AVS Forum banner

981 - 1000 of 1693 Posts

·
Registered
Joined
·
9,736 Posts
Ok, I've tried Autocal with mixed results, so I'm thinking about taking first steps into doing a manual calibration.
The idea is to try a 2-Point WB calibration in the SM first, and then see where I go from there.
Can I do that with Calman Home for LG and the iTPG of the LG CX? Can a full manual calibration (if I ever get there) be done with this setup?
Use PGenerator, iTPG of LG is missing patches with CalMAN.

It will never miss a single patch in the last two years we have tested with LG Template and ColourSpace/LightSpace.

The problem is missing patches it coming from CalMAN. It's expecting a fix to the CalMAN code in CalMAN 2020 R2 for that issue.

I'm very impressed it took only two years from CalMAN to find out and resolve a very easy to notice a problem:


Even if you measure 10 times ColorChecker or Grayscale, you will see that some patches will have a bad reading when you will re-measured them, you will see fewer errors.


However, we don't suggest any user use iTPG of LG for SDR because it is not a bit-perfect.

It's a better idea to invest in PGenerator, to be sure that you are generating bit-perfect RGB 16-255 / 16-235 patches as LG will require for calibration.
 

·
Registered
Joined
·
69 Posts
Use PGenerator, iTPG of LG is missing patches with CalMAN.

It will never miss a single patch in the last two years we have tested with LG Template and ColourSpace/LightSpace.

The problem is missing patches it coming from CalMAN. It's expecting a fix to the CalMAN code in CalMAN 2020 R2 for that issue.

I'm very impressed it took only two years from CalMAN to find out and resolve a very easy to notice a problem:


Even if you measure 10 times ColorChecker or Grayscale, you will see that some patches will have a bad reading when you will re-measured them, you will see fewer errors.


However, we don't suggest any user use iTPG of LG for SDR because it is not a bit-perfect.

It's a better idea to invest in PGenerator, to be sure that you are generating bit-perfect RGB 16-255 / 16-235 patches as LG will require for calibration.
Hmm, seems like that upgrade fee moola goes towards marketing more than “upgrades”
 

·
Registered
Joined
·
189 Posts
When you have JETI or PhotoResearch Spectro, then use x0.309 y0.3290.

If you have i1PRO's, then use the coordinates I have posted for i1PRO's (they are Leon's coordinates; he had both JETI and i1PRO to find out)
Perceptual Mtaching / Alternative Whitepoint
Ive done it today :) Long story. Done by perceptual matching to my Eizo with an ipro 3 and i1Display pro. But with this coordinated you can do it also without a spectro. My Result is the same as ted's coordinates.. You can use the i1Displaypro without a spectrometer by perceptual mattching to a reference screen or by profiling with an Spectometer:

I Display Pro Plus with FSI.edr no profile: 0,3163 0,3345
I1Displaypro plus profiled with i Pro 3: 0,3119 0,3360 nearly as Ted
When you got an expensive spectrometer like yeti: 0,309 0,3290

The picture now looks amazing. It's not to red not to green or blueish. Every Set looks like it should be. Skin tones are real. It's not looking so digital anymore. It looks more natural.
 

·
Registered
Joined
·
1,324 Posts
Perceptual Mtaching / Alternative Whitepoint
Ive done it today :) Long story. Done by perceptual matching to my Eizo with an ipro 3 and i1Display pro. But with this coordinated you can do it also without a spectro. My Result is the same as ted's coordinates.. You can use the i1Displaypro without a spectrometer by perceptual mattching to a reference screen or by profiling with an Spectometer:

I Display Pro Plus with FSI.edr no profile: 0,3163 0,3345
I1Displaypro plus profiled with i Pro 3: 0,3119 0,3360 nearly as Ted
When you got an expensive spectrometer like yeti: 0,309 0,3290

The picture now looks amazing. It's not to red not to green or blueish. Every Set looks like it should be. Skin tones are real. It's not looking so digital anymore. It looks more natural.
As stated in Post 954, the i1 Pro Alternate White Point coordinates that are "suggested" for users to try come from a previous experiment I did

The white point Ted has recommended above is actually my white point. This was derived from an initial perceptual match to a plasma using the same/very similar methods that D-Nice uses while using a 5nm Jeti Spectro. I actually arrived at the same white point (+- xy 0.004) as D-Nice, so most users that have a 5nm Spectro or a meter profiled to one will use the coordinates provided by D-Nice.

