Screenshot Comparisons 4K

[jh1123]

Really frustrated about this. For some reason the HDR has completely changed the color of the flare on the left. I know it was red theaters, and can't imagine WB completely changing it's color for HDR. It has to be my display/settings, since I hear it shows up red for other people. I've tried everything.. but nothing seems to be correcting it....

This is playing off an LG EF9500 OLED


IMG_4036.jpg
IMG_4035.jpg

[ray0414]

Originally Posted by jh1123

Really frustrated about this. For some reason the HDR has completely changed the color of the flare on the left. I know it was red theaters, and can't imagine WB completely changing it's color for HDR. It has to be my display/settings, since I hear it shows up red for other people. I've tried everything.. but nothing seems to be correcting it....

This is playing off an LG EF9500 OLED

what are your UHD color settings on the tv?

[Opips3]

Wow! Difference picture color.

Interesting... I watched three times BDs. I love the movie.

[Blu-21]

Originally Posted by jaaguir

I think the opposite, as someone else pointed out, there is more detail on the UDH, plus it looks very smooth and natural, whereas the BD looks so "electronic", harsh, almost EEd. (Blu-21's post was about the one shot 4kstream/BD "Salt" comparison)

I see what you mean about the control panel, but I was focusing on the hand. The BD has more grain/noise which looks like it carries more detail.

Originally Posted by I KEEL YOU

The 4K rip has compression issues, but it definitely has more detail on those digital numbers, where you can see each dot that makes up the digital numbers clearly.

However, that difference becomes almost invisible as soon as I move my head away 2-3 feet away from the screen.

Sure, but the hand looks more detailed to me on the BD IMO.

[James Freeman]

Originally Posted by jh1123

Really frustrated about this. For some reason the HDR has completely changed the color of the flare on the left.

This is predictable, you can't compare two different masters.

The regular BD mastered on a rec.709 100nit 2.2 Gamma display, while the UHD-BD mastered on DCI-P3 1,000nit ST.2084 display.

It was the colorist intention to under-saturate the red color because what he saw on his DCI-P3 gamut display looked saturated enough for the red flare.
What you see on your display might not be red enough because your display does not cover 100% DCI-P3 (few are) like the professional monitor the colorist used.

Do you actually know what gamut your display can cover?

Here are the calibration reports from my display:
*The dotted line is the desired color space while the colored line is what my display is capable of.

709


DCI-P3


2020


My main viewing display is Dell U2410 wide gamut computer monitor, calibrated to 3DLUT together with a rendered called madVR.
It can quite nicely cover DCI-P3 but lack some greens and yellows.

[cheez avenger]

Looks like UHD will need a "video essentials" disc of its own for folks to properly calibrate their sets.

[pawel86ck]

Even on this screenshot BD has better textures thanks to better compression
http://screenshotcomparison.com/comparison/164864

It really looks like DNR or something, but I only can see it with 100% zoom on my fullhd monitor. With downscaled picture I cant see it, and everything is sharp.

And here's Lego comparison, I really like how vibrant color look on UHD screenshot
http://screenshotcomparison.com/comparison/164888

[dvdmike]

Originally Posted by cheez avenger

Looks like UHD will need a "video essentials" disc of its own for folks to properly calibrate their sets.

Won't work, not all sets use the same HDR format neither do the players or the discs

[James Freeman]

The standard exists but the results are all over the place.
Anyone actually have an idea (measurement) what gamut your TV is capable of?

We really do need a calibration/test disc and a calibration device to test what the current TV is capable of.
I would like to see some real data with i1 Display Pro (or a spectrophotometer) and HCFR.

There are a lot of questions unanswered.
1. How (if at all) the player actually knows the xyY Chromaticity coordinates of the TV?
2. Who and How does the color conversion from rec.2020 on the UHD-BD disc to the gamut of the TV?
3. What colorimetric intent is used to map the gamut (or is it just clipped)?

Something to think about:
The TV may just signal the Player "I'm a DCI-P3 capable TV" and the player will remap the rec.2020 to DCI-P3, but the TV actually nowhere near DCI-P3 but more like 75% or 90% of it.
Do you think the colors will be off?.. They most certainly will!
I can bet that the player has absolutely no clue of the TVs actual xyY coordinates, just a simple 709/DCI-P3/wide EDID "words" and nothing more.

Look here: http://www.rtings.com/tv/tests/movie...ci-p3-rec-2020
None of them actually covers DCI-P3 100%.

[dvdmike]

Originally Posted by James Freeman

The standard exists but the results are all over the place.
Anyone actually have an idea (measurement) what gamut your TV is capable of?

We really do need a calibration/test disc and a calibration device to test what the current TV is capable of.
I would like to see some real data with i1 Display Pro (or a spectrophotometer) and HCFR.

There are a lot of questions unanswered.
1. How (if at all) the player actually knows the xyY Chromaticity coordinates of the TV?
2. Who and How does the color conversion from rec.2020 on the UHD-BD disc to the gamut of the TV?
3. What colorimetric intent is used to map the gamut (or is it just clipped)?

