Display Calibration

[huhn]

Originally Posted by Nullack

Well you seem to be the only one saying that. Everywhere else I look, is pluge patterns to say 16D blackness and then ones specifically different coded to RGB where zero is the black with no blacker than black unlike the 16D TV style levels ones. Please explain.

a limited range source is rendered at full range so i can't follow you.

[foxyshadis]

Nullack, how many times do you have to be told to ignore old Studio RGB standards? Every video codec/renderer decodes to 0-255 RGB, your monitor displays 0-255 RGB, that's all that should matter. 16D isn't even a real term.

[Nullack]

The simple fact is, not everyone agrees with Foxyshadis. I would hope you to have the maturity to accept that your own theories may not be shared universally by all. Many people particularly on the consumer electronics side consider TV level calibrations to be the right way to go on any TV, including digital ones being fed digital signals from a HTPC.

However foxyshadis if you bothered to actually read my post, you would see *Im trying to do PC level* calibrations. Even a most cursory look at my post would have given you that. Instead of going off on an unrelated point, how about addressing what I actually said, which is about finding PC level 4K calibration test patterns.....

[huhn]

sorry the issue here is that you don't understand calibration

[Nullack]

Originally Posted by huhn

.
same for resolution.

So you have invented a perfect scaler then and can miraculously do test patterns like sharpness in an optimal way even when you have to scale the test patterns to have it display at native resolution

[Nullack]

Originally Posted by huhn

sorry the issue here is that you don't understand calibration

If you believe that, explain precisely where you think Im wrong. I do believe I do understand it.

[huhn]

Originally Posted by Nullack

So you have invented a perfect scaler then and can miraculously do test patterns like sharpness in an optimal way even when you have to scale the test patterns to have it display at native resolution

black boarders?

[Nullack]

Huhn I know you have allot of video knowledge and I'm thankful for insights you have offered in the past In this case though, I really dont think you get whats driving what Im asking and sometimes I get the impression english isnt your common language. Ill try to explain. The thing about test patterns is they need to be "reference quality". They need to be verified they are correct - for example the ISF have a process for this and they explained it when they published some ISF test patterns. As you know there is allot of different test patterns that do different things. The reason I want 4K test patterns in particular is to send *native" resolution images to the *native* resolution of the display so as to avoid any non reference artifacts in the image. I gave the example of sharpness tests. Those sorts of test patterns are useful in native resolution because of how fine the detail gets with testing sharpness. As you'd know, 4K is allot more pixels than full HD. The value seems obvious to me in these sorts of cases to be able to do native resolution test patterns.

Im not arrogant and I do listen to contrary arguments that are rational. If indeed your convinced I'm wrong about this topic, then please explain why you think so. Just saying that I dont know doesnt contribute anything of substance to the discussion.

[huhn]

like I already said levels doesn't matter if you 100% know how they work.

that's mean you know the render/desktop side, the GPU side and the TV side of "levels".

just a small example. if you set the GPU to YCbCr or limited range RGB than you will not get information in the WTW and BTB parts no matter what.

if you take a 1080p test image and want to use it at an UHD screen than you use black boarders for that (zoom 100 %). the image in the middle is still bit perfect.

works with http://www.lagom.nl/lcd-test/white.php
and works with these test pattern: http://www.avsforum.com/forum/139-di...libration.html

[Nullack]

Originally Posted by huhn

if you take a 1080p test image and want to use it at an UHD screen than you use black boarders for that (zoom 100 %). the image in the middle is still bit perfect.

Ok I see. Just dont scale it when using test patterns lower than the native resolution. While I realise that would work, I would prefer to deal with things like sharpness at full screen so I can carefully try to tune into the reference standard. Thanks

[huhn]

NN x2 scaling should work for most test patterns too.

a good sharpness test picture works like this: http://www.lagom.nl/lcd-test/sharpness.php

so a smaller picture helps a lot to do this test.
this test for example isn't affected really by NN x2 scaling.

[chros]

So, I have finished playing with profiling, so I try to summarize what I understood so far. And thank all of you for the great help.

Originally Posted by Asmodian

If you need the black point hack you almost always get crushed shadows.

Probably that's right if you use dispcal.exe at the first step (see below for alternative way), at least it worked like this on my setup.

Originally Posted by chros

About the black crush: I noticed that the cause of it is the "-G2.4 -f0 -k0" switches in the dispcal command. If I leave it out and '-b' switch is also present I won't get black crush.