I then measured the perceptually matched white point simultaneously with the Jeti and an i1 Pro 2 (which has since been sold to another forum member). So it it is up to you if you use it or not, bearing in mind X-Rites Inter-Instrument Agreement (meter variability).
Despite the official Inter-Instrument Agreement figures that are provided by X-Rite (Avg 0.4 dE94 - Max 1.0 dE94), then most 10nm i1 Pro Spectro's should be perfectly fine with those coordinates for an alternate white provided it is within spec, but this cannot obviously be guaranteed due to the calibration tolerances of consumer meters not being as tight as much more expensive reference meters. The fact you came up with the same coordinates for your i1 Pro in your own independent test is very interesting, so thank you for sharing.

But....with regards to the i1 Display Pro Plus (with EDR) coordinates, this once again shows that the meter variability of the retail meters is much larger. In fact much larger than I previously found with my own tests with multiple samples, along with information provided to me by other users (friends) with 5nm Spectros to use as a base reference to measure their i1d3 probes.

With a reference white patch of 0.309/0.329 (calibrated with a Jeti 5nm), all i1d3 probes (Retail and OEM) measured within a range of x 0.310-0.314 / y 0.330 - 0.334. Your i1d3 Pro Plus coordinates fall even further outside that range (on the y axis), proving that there can be no "set in stone" set of alternate coordinates for the i1 Display Pro meters, and that profiling to Spectro (Reference or Consumer) really can be of substantial benefit when calibrating WRGB OLED.
 

·
Registered
Joined
·
189 Posts
As stated in Post 954, the i1 Pro Alternate White Point coordinates that are "suggested" for users to try come from a previous experiment I did



Despite the official Inter-Instrument Agreement figures that are provided by X-Rite (Avg 0.4 dE94 - Max 1.0 dE94), then most 10nm i1 Pro Spectro's should be perfectly fine with those coordinates for an alternate white provided it is within spec, but this cannot obviously be guaranteed due to the calibration tolerances of consumer meters not being as tight as much more expensive reference meters. The fact you came up with the same coordinates for your i1 Pro in your own independent test is very interesting, so thank you for sharing.

But....with regards to the i1 Display Pro Plus (with EDR) coordinates, this once again shows that the meter variability of the retail meters is much larger. In fact much larger than I previously found with my own tests with multiple samples, along with information provided to me by other users (friends) with 5nm Spectros to use as a base reference to measure their i1d3 probes.

With a reference white patch of 0.309/0.329 (calibrated with a Jeti 5nm), all i1d3 probes (Retail and OEM) measured within a range of x 0.310-0.314 / y 0.330 - 0.334. Your i1d3 Pro Plus coordinates fall even further outside that range (on the y axis), proving that there can be no "set in stone" set of alternate coordinates for the i1 Display Pro meters, and that profiling to Spectro (Reference or Consumer) really is a requirement when calibrating WRGB OLED.
You'll have to recognise that my Coordinates are not exactly the same as yours. My white point differs from Ted's and Dnice's coordinates. So this is the reason my i1Dpro is fall even further outside that range. If i would go exactly to Dnices or Ted's coordinates it would be possibly in the range.
But these coordinates are my perseptual matching... Maybe it could be visually differ from your matching.

But what i have done is. Set the High Gain to 80ire. So in this case the RGB Balance is more parallel. Then i corrected the 100-2,5 Ire with 22 Gamma EQ. And my tv is no set to RGB low as Ted mentioned.
 

·
Registered
Joined
·
1,324 Posts
You'll have to recognise that my Coordinates are not exactly the same as yours. My white point differs from Ted's and Dnice's coordinates. So this is the reason my i1Dpro is fall even further outside that range.
Your i1 Pro coordinates are almost identical.

My i1Pro2 0.3120/0.3356
Your i1Pro3 0.3119/0.3360

To all intents and purposes, they are the same. In fact, when I first posted them here on AVS many months ago, I actually rounded them to 0.312/0.336. So as I say, they’re the same. It’s the i1d3 meters that have a much larger variation 😉

But yes, the perceptual difference of white as seen on WRGB OLED can vary from person to person. That is why I said before that you should first shoot for D65. If it looks incorrect, try the alternate coordinates provided. If neither of those two options look correct, then you will have to perceptually match to an alternate display type yourself.

Anyway, it’s really nice to see others actually going out of their way to do these things. I really appreciate you posting your results, thank you 👍🏻
 
  • Like
Reactions: mrtickleuk

·
Registered
Joined
·
189 Posts
Your i1 Pro coordinates are almost identical.

