe to go with the measured value from DisplayCAL.
rickardl is offline  
post #641 of 669 Old 04-26-2017, 04:40 AM
AVS Forum Special Member
 
rickardl's Avatar
 
Join Date: Nov 2003
Location: Stockholm, Sweden
Posts: 2,144
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 123 Post(s)
Liked: 95
Quote:
Originally Posted by DrFaxe View Post
Looking at your setup, I would not include the Oppo in the chain. Just use Laptop --> eecolor --> TV. Use extend display in your graphic card settings. Open the madLevelsTweaker of madvr and drag the window to the TV. Activate "Force PC" levels and click apply. Then reboot the laptop and check the display settings in the madvr TPG. You will see a device "SII Repeater". Make sure you select "Digital Monitor / TV" and "TV Levels (16-235) in the settings. Now Your levels should be correct.
I actually got spot on levels now with the Oppo in the chain so I will keep it there for now.

Quote:
Originally Posted by DrFaxe View Post
Regarding the patch settle time, argyll usualy set the optimized time right at the beginning by displaying a short patch sequence. This always worked fine for my Panasonic Plasma, I had something like 270-280ms.
What timing do you see with your TV?
I can't recall, I will have to check if I got some logs left. The times I posted above are not from my setup.
rickardl is offline  
post #642 of 669 Old 04-26-2017, 04:57 AM
Senior Member
 
DrFaxe's Avatar
 
Join Date: Nov 2012
Posts: 220
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 24 Post(s)
Liked: 91
Your 168ms patch settle time seems to be OK, as Samsung plasmas are more prone to image retention and afterglow then Panasonic.

But if You like to test other settle times, open the command line window and go into the argyllcms\bin directory and
type "set ARGYLL_MIN_DISPLAY_UPDATE_DELAY_MS=700" (without the ") if you want a delay of e.g. 700ms.
if you want to go back to automatic mode type "set ARGYLL_MIN_DISPLAY_UPDATE_DELAY_MS="

But I don't know if DisplayCal will change it back automatically. (I used command line in the past and not DisplayCal). But You can easy test this by setting a very long time like 10s or so and see if the setting is still active.

I did a lot of testings regarding the patch settle time in the past, but didn't notice any advantage over the automatic setting.

Great to hear You solved the level problems. What was the issue of the lifted blacks?

S/W: LightSpace HCC, Calman 5, DisplayCal/ArgyllCMS
H/W: CR-100, I1Pro2, I1D3, eeColor, RB8 Generator
Tools: ControlCal, Ted's LightSpace CMS Calibration Disk

Last edited by DrFaxe; 04-26-2017 at 05:11 AM.
DrFaxe is offline  
post #643 of 669 Old 04-26-2017, 05:22 AM
AVS Forum Special Member
 
rickardl's Avatar
 
Join Date: Nov 2003
Location: Stockholm, Sweden
Posts: 2,144
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 123 Post(s)
Liked: 95
Quote:
Originally Posted by DrFaxe View Post
Great to hear You solved the level problems. What was the issue of the lifted blacks?
When changing the graphic settings (from ycbcr) on my extended screen, it didn't ripple through
the laptop->oppo->eecolor->plasma chain correctly until I pulled out and re-inserted the hdmi cable from the laptop.
rickardl is offline  
post #644 of 669 Old 04-26-2017, 11:27 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,537
Mentioned: 63 Post(s)
Tagged: 1 Thread(s)
Quoted: 1732 Post(s)
Liked: 2131
Quote:
Originally Posted by rickardl View Post
I used 56 for the color setting on the display when I ran DisplayCAL yesterday and when I checked the contrast clipping and color clipping patterns on the Spears & Munsil v2 disk, it showed no clipping with the LUT enabled but showed color clipping with it disabled. So even though it clipped in the intial setting, the resulting LUT was good with no clipping and that made me wonder if I should increase it even more to cover more of the gamut to start with.
It's not a good idea to start a 3D LUT profiling with display which already clips, because software will reduce your peak output to fix that errors (so you sacrifice data to your signal for no reason) and you will have lower peak output at the end.
sillysally likes this.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is offline  
post #645 of 669 Old 04-27-2017, 01:29 PM
AVS Forum Special Member
 
