Jump to content

Calibrating QD-OLED TV's SONY A95K / Samsung S95B - It's Different.


Recommended Posts

So I've ordered and prepaid for a SONY A95K 65", ETA August 8th, and have been reading up on how things have been going regarding calibrating these new QD-OLED displays has been going at the AVS forum.

 

Having done my own calibrating for years I've decided I'm going to update my old XRITE i1Display Pro meter as well.......well they aren't even called XRITE anymore, and I know it won't cut it with these newer brighter displays.

And I think I'm going to move to HCFR from ChromaPure software for calibration few reasons....New Meter compatibility being one.

[Not that I dislike ChromaPure, but it will not be getting any more updates, as Tom Huffman, the owner of ChromaPure is retiring, plus he no longer has a software engineer working for him ]

 

I'll PM Tony O'Brien later and ask him some questions, and his recommendation on a replacement meter, but I'm pretty sure he's going to say get either a Calibrite ColorChecker Display PRO of PLUS.

These are basically renamed/branded/updated versions of the Xrite i1Display PRO.

https://calibrite.com/us/product-category/filmmakers/

 

The difference between the PRO and PLUS is the Pro is rated to 1000 nits and the PLUS to 2000 nits, plus both have ambient light sensors and measurement.

 

I have usually bought my calibration gear from Image Science in Melbourne, but I see they are only stocking the PRO version ATM [and the lower version]

I'm unsure if trying to get a PLUS version is needed, as 1000 nits is plenty bright already, something I expect Tony would have better knowledge of and hopefully post back about.

 

Anyway.

It seems some adjustments are going to be needed if anybody intends to try doing their own calibration of a QD-OLED at home with their own calibration gear, as XRITE based meters are currently not getting correct readings.

 

Having seen this sort of thing happen years ago when Samsung changed the pixel structure to a T shape on a range of it's TV's, I suspect the Pyramid like pixel structure of these Samsung sourced QD-OLED panels is causing at least part of the problem.

The added Blue layer over a standard WOLED is also something that is causing problems with readings.

 

What's been going on at the AVS forum is some heavy hitters in the calibration world have been doing calibration tests using the Samsung S95B [They haven't been able to get hold of a A95K], which the SONY A95K uses the same panels.

They have been getting strange results, and to cut a long story short, they have been trying to workout how to compensate for the readings they have been getting.

 

The person that is leading the charge [AVS forum member dwalete  ] has been using some very high spec'd calibration gear [ Jeti 1501 (4.5nm) spectro ] , and is using a Pioneer KURO Plasma display as the Standard to aim for [Next best thing to a old CRT display, which is still the best ]

 

He in turn has made Spectral Data from his measurements available to another [AVS forum member aron7awol ] who has painstakingly tuned this data into a usable .ccss file, so with this data it could be used with the ArgyllCMS-enabled software such as HCFR/DisplayCal.

 

You can read about this in the link below [POST #12,416 - 17th June ] , plus aron7awol has kindly posted this .ccss file as a PDF in the same post, so if you are using HCFR , from my understanding, you should be able to use it as well to calibrate a QD-OLED

https://www.avsforum.com/threads/samsung-4k-s95b-qd-oled-owners-thread-no-price-talk.3240720/page-621#post-61758430

If reading that post, you should really read the several pages before and after it, as it explains how the original calibration took place, and the reason for the choices made

 

Having not used the HCFR software in donkeys years, how one goes about putting the .ccss file data into HCFR so XRITE meters reads more like a US$7000 Jeti 1501 spectro is unknown to me, so if anybody could reveal that and post in this thread, I think everyone else who uses HCFR and intends to calibrate a QD-OLED would find it most helpful.

I know I certainly would.

 

Seeing that HCFR is open source software, I'm hoping that somebody will soon make this process easy, and selection of a preset for calibration of QD-OLED displays will be all that is needed, but ATM that isn't the case.

 

Early days.

 

 

 

 

 

Link to comment
Share on other sites



  • 1 month later...

Well further news regarding calibrating ANY QD-OLED display....Unfortunately its not good news.😫😡

 

I only received my SONY A95K last week, and have been waiting until it had at least 200 hours on it before I attempted a calibration, it now seems I'd be wasting my time.

 

I've been quite specific in asking questions of the ISF calibrators at the AVS forum about DIY calibration of QD TV's, and the repeated reply has been unless you have a high end spectro,  it can't be done.

 

I'll quote some of the questions I asked with their replies, which will give you a idea of the reasons why a attempt at home calibration without a high end spectro is futile.

 

This was a reply to another forum members question about calibrating a QD TV.