My i1Pro2 0.3120/0.3356
Your i1Pro3 0.3119/0.3360

To all intents and purposes, they are the same. In fact, when I first posted them here on AVS many months ago, I actually rounded them to 0.312/0.336. So as I say, they’re the same. It’s the i1d3 meters that have a much larger variation 😉

But yes, the perceptual difference of white as seen on WRGB OLED can vary from person to person. That is why I said before that you should first shoot for D65. If it looks incorrect, try the alternate coordinates provided. If neither of those two options look correct, then you will have to perceptually match to an alternate display type yourself.

Anyway, it’s really nice to see others actually going out of their way to do these things. I really appreciate you posting your results, thank you 👍🏻
Your welcome.
I've done the D65 Calibration before. But it doesn't look natural to me. To blueisch / redish.
Compared to my eizo D65 it looks very different.-
Because of i like my eizo color i used this for perceptual m.

Have you recognized that my I1displa pro coordinates without profile was measured with the fsi.edr and not xyz...?
 

·
Registered
Joined
·
7 Posts
Do you have a problem with near black before calibration?

Are you talking about SDR signaling?
Hi Ted, yes this was happening before calibration and yes I believe so.
Any type of YCbCr signal selected in the AMD GPU driver causes the screen to noticeably darken and cause the blacks to crush compared to an RGB signal.
I will test this outside of PC mode on the TV just in case that’s causing the problem.
It could be an graphics driver bug so I’ve reported the issue to AMD.
 

·
Registered
Joined
·
1,324 Posts
Have you recognized that my I1displa pro coordinates without profile was measured with the fsi.edr and not xyz...?
Yes, the range I mentioned earlier were all i1 Display Pro meters while using the FSI EDR. They were also a mixture of Rev.B Retail and OEM (the Pro Plus is an OEM with a Retail lock code).

Either way, the point is that you have managed to perform your own independent perceptual match with an i1 Pro Spectro and it aligns near as damn it perfectly with my own results using a 5nm Spectro and reading that patch with an i1 Pro also. This is the part I find very interesting :)
 

·
Registered
Joined
·
189 Posts
Your i1 Pro coordinates are almost identical.

My i1Pro2 0.3120/0.3356
Your i1Pro3 0.3119/0.3360

To all intents and purposes, they are the same. In fact, when I first posted them here on AVS many months ago, I actually rounded them to 0.312/0.336. So as I say, they’re the same. It’s the i1d3 meters that have a much larger variation 😉

But yes, the perceptual difference of white as seen on WRGB OLED can vary from person to person. That is why I said before that you should first shoot for D65. If it looks incorrect, try the alternate coordinates provided. If neither of those two options look correct, then you will have to perceptually match to an alternate display type yourself.

Anyway, it’s really nice to see others actually going out of their way to do these things. I really appreciate you posting your results, thank you 👍🏻
And maybe the i1displaypro differs because of differnet panelsizes??? The FSI.edr is avalable for 55 and 65 inches. But maybe they doesn't working godd for my 77 Inch, äh TV!
 

·
Registered
Joined
·
2,271 Posts
1+2) SM WB = Service Menu White Balance.

It has 2-Point controls there, RGB-LOW and RGB-High.

RGB High @ 192.192.192 provides the maximum panel dynamic range.

For keeping it at maximum, keep Red @ 192 and reduce only the other two color channels to reduce RGB balance errors.


3) Calibrate with a meter, but not say to one specific pattern....open OSD menu....and perform real-time RGB adjustments.

Measure the whole Grayscale, display a black patch for the panel to cool down, open the OSD menu, ...look at the software RGB balance charts, and adjust all controls at once with the values you believe that they will reduce RGB balance errors.

Then close the OSD menu and re-measure using a grayscale sweep.

4) LG OLED 2019/2020 PQ Curve Upload Free Template for DeviceControl Interface
Thanks for this, much clearer now.

As I'm using HCFR to calibrate, is there any way to use the internal pattern generator of the CX?

Otherwise, the only way I can play the patterns with my current setup is via Plex (I have AVSHD SDR patterns + Masciola's HDR patterns).
 

·
Registered
Joined
·
1,324 Posts
And maybe the i1displaypro differs because of differnet panelsizes??? The FSI.edr is avalable for 55 and 65 inches. But maybe they doesn't working godd for my 77 Inch, äh TV!
Ahhhhh, you have a 77” panel, I didn’t realise that. It could possible why there is a difference then. But nonetheless, there is definitely variability in i1 Display Pro meters.

None of it really matters anyway as you have a Spectro.