rickardl's Avatar
 
Join Date: Nov 2003
Location: Stockholm, Sweden
Posts: 2,144
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 123 Post(s)
Liked: 95
Quote:
Originally Posted by DrFaxe View Post
Your 168ms patch settle time seems to be OK, as Samsung plasmas are more prone to image retention and afterglow then Panasonic.
Those figures were not from my display. These are from mine:
Code:
Measured display update delay of 201 msec,  using delay of 368 msec & 0 msec inst
but keep in mind that I have the chain laptop -> oppo -> eecolor -> display here.

Last edited by rickardl; 04-27-2017 at 02:07 PM.
rickardl is offline  
post #646 of 669 Old 04-27-2017, 01:38 PM
AVS Forum Special Member
 
rickardl's Avatar
 
Join Date: Nov 2003
Location: Stockholm, Sweden
Posts: 2,144
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 123 Post(s)
Liked: 95
Quote:
Originally Posted by ConnecTEDDD View Post
It's not a good idea to start a 3D LUT profiling with display which already clips, because software will reduce your peak output to fix that errors (so you sacrifice data to your signal for no reason) and you will have lower peak output at the end.
I understand your reasoning.
But in a small test (73 patches), with color at 50 (default) the gamut coverage/volume (sRGB) increased from
87.8% / 98.3 % to
89.0% / 104.1 % with color at 56 while the measured luminance basically remained the same:
156.1 cd/m² vs
155.6 cd/m².


Edit: but maybe the gamut coverage/volume is not needed/relevant, I don't know. I am here to learn.

Last edited by rickardl; 04-27-2017 at 11:22 PM.
rickardl is offline  
post #647 of 669 Old 04-28-2017, 04:13 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,537
Mentioned: 63 Post(s)
Tagged: 1 Thread(s)
Quoted: 1732 Post(s)
Liked: 2131
Quote:
Originally Posted by rickardl View Post
I understand your reasoning.
But in a small test (73 patches), with color at 50 (default) the gamut coverage/volume (sRGB) increased from
87.8% / 98.3 % to
89.0% / 104.1 % with color at 56 while the measured luminance basically remained the same:
156.1 cd/m² vs
155.6 cd/m².

Edit: but maybe the gamut coverage/volume is not needed/relevant, I don't know. I am here to learn.
Using Color (global luminance control) you are not expanding the gamut; inless you have low primary color luminance generaly, when you have selected the option to your display where it gives you the more saturated primaries (native gamut), color slider is adding more luminance to primaries, so your gamut by looking the CIE Chart will remain the same, only the Luminance will be increased (which is not visible in CIE chart..it's the 3rd dimension).

These measurement you have taken are related with pre-calibration gamut coverage, in post calibration after the 3D LUT correction loaded, to see the real coverage you have to measure the post-calibration...but 73 colors are not capable to give you an accurate calculation number either, you need some thousands points to have an accurate presentation of the coverage percentage.

During pre-calibration you can increase color slider (to give more luminance to primaries), if you see that one primary has lower luminance from the ideal number, but if it's more saturated; as example lets say that your 100% Red has less luminance than required but the same time is more saturated from ideal....the software to de-saturate it will add equal amount of Green+Blue (when it will do this, it's Luminance will be increased, to there times you don't need to add as much luminance during pre-calibration to some cases, which will avoid clipping also; unless you have check that you have no color clipping using some reference patterns.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is offline  
post #648 of 669 Old 04-30-2017, 01:53 AM
AVS Forum Special Member
 
rickardl's Avatar
 
Join Date: Nov 2003
Location: Stockholm, Sweden
Posts: 2,144
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 123 Post(s)
Liked: 95
In the log window, I see these messages, the "out of range" message, is that something I should be concerned about?