 

You probably will not find very much about those displays in the DIY calibration areas. Their wide and "spiky" color gamuts really need measurements with a professional-grade high-resolution spectrophotometer (read "$$$$$") for accurate results, and most home calibrators don't have access to these.

 

I replied with this question.

 

How much of a difference are we talking about in the accuracy of colour readings between a Calibrite i1Display Plus and a high level spectro ?
I keep reading posts that are basically hinting that you would be totally wasting your time even attempting to calibrate a A95K or Samsung S95B using anything less than a high end spectro.
Mind you, those posts all seen to be originating from professional calibrators, where a very high level of calibration exactitude exists
Surely that can't be the case ?
OK, I'm obviously not expecting to get as good results calibrating my A95K with the above meter once my A95K has had a suitable amount of bed-in time, but I'm also not expecting it not to be able to measure and adjust settings well enough to get pretty close.

 

The reply I got below.

There is no way of getting around the fact you need a 5nm or better spectro to use on QD-OLEDs. Attempting to calibrate with anything less is pretty much pointless where you will end up something equal or worse than what you get OOTB.

 

I asked another question.

Thanks for the disappointingly bad news.
Why is does this problem exist with QD-OLED and not standard WOLED ?
Is it due to the different pixel structure in QD-OLED ?
If memory serves me correctly, years ago Samsung had a T shaped shaped array on its displays, when Green was larger and caused luminance error readings, is the same sort of thing happening with QD-OLED

 

Reply below

It is basically anything with Quantum Dots, whether OLED or LCD based. The QDs produce very narrow, spiky, high-amplitude primaries that are difficult for lower-resolution spectrophotometers like the I1 family to "see" and report properly. The farther the primaries get pushed out toward DCI P3 and BT. 2020, the worse this gets.
Laser projectors have a similar issue with their primaries.

 

Praying for at least some possibility of even being able to do any calibration I asked this question.

 

Does this limitation of i1 type meters still apply for calibrating at REC 709 with a QD display ?
Form my understanding of what both D_NICE and yourself are saying, the limitations of the i1 type meter with calibrating QD is mainly with individual colours, and especially at the outer ends of the HDR colour gamut.

I was hoping that I could at least measure and calibrate the grey scale and white balance at REC 709, as the White balance carries over to HDR anyway on SONY TV's.

There was a post a while ago where the profile from a high end spectro was put up as a attachment [which I have downloaded] so it could be used with a i1 type meter.
Would this help matters if I were to apply it for use with my new Calibrite meter ?

 

Reply below.😭

It is with everything. You need a 5nm or better spectro. There is no way around it.

 

So that's the state of things

It seems Tony O'Brien of Clarity Calibrations is going to very busy if anybody wants their QD TV calibrated

https://www.claritycalibration.com.au/

 

AVS forum member and ISF calibrator D_NICE usually posts his recommended settings for new TV's at the forum, but there was  a recent run-in and short banning on the forum [I have no idea what it was about], and now reinstated has said he refuses to post those settings for a SONY A95K until he gets a satisfactory explanation on why he was banned.

 

BUT he said he will give those settings to those that ask, which I have now just done and await a reply.

I'll let people know if and when I get them.

 

 

 

  • Like 2
Link to comment
Share on other sites

I had requested D_NICE's calibrated setting s for the SONY A95K via PM at the AVS Forum.

 

He replied that PM's at the AVS are not private, and can be read by all and sundry, so he is only sending his settings out to those that request them via Email.

I used the email info@av-fi.com to contact him with the request, and he just sent them to me.🙂

 

The caveat to him sending them to me were on the understanding that I would not send them to anybody else, or post the settings on any website.

So out of gratitude and respect I will follow his wishes.

 

His calibration settings for the A95K are quite comprehensive, and you would have to have installed the Calman App on your TV to gain the extra two additional Expert 1 and Expert 2 preset options to input them all [Just type in Calman in the Google search in the App section on your TV to get it ]

This also is the only way to open up the 20 point white balance controls on the A95K [The default is 10 point otherwise]

 

The settings included are for.

Custom Pro 1 [SDR]

Custom Pro 2 [SDR]

Custom Pro 2 [HDR]

Dolby Vision [Dark] He has mentioned in other posts that Dolby Vision Dark is the most accurate.

 

I just had a brief look at what the main difference is between the two SDR calibrations, as it's not specified, and I see the brightness setting is much lower for the Custom Pro 1 {SDR] calibration, so I think it's safe to say that the first one is for a darkroom viewing, and the second is for bright room viewing.

 

As anybody that does calibration will tell you, each TV will measure differently, so as a general rule it is not wise to input somebody else's calibration settings, as the results can be vastly different when inputted into your TV.