Originally Posted by chros

b) with dispalGUI the situation is pretty interesting, it uses different method: setting measured (as target) everywhere (the full log is here: http://pastebin.com/nPrEspS1 )
here are the commands (timecodes are included):

00:39:30,080 dispread.exe -F "-P0.5,0.5,9.486833" -v -dmadvr -c1 -yn -X CCFLFamily_07Feb11.ccss -K "madVR 2015-12-06 00-38 S XYZLUT.cal" "madVR 2015-12-06 00-38 S XYZLUT"
00:42:46,055 colprof.exe -v -qm -ax -bn "madVR 2015-12-06 00-38 S XYZLUT"

And that's pretty interesting indeed, if you leave the tone curve setting at 'measured' on Calibration tab in dispcalGUI then it won't use dispcal.exe at the first step but instead:
- it will use dispread.exe with predefined (!) .cal (calibration) and ti1 (testchart with 154 patches) to create a measured .ti3 file
- then colprof.exe will provide the necessary .icm file from the above .ti3 file to be able to use targen.exe in the next step
As I reported, I don't have any problems with this method. So I went for this.

Originally Posted by chros

(-i switch) Asmodian used -ims , dispalGUI offers -iaw by default. The difference is the -ims is relative, while -iaw is absolute. They state that -iaw supposed to be more accurate (I didn't had the time to test it on my setup though).

So, I tried out all of them: only the types that belong to the absolute group produced acceptable results and I couldn't really see the difference between them. For example -ims , -is switches produced oversaturated colors that was so visible as well. So I settled down with the -iaw switch (Absolute with white point scaling).

Originally Posted by e-t172

In fact, that's one of the reasons why BT.1886 was introduced in the first place (i.e. achieve correct results in displays with imperfect black).

Originally Posted by markanini

I wonder what implications -IB:x.x:2.4 switch or output offset slider has for stuff level adjustments etc. I assume defaults are the most accurate ...

Originally Posted by huhn

yes. BT 1886 is currently the correct "SPEC".

Originally Posted by huhn

no BT 1886 it does the rest by it self that's why we have it now.

I would argue about this. This is the reason why:

Originally Posted by huhn

in the end gamma depends on your ambient light. this is for bat-caves.
in a day light flooded room you are not going to like a gamma of 2.4

Originally Posted by huhn

so judging bt 1886 doesn't work without the screen.

These are the most important facts.

Originally Posted by Arm3nian

I targeted a specific white level and white point to help me set it as desired, then went back and changed to as measured. The windows calibration looks great. From that, I made the 3D lut, and it looks good in movies. I'm guessing it is better to create a new profile for madVR and target Rec. 1866 instead of 2.2 gamma? It feels weird for the tone curve to be gamma 2.2, and then the 3D lut to be set to Rec. 709?

Originally Posted by seiyafan

I see, I have an EIZO IPS with 1000:1 contrast ratio, but nowhere near as black as any OLED. Is it still okay to go with BT 1886?

Originally Posted by markanini

I wonder what implications -IB:x.x:2.4 switch or output offset slider has for stuff level adjustments etc. I assume defaults are the most accurate but I too have found the shadows to be somewhat lifted on my IPS panels so I'm tempted to try reducing that now.

If someone hasn't seen it yet, I made a post about gamma here: http://www.avsforum.com/forum/139-di...l#post40108362

[Arm3nian]

Is it possible to get non oversaturated colors on a high gamut display using dispcal and a color managed program?
For example, if I choose sRGB tone curve, and create an icc profile and import it in firefox, will colors look normal?

I still need a monitor (returned the pg279q lol). 4k bluray is coming in less than a month and is has a larger color space and is 10bit. The new dell ultrasharp's have ~98% dci-p3 and are cheap ($700) for the supposed quality you get; gb-led, 10bit, 1440p etc. They currently seem like a good purchase.

What about madVR? Do you need a 3Dlut for proper color space conversions, i.e rec. 2020 to .709? What about 4k bluray on a display that has high dci-p3 coverage, works by default?

[Siso]

Is madvr yCMS calibration option useful for sRGB monitors?

[Asmodian]

Originally Posted by Arm3nian

Is it possible to get non oversaturated colors on a high gamut display using dispcal and a color managed program?
For example, if I choose sRGB tone curve, and create an icc profile and import it in firefox, will colors look normal?

I still need a monitor (returned the pg279q lol). 4k bluray is coming in less than a month and is has a larger color space and is 10bit. The new dell ultrasharp's have ~98% dci-p3 and are cheap ($700) for the supposed quality you get; gb-led, 10bit, 1440p etc. They currently seem like a good purchase.

What about madVR? Do you need a 3Dlut for proper color space conversions, i.e rec. 2020 to .709? What about 4k bluray on a display that has high dci-p3 coverage, works by default?