As I'm using HCFR to calibrate, is there any way to use the internal pattern generator of the CX?
The LG iTPG cannot be used with HCFR
 

·
Registered
Joined
·
2,271 Posts
The LG iTPG cannot be used with HCFR
So apart from using the SDR & HDR patterns played via Plex app in the CX, is there any better way to generate the patterns for calibration?
I know the PGenerator is an option, but was hoping to avoid needing to build an entire automatic generator since I have the patterns already and just need an easy and accurate way to display and calibrate with them.
 

·
Registered
Joined
·
18,383 Posts
Perceptual Mtaching / Alternative Whitepoint
Ive done it today :) Long story. Done by perceptual matching to my Eizo with an ipro 3 and i1Display pro. But with this coordinated you can do it also without a spectro. My Result is the same as ted's coordinates.. You can use the i1Displaypro without a spectrometer by perceptual mattching to a reference screen or by profiling with an Spectometer:

I Display Pro Plus with FSI.edr no profile: 0,3163 0,3345
I1Displaypro plus profiled with i Pro 3: 0,3119 0,3360 nearly as Ted
When you got an expensive spectrometer like yeti: 0,309 0,3290

The picture now looks amazing. It's not to red not to green or blueish. Every Set looks like it should be. Skin tones are real. It's not looking so digital anymore. It looks more natural.
AWPs are not needed for the 77” WRGB OLEDs. Just target standard D65. I’ve verified this again earlier this week.
 

·
Registered
Joined
·
1,324 Posts
So apart from using the SDR & HDR patterns played via Plex app in the CX, is there any better way to generate the patterns for calibration?
I know the PGenerator is an option, but was hoping to avoid needing to build an entire automatic generator since I have the patterns already and just need an easy and accurate way to display and calibrate with them.
I’m not sure what you mean by “better way”? You can either have automatically generated patterns, or pre-made manual ones (Such as Ted’s Disk, R.Masciola’s USB, etc).

If you are using HCFR and want automatic patterns, you will need an external signal generator (as you said, PGenerator is an option).

Otherwise you are limited to manual patterns (Disk/USB/Plex/etc).
 

·
Registered
Joined
·
2,271 Posts
I’m not sure what you mean by “better way”? You can either have automatically generated patterns, or pre-made manual ones (Such as Ted’s Disk, R.Masciola’s USB, etc).

If you are using HCFR and want automatic patterns, you will need an external signal generator (as you said, PGenerator is an option).

Otherwise you are limited to manual patterns (Disk/USB/Plex/etc).
Yes, sorry what I mean is that I want manual patterns, but wondering if internal Plex app is an accurate way of playing them for calibration purposes?

Or is using a USB stick preferred?
 

·
Registered
Joined
·
1,324 Posts
Yes, sorry what I mean is that I want manual patterns, but wondering if internal Plex app is an accurate way of playing them for calibration purposes?

Or is using a USB stick preferred?
This is only something you can decide. Neither are a particularly great choice as it can be an awkward and cumbersome experience manually switching the patterns, putting up with the players UI whilst also going in and out of the TVs picture menu settings to make adjustments etc, etc. This is why most people prefer to use an external generator because none of this is an issue. So if your determined to use manual patterns, it may be easier to play them back via an external device (Blu-Ray player etc) because at least then you can manually change the patterns and control any UI overlay with a separate remote, whilst using your TV remote to make the adjustments.
 

·
Registered
Joined
·
189 Posts
Yes, the range I mentioned earlier were all i1 Display Pro meters while using the FSI EDR. They were also a mixture of Rev.B Retail and OEM (the Pro Plus is an OEM with a Retail lock code).

Either way, the point is that you have managed to perform your own independent perceptual match with an i1 Pro Spectro and it aligns near as damn it perfectly with my own results using a 5nm Spectro and reading that patch with an i1 Pro also. This is the part I find very interesting :)
Yes thats really amazing. That shows us, that the ipro spectro is a good thing! Thanks for your response.
 

·
Registered
Joined
·
189 Posts
AWPs are not needed for the 77” WRGB OLEDs. Just target standard D65. I’ve verified this again earlier this week.
you are making me crazy :). I've done D65 3-4 times. It always looked to cool and blueish / redish. And it doensn't look like my eizo D65. So this is the reason why i choose perseptual matching with my eizo. And the result is like Ted's Result. Now you make me unsure....
How did you verified it?
Have you got a reference Monitor side by side?
I can do it too exactly to D65. And the results looks great in calman. But the picture not.
Got an 77C9.
 
981 - 1000 of 1693 Posts
Top