 

But this case, seeing there is no other option other than hiring a ISF to come around and calibrating your QD-OLED TV, or you forking out well over $11K for a high end spectro to do the job yourself, you might as well give D_NICE's calibration settings a try.....you can always reset the preset if you don't find it better, or if you think it actually makes your TV look worse.

 

I'm unsure if D_NICE is only sending out his calibration settings to AVS forum members, I think he is obliging to anybody that asks.

BUT I do know when he first offered to send therm out to people, he waited until he had quite a few requests, and then he sent the settings out en mass.

 

So if you do politely request his settings, don't be surprised if you have to wait a few days for a reply.

Link to comment
Share on other sites

If anybody wanted to know what sort of OOTB performance is like on a typical SONY A95K TV, DWanye from the AVS forum posted his pre calibration report of the one done for the Value Electronics 4K shootout.

Its not perfect by any means, but its a lot better than you can expect from most TV's.

 

When I first turned on my A95K I had it on the Standard preset.

The Red was very much more present than any other colour, and the pre calibration test shows that to be the same on the A95K he measured.

 

I visited the FlatpanelHD website for their review of the SONY A95K [also has pre and post calibration reports] and copied the basic calibration settings they used in the Advanced colour menu into the TV's Custom preset, this sorted the Red push.

https://www.flatpanelshd.com/review.php?subaction=showfull&id=1655279478#4

1575830823_SONYA95Kprecalibrationreport_page_1.thumb.jpeg.a2927b55b11214fec0c319f1495f7eb5.jpeg570325538_SONYA95Kprecalibationreport_page_2.thumb.jpeg.8e1a12d75d0c1bef589184d21820ffd5.jpeg1774741871_SONYA95Kprecalibrationreport_page_3.thumb.jpeg.086cb340f5f42b87b9d34d06ff948954.jpeg428798817_SONYA95Kprecalibrationreport_page_4.thumb.jpeg.74d1a2faf109b06cebff2d5915e6a4d5.jpeg

 

 

 

 

Edited by Tweaky
Link to comment
Share on other sites

After further discussion about possible work around on calibrating a QD display, particularly  the likes of the SONY A95K or Samsung S95B [applies to their QD LED/LCD displays as well ] lets just say it's been put in the too hard basket, as for a starter you would need a fully calibrated to D65 WOLED, Plasma or CRT display to perceptually match the white balance.

 

On the topic of just how hopeless it would be to try and calibrate a QD display with a i1 type meter, any idea that it could get close enough to do so has now definitely been put in it's casket, the lid put on and firmly nailed down.....it's dead.

Totally forget about it.

It's news I'm unhappy to report, but at least it's better than being lied to and have you in a state of false hope.

 

AVS forum member and calibrator, D-NICE , who gave me a copy of his calibration settings for my 65" SONY A95K, recently posted a set of comparison screen shots of the results of trying to calibrate a A95K with a I1display Pro 2, compared to  the results gotten when calibrating the same TV with a high end CR-300 spectro.

 

The differences are obvious, and clearly show that any attempt at trying to calibrate a QD display with a i1 type meter will, in all likelihood, make your TV look a lot WORSE.

See for yourself.

And the results with the I1Display Pro 2 were of that meter using a profile that was calibrated to the CR-300 spectro.

As you can se from the results, that didn't help matters either,.

 

When you consider that 99% of "At Home Calibrators" who use a I1 type meter, won't have their meters calibrated to a high end spectro, their results will be hopelessly out of whack.

 

The main issue with the SONY A95K at least, is that there is a strong RED push when not using the Custom preset when using OOTB settings.

The settings given at the FlatpanelsHD website I gave a link to in a post above rectifies the worse of this.

 

But you can see below that the i1Display calibration still has this, the meter just isn't up to the task to measure and rectify it.

 

The same can be said of adjusting the black levels in the lowest IRE's, which is another place where the A95K can show problems.

When watching a 4K Disc, this problem in the lowest IRE's show themselves as having little separation between total black and the two IRE's directly above them.

 

I recently bought all 11 of the Star Wars movies on 4K disc's, this lack of separation in the lowest IRE's only really made it's presence known when seeing something like Darth Vader's clothing.

This is using the OOTB settings.

You couldn't see the folds in his all black attire, especially if he was partially in the shadows like he is often seen when inside the Death Star, which is predominantly a Black background.....Not exactly a deal killer, and you'd have to know what to look for to notice it, but once you do know what to look for, it sort of spoils things if you are a perfectionist regarding getting the most out of your TV.

 

I'm sure some calibration setting will eventually be posted for all to use that will fix this , so it's nothing to worry about long term.