Yes, it is possible to get non oversaturated color on a high gamut display as you stated. A lot of applications that use ICC profiles only use the simple 1DLUTs (gamma ramps) for calibration which does not correct the gamut but Firefox, Chrome, Photoshop, etc. will.

For madVR you only need a 3DLUT if you want to use your display's true gamut but if your display's native gamut is close to one of the options in "this display is already calibrated" madVR does the proper conversion from the source gamut to the display's gamut.

Options available:
BT.709
SMPTE C
EBU/PAL
BT.2020
DCI-P3

The one important option missing is AdobeRGB but for those new ultrasharps DCI-P3 would probably work pretty well.

The biggest issue I have with any wide gamut display is streaming video, I have never found a method to correct streaming video.

Originally Posted by Siso

Is madvr yCMS calibration option useful for sRGB monitors?

Not really IMO, only if you aren't really very close to sRGB.

[Siso]

Originally Posted by Asmodian

Yes, it is possible to get non oversaturated color on a high gamut display as you stated. A lot of applications that use ICC profiles only use the simple 1DLUTs (gamma ramps) for calibration which does not correct the gamut but Firefox, Chrome, Photoshop, etc. will.

For madVR you only need a 3DLUT if you want to use your display's true gamut but if your display's native gamut is close to one of the options in "this display is already calibrated" madVR does the proper conversion from the source gamut to the display's gamut.

Options available:
BT.709
SMPTE C
EBU/PAL
BT.2020
DCI-P3

The one important option missing is AdobeRGB but for those new ultrasharps DCI-P3 would probably work pretty well.

The biggest issue I have with any wide gamut display is streaming video, I have never found a method to correct streaming video.



Not really IMO, only if you aren't really very close to sRGB.

I see, thank you for the answer, I'll stick with 3DLUT

[Arm3nian]

Originally Posted by Asmodian

Yes, it is possible to get non oversaturated color on a high gamut display as you stated. A lot of applications that use ICC profiles only use the simple 1DLUTs (gamma ramps) for calibration which does not correct the gamut but Firefox, Chrome, Photoshop, etc. will.

For madVR you only need a 3DLUT if you want to use your display's true gamut but if your display's native gamut is close to one of the options in "this display is already calibrated" madVR does the proper conversion from the source gamut to the display's gamut.

Options available:
BT.709
SMPTE C
EBU/PAL
BT.2020
DCI-P3

The one important option missing is AdobeRGB but for those new ultrasharps DCI-P3 would probably work pretty well.

The biggest issue I have with any wide gamut display is streaming video, I have never found a method to correct streaming video.

I see... Why doesn't windows have proper color management by now

So unless your monitor is in Photoshop 24/7, there really isn't a point in wide gamut. I guess you could get accurate colors on the web, but content is basically all sRGB, so you're just wasting money. With UHD and madVR you could benefit, but a higher contrast ratio is probably more important than a larger color space, which rules ips panels out for viewing, as opposed to creating.

[Asmodian]

Originally Posted by Arm3nian

I see... Why doesn't windows have proper color management by now

So unless your monitor is in Photoshop 24/7, there really isn't a point in wide gamut.

I agree that it is very annoying that Windows still lacks the ability to do gamut conversion.

I would even go so far as to say that there are large negatives to a wide gamut display unless your monitor is in Photoshop 24/7.

A DCI-P3 OLED for movie watching with madVR would still be quite enticing, assuming I could get any content in wide gamuts.

[Arm3nian]

I got my UP2716d today (ordered it before our discussion on wide gamut).

The first thing that came to mind was wow, vibrant colors, not over saturated and bad looking. It looks a bit like a samsung phone, which imo looks the best out of all phones screens. This is the best looking display I've seen, and it's not like I paid 2,000 for it, only $700, which is less than my previous monitors. It comes default in adobeRGB color space. sRGB looks very faded, I wouldn't use it, unless I calibrate it. But I don't think I need to use it, because adobeRGB looks excellent. DCI-P3 mode has very nice colors, but it's dim (obviously for movies), but I don't have any content in DCI-P3 yet. UHD movies come in less than a month I think. I have not calibrated it yet, but ran a color test chart in dispcal and I see colors I've never seen before. They don't look incorrect.

I can select 10bpc in NVCP with my 980ti. Do you know how to use madTPG for normal calibrations, or is it not good to do that. IIRC madTPG has 10bit output?

[Arm3nian]

I've found something a bit weird. The monitor on default is in "standard" mode. Everything looks great, and dispcal shows 100%sRGB and 99.7% adobe RGB and 93% dci-p3. When I switch from standard mode to color space, and select adobe RGB, reds turn orange... sRGB emulation mode also looks makes reds look orange. wtf lol