AVS Forum banner
41 - 60 of 1,661 Posts
Discussion starter · #41 ·
ColourSpaceZRO Beta 1.0.0.856 (13 January 2021)

Release Notes


Fixed issue with Sub Space and graphs, when doing manual measurements.

Temporary change to dpi‐aware operation to fix layout when display scale is over 100%.

NOTE: New Download Instructions

Please read about the new download instructions here. (Please refresh your browser cache to see the new instructions)

Installation Notes


ColourSpace Installer will completely overwrite any previously installed version, not just update it.

It will not require to un-install older releases before installing a newer version.

Update Notes

While ColourSpace is working in beta, when the users will start ColourSpace, it will not pop-up notification to inform about the availability of newer beta.

However, I will update this thread but also refresh the form page with release notes text per each beta.
 
Last night I experienced something strange.

I was confirming my calibrations on my Sony 950H

Custom/Warm : BT.1886/D65/@100nits
Pro1/Expert1: Rec. 709/2.2/D65@100nits
Pro2/Expert2: Rec. 709/2.2/D65@500nits

The first two calibration lined up pretty well, and with only minor adjustments to the luminances across the 21 pt adjustments. For both I was able to get the dE below .5 across the grey scale/primary and secondary colors.

Then I set into checking Pro2/Expert2 @ Max brightness. The prior calibration I had, which was accurate when I did it, was now quite a bit off. Mostly in luminance, Y being too low. As I worked my way from 95%-10% adjusting each point to target I was able to get things dialed in. However, when I would run a full sweep it would be off, this time, too bright on the Y. SO I'd go back down adjusting from 95%-10% (19-1 on the 20pt adjustments). I finally got it dialed in again, but once again the full sweep was off again, with all the Y values elevated evenly across the grey scale. Then I checked the difEOTF and it was an absolute train wreck. The lower luminances were okay, but starting around 40% it sky rocketed, with a deep plunge at 100%. I looked at each 20 pt setting and I had arrived at a mess. neighboring points were not in line, some grossly elevated, some grossly depressed.

Thinking something was amiss I went back and checked my first two presets both of which were measuring perfectly before, and low and behold, they were both measuring too bright as well now.

Any ideas why this is happening?

Thanks in advance.
 
Discussion starter · #43 ·
Last night I experienced something strange.

Any ideas why this is happening?

Thanks in advance.
Hi,

Be sure at your ColourSpace -> Settings, and the Profile Luma -> Auto is ticked for ColourSpace engine to calculate your charts properly based on your measured Reference Black/White values.

(Sub-Space unticked also)

After these details, if you see something wrong, it's coming from the display.
 
Hi,

Be sure at your ColourSpace -> Settings, and the Profile Luma -> Auto is ticked for ColourSpace engine to calculate your charts properly based on your measured Reference Black/White values.

(Sub-Space unticked also)

After these details, if you see something wrong, it's coming from the display.
Thanks Ted. I'll try it with auto Luma on.

The calibrations should be reproducible day-day with only minor variances?

Thanks

David

Sent from my Pixel 2 using Tapatalk
 
Discussion starter · #45 ·
Thanks Ted. I'll try it with auto Luma on.

The calibrations should be reproducible day-day with only minor variances?

Thanks

David
Patch generation settings and meter settings are important also.

Wait one hour before starting measurements for the TV to stabilize.
 
Patch generation settings and meter settings are important also.

Wait one hour before starting measurements for the TV to stabilize.
My routine has been to plug in the i1 to the laptop, take some repeat measurements on the laptop screen for 20-30 minutes prior to starting

The TV typically stays on most the day.

Last night I ran some repeat grey scale patches on the tv for 20 minutes or so before making measurements.

Should I still wait an hour after that?

**Should Auto Luma only be used during sweeps, or while making adjustments as well??

General Settings;
Target Gamut/EOTF
BT.1886 or Rec709_2.2 (user)

Sub Space - off

Luma Target
I usually use "update" both Max and Min before starting measurements

Profile Luma
Auto - off (should this be on while making adjustments, or just during sweeps??)

Stabilisation
OFF

Patch Size
L30

Patch Scale
16
235

Pre-roll CSV sequence
NONE

Active LUT - NONE


For the i1 display Pro here are my meter settings:

Calibration Settings:
Generic CMF
Intelligent Int. OFF
Average Low Light : OFF
Enable Visible Feedback: OFF

Integration Time:
This is variable. Sometimes I use 2.0 seconds for sweeps. Sometimes .75 for sweeps.