17:56:56,606 Test display patch window, Set Video LUTs, Install profiles, Version 1.9.2


17:56:56,607 Author: Graeme W. Gill, licensed under the AGPL Version 3


17:56:56,607 Diagnostic: -d parameter '2' is out of range


17:56:56,607 usage: dispwin [options] [calfile]
...
rickardl is offline  
post #649 of 669 Old 04-30-2017, 06:09 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 945
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 261 Post(s)
Liked: 196
Quote:
Originally Posted by rickardl View Post
In the log window, I see these messages, the "out of range" message, is that something I should be concerned about?
Something/someone has invoked dispwin with an invalid display selection - i.e. there is no "display 2" on your system.
gwgill is offline  
post #650 of 669 Old 05-06-2017, 01:16 AM
Member
 
Bloodwound's Avatar
 
Join Date: Apr 2007
Posts: 195
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 126 Post(s)
Liked: 37
@gwgill :
Was curious how my old dtp94(b)s measure on a modern monitor, so I did a test measure with W and they measured about 6100k. So a bit off, but a superb inter-instrument agreement as always. Anyways that wasn't what i was gonna ask about.

Did you remove the need for initial black reading on the dtp94/dtp94b (and each 10mins afterwards) in v1.92. If so, was this really unneeded all this time? That would actually make these meters quite a bit more usable.
Bloodwound is offline  
post #651 of 669 Old 05-07-2017, 05:40 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 945
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 261 Post(s)
Liked: 196
Quote:
Originally Posted by Bloodwound View Post
@gwgill :
Did you remove the need for initial black reading on the dtp94/dtp94b (and each 10mins afterwards) in v1.92. If so, was this really unneeded all this time? That would actually make these meters quite a bit more usable.
I'm not sure what you mean. I don't think the DTP94 behaves this way. Perhaps you are confusing it with the i1Pro ?
gwgill is offline  
post #652 of 669 Old 05-07-2017, 11:21 PM
Member
 
Bloodwound's Avatar
 
Join Date: Apr 2007
Posts: 195
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 126 Post(s)
Liked: 37
Quote:
Originally Posted by gwgill View Post
I'm not sure what you mean. I don't think the DTP94 behaves this way. Perhaps you are confusing it with the i1Pro ?
It does in Calman and other packages that use the official driver, doesn't it? I wrongly assumed the same was the case with the Argyll driver.
But you are in fact saying that the black calibration isn't needed on DTP94s? Maybe you're right about that. I'll do some tests later.
Bloodwound is offline  
post #653 of 669 Old 05-08-2017, 05:31 AM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 945
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 261 Post(s)
Liked: 196
Quote:
Originally Posted by Bloodwound View Post
But you are in fact saying that the black calibration isn't needed on DTP94s? Maybe you're right about that. I'll do some tests later.
From memory, while it's possible to do a black calibration, it isn't required.
gwgill is offline  
post #654 of 669 Old 06-23-2017, 05:16 AM
Member
 
Alaric's Avatar
 
Join Date: Feb 2001
Location: Huntingdon, Cambridgeshire, UK
Posts: 187
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 11
Hi there,

Just picked up an eecolor box as my Mitsubishi HC5 (HC9000) has some odd CMS stuff. I had been playing with HCFR and tweaking it and got reasonable on the Grey scale, but adjusting red interacted with green and blue and despite doing loads of runs I could never hit a happy place, along with the strange gamma controls, I started looking for another way.

I've got a Colormunki Display which replaced my OLD Spyder2 from my Barco days. It works well with HCFR and is good for DisplayCal / Argyll though Calman only supports it as a Dev option (how stupid) and Lightspace doesn't even get that far (Which considering Free software supports seems somewhat daft). So DisplayCal and the eecolor box are my only real options (Why would someone buy a new meter to use expensive software to support expensive hardware when seemingly the best option is also the most cost effective?!?!?)