D-NICE has usually done this with new TV's, but ATM at least, he is only giving his setting for the A95K if you email him a request for them, email address below.

info@av-fi.com

 

GREY SCALE with CR-300 spectro

272358135_CR-300greyscaleA95K.thumb.jpg.59f96d7b0d0b61a55cc8f3d3e04e2a8a.jpg

 

GREY SCALE with i1Display Pro 2

 

151378550_I1Pro2GreyscaleA95K.thumb.jpg.7ca512a0001d18b98d7f15b05995da85.jpg

 

GAMUT with CE-300

643351860_CR-300GamutA95K.thumb.jpg.55f4c9dc99bf7535123d01a3d833f05a.jpg

 

GAMUT with i1Display Pro 2

216214141_i1Pro2GamutA95K.thumb.jpg.75dbe75545d753d087cd4e8e55df4f38.jpg

Edited by Tweaky
  • Like 1
Link to comment
Share on other sites



Compare how far out D-Nice's results are using the cheap ($400) Colorimeter vs the (15k) Spectro. Amazing.

This is why the Calman for Home Autocal software is getting so many people saying the autocal has made their picture look worse.
 

Take a look at this video by Classy Tech, he shows the variance between the meter profiled with a spectro, vs the meter on its own. Huge.

 

 

 

Even if you have a meter and the Calman for home software you will not get the optimum result.

The issue here is you need a real calibrator with real equipment to calibrate these new tv's, someone like Tony O'Brien.

 

Edited by agisthos
  • Like 1
Link to comment
Share on other sites

Look at the graphs for D-Nice on the Sony A95K, and ClassyTech on the LG C1/G1.

Both calibrators, on differing OLED panel types, saw the idisplay Pro 2 meter pick up too much red and not enough green. It's not just on the WOLED, it seems QD-OLED as well.

 

If you were to continue calibration with the basic colorimeter, you will (wrongly) reduce red and increase green, resulting in a green tint on everything, which is the common complaint with OLED's.

Link to comment
Share on other sites

3 hours ago, agisthos said:

Look at the graphs for D-Nice on the Sony A95K, and ClassyTech on the LG C1/G1.

Both calibrators, on differing OLED panel types, saw the idisplay Pro 2 meter pick up too much red and not enough green. It's not just on the WOLED, it seems QD-OLED as well.

 

If you were to continue calibration with the basic colorimeter, you will (wrongly) reduce red and increase green, resulting in a green tint on everything, which is the common complaint with OLED's.

 

It's especially a problem on QD-OLED's

I'd be very wary of watching any YouTube videos with recommended settings for a QD-OLED other that contrast/brightness and setting up VRR for gaming.

Unless you know they have used a high end spectro, they are going to be way out of whack.

You'd be better off copying the basic calibrated settings from the FlatpanelsHD website and using them.

 

It will be interesting to see just how few YouTube videos giving best settings for both the A95K and S95B show up.

I'm sure a lot of those YouTube guys won't realize what the problem is when they try and calibrate a QD-OLED or any QD TV for that matter, with a i1 type meter, a lot of head scratching I suspect.

Or they might not notice and post them anyway, then everyone's TV will look like a Hisense in Vivid mode.

 

I've never used Calman, and never used Autocal.

I thought about getting Calman years ago,  but when Calman changed from being a calibration software which could be used on any TV or PJ, to one where you had to buy [or should I now say Rent] software for a particular model of TV, I said no bloody way.....you can go the same way as Adobe with your subscription model and take a flying leap.

 

I've always done manual calibration using ChromaPure, and was intending to give HCFR a go, but since the problems with QD-OLED, and me now owning one, there is no point for me.

 

Well unless I fork out for a high end spectro, and that's very unlikely to happen unless there is a very sudden windfall.

 

Edited by Tweaky
Link to comment
Share on other sites

1 hour ago, Tweaky said:

I've always done manual calibration using ChromaPure, and was intending to give HCFR a go, but since the problems with QD-OLED, and me now owning one, there is no point for me.

 

Well unless I fork out for a high end spectro, and that's very unlikely to happen unless there is a very sudden windfall.

 

 

Same here, I use Chromapure and know how to do my own 2-10 point white balance, but that's about it.

 

I have been out of this for a while and only just got back in with an LG G2 4k. Naively I thought you could just plug in the iDisplay, use the cheap Calman autocal for Home, and get a good calibration. That's seems to not be the case. You still need a pro like @Clarity Calibration

 

The only saving grace for me, I purchased the Chromapure+Meter package, so my iDisplay has a custom offset that is profiled against an OLED by Tom Huffman. This is his OLED he has onsite, so it will never be as good as my unique panel, but the corrections should make it better than a stock colorimeter by itself.