I usually adjust to .75 while making repeat measurements for the higher luminances, and move it to 2 seconds for lower luminances. Then do sweeps at 2.0 seconds

Maximum Exposure Time
default

Sync Frequency
NONE

Extra Delay Time
0.5

Probe Offset 0.00 x 0.00 y
 
Discussion starter · #47 ·
Set AUTO @ Luma. (leave it ticked forever)

AIO mode 0.75 sec, Inteligent Integrator 2.

If you see that the TV dims after you will display a patch, then try Full-Field.

If it still dims, use 255.255.255 BG with a 50% Area patch.
 
Set AUTO @ Luma. (leave it ticked forever)

AIO mode 0.75 sec, Inteligent Integrator 2.

If you see that the TV dims after you will display a patch, then try Full-Field.

If it still dims, use 255.255.255 BG with a 50% Area patch.
Thanks again Ted.

Intelligent Integrator at 2.0 would average the readings for all luminances below 2.0 nits correct?
 
Discussion starter · #49 ·
Thanks again Ted.

Intelligent Integrator at 2.0 would average the readings for all luminances below 2.0 nits correct?
Intelligent Integration is not averaging any measurement.

When you set 2, its the nit value below which the i1Display PRO will perform a second, third, fourth, etc. reading, with increasing integration times, up to the maximum integration time of 6 seconds.

The readings will stop as soon as a stable and valid reading is achieved.

When averaging is used for measurement, it will not provide any real improvement in inaccuracy.

Averaging works as a smooth filter with no attempt to get a more accurate measurement.
 
I discovered a quirk when using ”update“ for luminance targets vs auto update. I’m not sure if this is design intent.

I’m using AWP for my calibration (using D-Nice AWP for Sony X950H). When I use the individual min and max luminance update buttons, then go back to the measure tab, the xy measurement values are populated with standard Rec709 coordinates at least temporarily. Only when I use “auto” that my AWP gets populated properly. However, I believe that once measurement sweeps start, the proper AWP is being used.

Edited for accuracy, see below for resolution.
 
Discussion starter · #51 ·
I discovered a quirk when using ”update“ for luminance targets vs auto update. I’m not sure if this is design intent.

I’m using AWP for my calibration (using D-Nice AWP for Sony X950H). When I use the individual min and max luminance update buttons, then go back to the measure tab, the xy are populated with standard Rec709 targets at least temporarily. Only when I use “auto” that my AWP gets populated properly. However, I believe that once measurement sweeps start, the proper AWP is being used.
Manual Update can be used (with AUTO unticked) can be used for many different measurements reasons.

For example, when you want to set your peak output to specific nits, enter with your keyboard 100, so ColourSpace will populate all charts based on that target.

If you want to recalibrate to exact nits, for example:


Or if you want to evaluate a measurement file to specific nits to see which deviations you have.

When you want to see a saved measurement or perform a manual calibration, users have to tick AUTO of Profile Luma for the software to re-calculate always the charts/targets based to display black/white levels.

When you want to use a custom colorspace with AWP, you have to create a new colorspace based on REC.709 for SDR, save it with a different name, and select it as Target Gamut / EOTF.

The more stuff you know about how the software works, the better understanding and the results you will have.

It's the reason LightIllusion keeps updated all on-line guides for users to read how the software works.
 
Thank you Ted. I’ve been using custom user color space with my AWP as the target gamut. Even with it loaded I just saw the display quirk with the non-custom x-y coordinates being show as “Actual“ on the measure tab when not using “auto”. As I said the coordinates change to my custom AWP as soon as it runs through the calculation.

Also thank you for the explanation on pre-calibration on LG OLED. Seems like Sony has something equivalent in SDR - default contrast is 90, but when set to 100, above 235 is shown with associated bump in max luminance. At contrast = 100 is also when the 20 adjustment points line up with the patch list.

EDIT: the "quirk" I referred to is not the target values (those show my AWP) but the "actual" after a measurement, so I edited. So this may be about nothing - I saw some numbers change on the measure tab, but in fact the measured values don't populate until I actually make a measurement and not manual update of luminance target.
 
Ted, just so I'm clear:

To set the pre-calibration white point on the Sony 950H:

Contrast on tv to 90
Turn OFF Auto Luma and set the peak target at 100 nits in colourspace
Patch Scale 16-235 in ZRO
Measure 100% (235,235,235)
Adjust the *balance with the Basic Color Controls RGB Gain until RGB are at 0 in the RGB balance graph/Measurement tab.
Adjust Brightness slider on tv to get as close to 100 nits as possible.
Peak white is now aligned
Turn on AUTO luma and begin 20 pt grey scale adjustments.
 
Discussion starter · #54 ·
Thank you Ted. I’ve been using custom user color space with my AWP as the target gamut. Even with it loaded I just saw the display quirk with the non-custom x-y coordinates being show as “Actual“ on the measure tab when not using “auto”. As I said the coordinates change to my custom AWP as soon as it runs through the calculation.