I set up my Windows 10 basic net-top last night which has a HDMI out though on Intel graphics and got an extended desktop, though that took some doing in itself, particularly going through my Nakamichi pre and the eecolor box, but I made it!

I managed to do a short run using the internal DisplayCal paterns and using a bit of the work through here
http://www.hometheatershack.com/foru...-under-1k.html

Set my PJ to a new user setting and worked out 6000K gave a reasonable reading for DisplayCal to work off. Better than 6500 oddly for matching color!

Took about 15mins to load the unity luts to reset the box to 6 lots of Unity and then I did about 150 patches which probably took 40mins to read and then about the same to calculate it all. Alas ran out of evening to actually upload the basic lut and see what HCFR thought about it, but I'm hopeful!

I probably need to set aside a couple of evenings next week to get the Madvr test patterns up and check the levels and then get a full 1500 patches run through it (am betting 3+hrs) but after a lot of reading arround the subject I'm really liking the idea of the eecolor box and 3D Luts to balence out the image on the PJ.
It reminds me a little of my Dirac box but for image rather than sound

Thanks,
Lee

Lee Edward Armstrong
My Clothes Will Impress You, And My Claws Will Undress You. - The Damned
Alaric is offline  
post #655 of 669 Old 06-23-2017, 05:36 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,537
Mentioned: 63 Post(s)
Tagged: 1 Thread(s)
Quoted: 1732 Post(s)
Liked: 2131
Quote:
Originally Posted by Alaric View Post
I've got a Colormunki Display which replaced my OLD Spyder2 from my Barco days. It works well with HCFR and is good for DisplayCal / Argyll though Calman only supports it as a Dev option (how stupid) and Lightspace doesn't even get that far (Which considering Free software supports seems somewhat daft). So DisplayCal and the eecolor box are my only real options (Why would someone buy a new meter to use expensive software to support expensive hardware when seemingly the best option is also the most cost effective?!?!?)
Hi Lee,

LightSpace supports you meter, the ColorMunki Display, see there for more info: LightSpace 3D LUT Home Cinema Calibration Software

The free version of LightSpace DPS can be used for measurements only not for display characterization providing a 3D LUT correction generation (which requires the paid version of software.)
Alaric and RIppolito like this.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is offline  
post #656 of 669 Old 06-25-2017, 12:14 AM
Senior Member
 
Join Date: Aug 2015
Posts: 346
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 243 Post(s)
Liked: 48
Quote:
Originally Posted by Alaric View Post
Took about 15mins to load the unity luts to reset the box to 6 lots of Unity and then I did about 150 patches which probably took 40mins to read and then about the same to calculate it all. Alas ran out of evening to actually upload the basic lut and see what HCFR thought about it, but I'm hopeful!

I probably need to set aside a couple of evenings next week to get the Madvr test patterns up and check the levels and then get a full 1500 patches run through it (am betting 3+hrs) but after a lot of reading arround the subject I'm really liking the idea of the eecolor box and 3D Luts to balence out the image on the PJ.
It reminds me a little of my Dirac box but for image rather than sound
The Colormunki display has an enforced slow read coded into the firmware, which makes it a bit painful to use for profiling purposes (it is OK fro small numbers of patches). Probably worth trying to pick up an i1d3 if you can if you're going to be doing 1000+ patch runs. WEX photographic do them new for £159 at the moment; you'd probably get a reasonable amount for the colormunki on eBay (I've seen them go as much as £100 used)
bobof is offline  
post #657 of 669 Old 06-26-2017, 05:23 AM
Member
 