 

But you can surely presume the meter was profiled against a WOLED, and not a QD-OLED, so if you have the new Sony or Samsung QD-OLED this correction is probably useless.

Edited by agisthos
Link to comment
Share on other sites

8 minutes ago, agisthos said:

 

Same here, I use Chromapure and know how to do my own 2-10 point white balance, but that's about it.

 

I have been out of this for a while and only just got back in with an LG G2 4k. Naively I thought you could just plug in the iDisplay, use the cheap Calman autocal for Home, and get a good calibration. That's seems to not be the case. You still need a pro like @Clarity Calibration

 

The only saving grace for me, I purchased the Chromapure+Meter package, so my iDisplay has a custom offset that is profiled against an OLED by Tom Huffman. This is his OLED he has onsite, so it will never be as good as my unique panel, but the corrections should make it better than a stock colorimeter by itself.

 

But you can surely presume the meter was profiled against a WOLED, and not a QD-OLED, so if you have the new Sony or Samsung QD-OLED this correction is probably useless.

 

I got a profile that was from a high end spectro that was put as a attachment for use with a 1i type meter in a recent post at the AVS forum.

I haven't bothered to even load it, as I know its not going to be any help to me with QD-OLED due to the narrow, spikey , high amplitude primaries QD displays have, which i1 type meter have a hard time picking up.

And not just QD-OLED's, it applies to QD-LCD as well.

it might be a help with WOLED, but since I don't have one there is no point.

Link to comment
Share on other sites



If you want to try the 10 point white calibration settings posted on RTINGS for the A95K I copied them down, and give them below, it will save you going through all the screen shots and doing it yourself.......They are very conservative settings, and TBH I think they will make a minimal difference, but should help get rid of the worse of the over abundant Red on the A95K.

 

Just read the rest of what they have to say about the other settings you should use in the link.

https://www.rtings.com/tv/reviews/sony/a95k-oled/settings

 

                                        10 Point White calibration  

Basic White settings.                                               Advanced White Balance for 1 to 10

R Gain : Max                                                   1            2           3         4         5          6          7        8         9        10                                  

G- Gain : Max                    Red Offset         -1          -4         -4       -5         -1          -1         0       -1         -1         0

B Gain : -4                         Green Offset      -2         -4         -4       -6         -1          0         1         0         0         0

R Bias  : -2                          Blue Offset        -3         -5          -5       -5          0         -1         0       -1         -1         0       

G Bias : 0

B Bias : 0

 

Per Colour Adjustment ---- they only made changes on two of the colours, the rest stay at 0

                     RED                   BLUE

           HUE    0                         0

           SAT     -1                         -1

Lightness     0                         0

Link to comment
Share on other sites

  • 6 months later...

Hey @Tweaky, thanks for your efforts on this! I also own i1 display pro  and had no issues to calibrate my Panasonic 55 GZ1000. It’s amazing by using Calman. Now it’s time to upgrade for a bigger size and I was thinking about the A95K… well, I really was but after seeing your findings I think I’ll be switching to the A90J. I don’t think it’s same sort of issue due the WOLED panel. Am I right? Thanks mate. 
ps: sorry about the English, I’m not native speaker. 😊👍

Link to comment
Share on other sites

9 hours ago, Marcos said:

Hey @Tweaky, thanks for your efforts on this! I also own i1 display pro  and had no issues to calibrate my Panasonic 55 GZ1000. It’s amazing by using Calman. Now it’s time to upgrade for a bigger size and I was thinking about the A95K… well, I really was but after seeing your findings I think I’ll be switching to the A90J. I don’t think it’s same sort of issue due the WOLED panel. Am I right? Thanks mate. 
ps: sorry about the English, I’m not native speaker. 😊👍

 

The issues calibrating QD-OLED panels with a i1Dispaly type meter doesn't exist with WOLED based ones, although having a better [smaller than 10mm sensor] will obviously get you better results.

 If you are going to calibrate your own SONY TV, I suggest you first read the posts below from the AVS forum calibration section.

 

There are certain things you need to have the TV set at [ you might possibly need to get into the service menu to change a white balance point, especially on a A90J ], as SONY uses JUDD offset white  - not D-65 - mainly comes into play when calibrating for Dolby Vision where it can make a big difference.

https://www.avsforum.com/threads/sony-calibration-overview-for-enthusiasts.3231568/

 

  • Thanks 1
Link to comment
Share on other sites

  • Recently Browsing   0 members

    • No registered users viewing this page.




×
×
  • Create New...
To Top