Also thank you for the explanation on pre-calibration on LG OLED. Seems like Sony has something equivalent in SDR - default contrast is 90, but when set to 100, above 235 is shown with associated bump in max luminance. At contrast = 100 is also when the 20 adjustment points line up with the patch list.

EDIT: the "quirk" I referred to is not the target values (those show my AWP) but the "actual" after a measurement, so I edited. So this may be about nothing - I saw some numbers change on the measure tab, but in fact the measured values don't populate until I actually make a measurement and not manual update of luminance target.
Select your custom colorspace as Target Gamut, Tick AUTO @ Profile Luma, and then you are ready to start the measurements or see saved measurements.
 
Discussion starter · #55 ·
Ted, just so I'm clear:

To set the pre-calibration white point on the Sony 950H:

Contrast on tv to 90
Turn OFF Auto Luma and set the peak target at 100 nits in colourspace
Patch Scale 16-235 in ZRO
Measure 100% (235,235,235)
Adjust the *balance with the Basic Color Controls RGB Gain until RGB are at 0 in the RGB balance graph/Measurement tab.
Adjust Brightness slider on tv to get as close to 100 nits as possible.
Peak white is now aligned
Turn on AUTO luma and begin 20 pt grey scale adjustments.

The LG instructions had unticked the Profile AUTO because users want to target a specific nit level for 109% White manual pre-calibration of the TV.

You can use the same method, if you want 100 nits calibrated, to start with your 100% White adjustment, to help you reduce backlight and fix RGB balance.

After the 100% White adjustment, you can enable Profile LUMA again.
 
Ted, just so I'm clear:

To set the pre-calibration white point on the Sony 950H:

Contrast on tv to 90
Turn OFF Auto Luma and set the peak target at 100 nits in colourspace
Patch Scale 16-235 in ZRO
Measure 100% (235,235,235)
Adjust the *balance with the Basic Color Controls RGB Gain until RGB are at 0 in the RGB balance graph/Measurement tab.
Adjust Brightness slider on tv to get as close to 100 nits as possible.
Peak white is now aligned
Turn on AUTO luma and begin 20 pt grey scale adjustments.
This is what I understand now after thinking more about it.

Using Ted's LG example - the Sony X950H can display Video Legal Extended Range (up to 255.... when Contrast set to 90).

So then we need to calibrate to 109%. Patch Scale in ZRO should be 16-255. At patch level 235, the light output should be the desired target.
 
For some years I have been using my i1Display Pro Retail (Rev.A) profiled by Jeti Spectraval 1511 to calibrate my Panasonic OLED 65EZ950E using my AWP.

I adjusted 100% White in ZRO. This is result (dE 2000 0,0706):


These are probe settings:


Then I used Calman (same Correction Matrix, same Integration Time, same Pattern Generator, etc.).
I got this (dE 2000 1.55):


Ted, why this significant difference? Did I do something wrong?
 
For some years I have been using my i1Display Pro Retail (Rev.A) profiled by Jeti Spectraval 1511 to calibrate my Panasonic OLED 65EZ950E using my AWP.

I adjusted 100% White in ZRO. This is result (dE 2000 0,0706):


These are probe settings:


Then I used Calman (same Correction Matrix, same Integration Time, same Pattern Generator, etc.).
I got this (dE 2000 1.55):


Ted, why this significant difference? Did I do something wrong?
Something to consider...

ColourSpace no longer uses the traditional FCMM for meter correction. If you have used BPD correction files that were previously created within LightSpace, those BPD files will now be using ColourSpace FCVM meter correction, which on a WOLED display is far more accurate.

If your CalMAN meter correction is using the traditional FCMM meter correction, then it is possible there could be some difference.

However, the FCVM doesn't really affect the accuracy of White, as White is usually very accurate anyway with FCMM on WOLED.

If you could display the xyY of the readings that were taken in CalMAN, that could give some clue as to what the difference is, and why.

Just one question though....Did you ensure to apply "Probe Matching Active" in ColourSpace Graph Options? It is something that is easily overlooked until you become more familiar with the interface, and if it hasn't been enabled it would explain the difference.
 
Probe Matching Active: not checked. Is this?
 
So then we need to calibrate to 109%. Patch Scale in ZRO should be 16-255. At patch level 235, the light output should be the desired target.
If Patch Scale is set to 16-255, then you need to configure the sliders to 234,234,234 to obtain an actual output pattern of 235,235,235 (100%), and configure the sliders to 255,255,255 to obtain an actual output pattern of 255,255,255 (109%). If Patch Scale is set to 16-235, then you need to configure the sliders to 255,255,255 to obtain an actual output pattern of 235,235,235 (100%).
 
41 - 60 of 1,661 Posts
Top