| Forums | Register | Reply | Polls | Search | Statistics |
Bugs & WIBNIs Light Illusion Forums / Bugs & WIBNIs /  
 

Bug reports when attempting to calibrate HDR

 
Author astralstorm
ZRO
Male
#1 | Posted: 2 May 2021 19:01 
This pertains to ColourSpace 1.0.0.1095.

I've been taking my time with this software and so far found a few bugs:
1) Enabling stabilization patch puts additional one between drift measurement patch and actual measurement. This makes the drift measurement less accurate for screens that heat up and drift quickly, such as LG OLEDs in HDR mode.
2) Drift is greyed out for some reason until a measurement from current patch type is done. Abort one and you can see it being available and working.
3) I've had multiple error boxes trying to use algorithms other than Peak Chroma and Peak Luma for HDR 3DLUT attempts in calibration mode (gamma 2.2), with one exception. No idea what causes the problem. If needed, I can send a grid measurement that caused problems.

Author astralstorm
ZRO
Male
#2 | Posted: 2 May 2021 19:03 
Attached problematic BCS.

LG CX 2.2 Cube Cooled.bcs.zip Attached file:
One of measurements where top 3 generation modes fail.

 

Author Light Illusion


Male
#3 | Posted: 2 May 2021 19:24 
Thanks for the feedback.

1) All Stabilisation patches are generated after every other patch - Drift, or Measured.
That is correct, as any patch will cause possible 'heat' issues, so will always need a Stabilisation to counter that.
The Stabilisation patch will therefore have the same 'impact' of Drift, as every other patch.

2) We can't replicate this - can you please provide further info?

3) Again, we will need to have further info.
However, if you mean the Profile in the following post, that lacks full range stimulus values.
What was the patch set was used?
It appears to have Limit Range active?
That is a very special use case for burnt-in PQ HDR, and you need to understand its application, as per the User Guides.

Steve

Author astralstorm
ZRO
Male
#4 | Posted: 2 May 2021 20:36 | Edited by: astralstorm 
2) Opening a new patch, connecting to devices gives a grayed out Drift compensation checkbox. After switching Mode, it's still greyed out.
For additional info, it's i1d3 probe with MadTPG output, Automatic calibration patches setting.

3) Weird. I'm pretty sure I've exported a BCS that had no limit, measured in UHD Rec2020 and gamma 2.2 in calibration mode, with no limit range. I've opened it and checked. Limit range is unchecked but input box is active. Probably another bug?
Patchset was a cube, I think 11x11x11, with drift tracking (203 nit white) and stabilization black patch.

Author Light Illusion


Male
#5 | Posted: 2 May 2021 20:42 
Something is wrong with the patch set.
You can see that wit the RGB Separation graph - there are no plots.
That means there are no matching RGB patches that correspond to their equivalent grey patches.

Steve

RGB Separation
RGB Separation

Author astralstorm
ZRO
Male
#6 | Posted: 2 May 2021 20:47 
All the patches I've measured have no RGB separation when I reopen them, but have the cube. This is very weird.
Another patch with the same issue, faster 9x9x9 cube. It might be related to having a user color space? (which is UHD Rec2020 modified to gamma 2.2)

LG CX 2.2 CSP 2.bcs.zip Attached file:
Another faulty file

 

Author Light Illusion


Male
#7 | Posted: 2 May 2021 20:52 | Edited by: Light Illusion 
Are you trying ti use Sub-Space?
Again, there are no RGB full saturation patches in the profile.
You can see this by selecting 'Targets Only'.
The patches are all low gamut.

Steve

Targets Only
Targets Only

Author astralstorm
ZRO
Male
#8 | Posted: 2 May 2021 21:08 | Edited by: astralstorm 
I have a suspicion it could be tied to me running a greyscale check before profiling with the cube.
Yes, I did subspace with DCI P3 D65 gamma 2.2 in this. Apparently misunderstood what it does.

Author Light Illusion


Male
#9 | Posted: 2 May 2021 21:10 
No, running a Grey Scale check can have no impact one anything...
I think you have Sub-Space enabled?

Steve

Author astralstorm
ZRO
Male
#10 | Posted: 2 May 2021 21:20 | Edited by: astralstorm 
Regarding 1) for HDR calibration where the screen extremely quickly heats up (5% in seconds of hot patches unless very long stabilization patch is used for cooldown), a drift patch that is "cold" (e.g. 60 nit) placed right after the measured patch would probably be more effective for tracking the drift than one after the long-ish cooldown.

Author Light Illusion


Male
#11 | Posted: 2 May 2021 22:27 | Edited by: Light Illusion 
Sorry, but I have no idea where you got that concept from.
But there is no logic to it.
The whole concept of 'Drift' is it compensates for the display/probe combination, regardless of the patches before/after the Drift patch.
That is the whole idea of them.
Any display takes a considerable period of time to heat up/cool down.
But the change with the display response is faster.
Far more than a patch duration.
And that is why the Calman concept of heat measurement for calibration is so flawed.
(Stupid idea...)
By the time any heat measurement changes, the display will have already changed.

And separately, the other issue with WOLEDS is image burn-in.
The Stabilisation patches are there to counter that too, using the FW option.

Steve

Author astralstorm
ZRO
Male
#12 | Posted: 3 May 2021 19:27 
It seems the brightness quickly rising in measurement is an artifact of i1d3 itself and not screen, as the i1 Studio spectro has at most a single digits nits changes. Thank you for your time.

Bugs & WIBNIs Light Illusion Forums / Bugs & WIBNIs /
 Bug reports when attempting to calibrate HDR

Your Reply Click this icon to move up to the quoted message


Watch this topic / Get updates by Email
 (?)
Only registered users are allowed to post here. Please, enter your username/password details when posting a message, or register first.

 

 
Online now: Guests - 2
Members - 1 [ Omardris ]
Max. ever online: 312 [11 Jun 2021 17:33]
Guests - 312 / Members - 0