Alaric's Avatar
 
Join Date: Feb 2001
Location: Huntingdon, Cambridgeshire, UK
Posts: 187
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 11
Quote:
Originally Posted by bobof View Post
The Colormunki display has an enforced slow read coded into the firmware, which makes it a bit painful to use for profiling purposes (it is OK fro small numbers of patches). Probably worth trying to pick up an i1d3 if you can if you're going to be doing 1000+ patch runs. WEX photographic do them new for £159 at the moment; you'd probably get a reasonable amount for the colormunki on eBay (I've seen them go as much as £100 used)
Yeah. I've read it adds a second in the firmware (surprised it's not been hacked actually). However given my net-top doesn't have the fastest processor in the world, the whole rig is on the slow side. Still I have a nice plasma TV rig in the living room to and I can leave the set-up churning away and watch a movie or even on over night. Once you get into 30+ readings you're gonna be more than just twiddling your thumbs anyway! I also recall how slow my Spyder2 was with my Barco and a *VERY* old version of Calman....I'm hoping once set-up i'll pack away the meter, so it's not a major biggy!

Lee
bobof likes this.

Lee Edward Armstrong
My Clothes Will Impress You, And My Claws Will Undress You. - The Damned
Alaric is offline  
post #658 of 669 Old 07-21-2017, 04:02 PM
Member
 
zachfine's Avatar
 
Join Date: Apr 2007
Location: Beijing, China
Posts: 15
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 10
I'm trying to use DisplayCAL to create LUTs for an eecolor box, using an i1displaypro as probe and DaVinci Resolve as output (rather than MadVR). I got DisplayCAL to communicate with the CalMan calibration code in Resolve and managed to upload a 3D LUT to the eecolor and it looked good. But having read the page at Light Illusion about generating 3D LUTs for the eecolor, I wondered if I had missed a step.

Specifically, they include these steps:

REPLACING THE 1D LUTS

The eeColor, for reasons best know to itself, has an output path that includes a first 1D LUT, followed by a 3x3 matrix, followed by a second 1D LUT. The first 1D LUT undoes a default 2.2 gamma, while the second 1D LUT reapplies the 2.2 gamma. It is a good idea to replace the 1D LUTs with Null versions, to avoid any potential issues.
  1. Generate a default Bypass LUT within LightSpace using 'File/New'
  2. Export the LUT in both eeColor 1D formats
  3. Make 3 copies of each 1D LUT and re-name as 'first1dxxx.txt and second1dxxx.txt', where xxx is the LUT colour red/green/blue
  4. Go to directory C:\Program Files\Entertainment Experience LLC\TruVue eeColor Application\Data\default
  5. Replace the three original 1D LUTs with the new Null LUTs exported from LightSpace

etc etc.
Is this also part of the DisplayCAL/Argyll workflow for eecolor? Is there a way to generate a null 1D LUT using these applications? I'm a little worried that this section of the workflow is not represented in the tutorial.

Thanks for any information.
zachfine is offline  
post #659 of 669 Old 07-22-2017, 07:06 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,537
Mentioned: 63 Post(s)
Tagged: 1 Thread(s)
Quoted: 1732 Post(s)
Liked: 2131
first1dxxx.txt and second1dxxx.txt

Quote:
Originally Posted by zachfine View Post
I'm trying to use DisplayCAL to create LUTs for an eecolor box, using an i1displaypro as probe and DaVinci Resolve as output (rather than MadVR). I got DisplayCAL to communicate with the CalMan calibration code in Resolve and managed to upload a 3D LUT to the eecolor and it looked good. But having read the page at Light Illusion about generating 3D LUTs for the eecolor, I wondered if I had missed a step.

Specifically, they include these steps:
REPLACING THE 1D LUTS

The eeColor, for reasons best know to itself, has an output path that includes a first 1D LUT, followed by a 3x3 matrix, followed by a second 1D LUT. The first 1D LUT undoes a default 2.2 gamma, while the second 1D LUT reapplies the 2.2 gamma. It is a good idea to replace the 1D LUTs with Null versions, to avoid any potential issues.
  1. Generate a default Bypass LUT within LightSpace using 'File/New'
  2. Export the LUT in both eeColor 1D formats
  3. Make 3 copies of each 1D LUT and re-name as 'first1dxxx.txt and second1dxxx.txt', where xxx is the LUT colour red/green/blue
  4. Go to directory C:\Program Files\Entertainment Experience LLC\TruVue eeColor Application\Data\default
  5. Replace the three original 1D LUTs with the new Null LUTs exported from LightSpace

etc etc.
Is this also part of the DisplayCAL/Argyll workflow for eecolor? Is there a way to generate a null 1D LUT using these applications? I'm a little worried that this section of the workflow is not represented in the tutorial.

Thanks for any information.
Hi, I have posted about this here: LightSpace 3D LUT Home Cinema Calibration Software

If you upload one of first1dxxx.txt & one second1dxxx.txt file, I can import them to LightSpace and let you know if there bypass LUT's.

If you can't upload them here, send me a msg: http://www.displaycalibrations.com/contact_us.html

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is offline  
post #660 of 669 Old 07-24-2017, 11:02 PM
Member
 
zachfine's Avatar
 
Join Date: Apr 2007
Location: Beijing, China
Posts: 15
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 10
Quote:
Originally Posted by ConnecTEDDD View Post
If you upload one of first1dxxx.txt & one second1dxxx.txt file, I can import them to LightSpace and let you know if there bypass LUT's.
Here's a link to the LUTs that Displaycal3 generated:
https://www.dropbox.com/s/f40kegekmm...n%202.zip?dl=0

I may have misinterpreted the instructions. I had somehow gotten the impression that the only file that I needed to upload to the eecolor was the single 3D matrix .txt file, in my case renamed to "3dLUT_2.txt", and not the full set of "first1dblue, first1dgreen, etc" files. That somehow the single .txt file was for video levels (and my devices output video levels, though I could change that if a reason popped up), and then for some reason full range signals required not the 1 3D LUT but the other 6 files (and I'm not entirely sure where those six files would be placed for the transfer). My impression was formed by this bit of text earlier in this thread:

VI. Upload 3DLUTs to eeColor box

Follow Buzz's guide (2nd post) if you are using YCbCr flow.

For full-range RGB in/out 3DLUTs you will also need to replace the following 6 files in the directory truvue\Data\default:

first1dred.txt
first1dgreen.txt
first1dblue.txt
second1dred.txt
second1dgreen.txt
second1dblue.txt
So the only file that I actually uploaded to the eecolor is the one I'd renamed "3dLUT_2.txt" so that it'd show up as preset #2 on the eecolor. I also made a custom OSD graphic for the LUT to differentiate it from the others on the device.

Video displayed using this LUT looked closer to video on my reference monitor than video displayed with the LUT disabled - though my attempts to use DisplayCAL to verify the result were not so hot. Maximum ?E*00 of 13.7, avg around 7. I am not sure I'm doing the verification properly, as I find the settings a bit confusing even after reading the documentation. And the fact that the eecolor has buttons for switching between "D65 W" and "Max W" that possibly affect each preset make me wonder whether those buttons are adding or removing additional LUTs to the pipeline of which I need to be aware.

It feels like I'm getting pretty close despite the bad verification numbers. I'd like to manage to do a verification pass that I trust, and to figure out for certain which LUT I should be applying and what's happening to it within the eecolor box (i.e. is it hit with any LUTs before and after application that need to be bypassed, and is there a D65/native-white LUT that inserts itself into the pipeline?

BTW, thanks so much to all who have posted information and have been so helpful.
Attached Thumbnails
Click image for larger version

Name:	2017-07-24 at 10.51 PM.jpg
Views:	11
Size:	87.8 KB
ID:	2257881  
zachfine is offline  
Sponsored Links
Advertisement
 
Reply Display Calibration

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