Standards, Phillips? Dolby? The standard HDR?
The TV sets all have diffrent ones and that's by set, not even by make

[doctor_who]

Someone please explain to me how a screenshot comparison is supposed to be effective on my computer screen when my computer screen does not display close to 4k. I just don't understand. Lol.

[scorpiontail60]

Originally Posted by doctor_who

Someone please explain to me how a screenshot comparison is supposed to be effective on my computer screen when my computer screen does not display close to 4k. I just don't understand. Lol.

Get a proper computer screen?

I use my Samsung 65JS9500 as my computer monitor and these screenshots look fine!

[James Freeman]

Originally Posted by dvdmike

Standards, Phillips? Dolby? The standard HDR?

SMPTE standards.

The UltraHD Premium brand TVs are not even here but they are should do "More than 90% of DCI-P3 colours" to qualify.

I believe most screenshots we have seen till now, the Player converts to rec.709 that is why the colors actually are close to the same saturation when comparing screen photos of the BD with UHD-BD.
I think that none of the TVs actually called for DCI-P3 as of yet.

[dvdmike]

Originally Posted by James Freeman

SMPTE standards.

The UltraHD Premium brand TVs are not even here but they are should do "More than 90% of DCI-P3 colours" to qualify.

I believe most screenshots we have seen till now, the Player converts to rec.709 that is why the colors actually are close to the same saturation when comparing screen photos of the BD with UHD-BD.
I think that none of the TVs actually called for DCI-P3 as of yet.

Yeah. No

[pawel86ck]

Originally Posted by doctor_who

Someone please explain to me how a screenshot comparison is supposed to be effective on my computer screen when my computer screen does not display close to 4k. I just don't understand. Lol.

Just use zoom feature. Pictures will be cropped, but with 1:1 scaling.

[jh1123]

Originally Posted by Opips3

Wow! Difference picture color.

Interesting... I watched three times BDs. I love the movie.

The color is striking at times, but I got the sense that something was 'off' with the UHD version. The change in color of the flare confirmed that something had been drastically changed. Personally, I'm sticking with the Blu Ray.

Originally Posted by ray0414

what are your UHD color settings on the tv?

Deep Color On
Gamut set to Wide
Color 50
Tint 0
Gamma 2.2
Brightness 50

UHD Player
4:4:4
Deep Color Off

Originally Posted by James Freeman

This is predictable, you can't compare two different masters.

The regular BD mastered on a rec.709 100nit 2.2 Gamma display, while the UHD-BD mastered on DCI-P3 1,000nit ST.2084 display.

It was the colorist intention to under-saturate the red color because what he saw on his DCI-P3 gamut display looked saturated enough for the red flare.
What you see on your display might not be red enough because your display does not cover 100% DCI-P3 (few are) like the professional monitor the colorist used.

Do you actually know what gamut your display can cover?

Here are the calibration reports from my display:
*The dotted line is the desired color space while the colored line is what my display is capable of.

709


DCI-P3


2020


My main viewing display is Dell U2410 wide gamut computer monitor, calibrated to 3DLUT together with a rendered called madVR.
It can quite nicely cover DCI-P3 but lack some greens and yellows.

I know the EF9500 doesn't meet the minimum specs. By how much, I don't recall (15-20% maybe). I think that's the reason for what I'm seeing, but at the same time, people with different TV brands are reporting the same images for Mad Max. But I don't know how their TVs are at HDR either.

[Geoff D]

Originally Posted by dvdmike

Standards, Phillips? Dolby? The standard HDR?
The TV sets all have diffrent ones and that's by set, not even by make

Sure, but a UHD test disc mastered with the mandatory HDR10 would be a start, no? And having primary/secondary colour slides would at least enable people to check whether they're hitting the correct chromaticities for P3 and 2020. [edit] And given the discrepancies that people are seeing with Fury Road, just calibrating for the colour alone would help to put a lot of questions to rest IMO.

TVs will then do their own thing when it comes to HDR performance, especially the HDR 'lite' TVs that are compatible with the signal but don't actually do real HDR, BUT if people can start collating that sort of data and put it into correction matrices for popular colourimeters - which is what's done at the moment with the various types of display tech - then we might start getting somewhere.

I think there's light at the end of the calibration tunnel for sure, but it's gonna take time.

[Geoff D]

Originally Posted by James Freeman

The standard exists but the results are all over the place.
Anyone actually have an idea (measurement) what gamut your TV is capable of?

We really do need a calibration/test disc and a calibration device to test what the current TV is capable of.
I would like to see some real data with i1 Display Pro (or a spectrophotometer) and HCFR.

People can do that now, HCFR can check for P3 coverage and even has an HDR setting now.

Here's something I pulled from my own Sony KD55X9005B TV re: P3 coverage using HCFR and a Colormunki Display meter (basically a badged I1D3 with my own TV-specific correction matrix applied using a spectro).

It's useless in real world terms (much like my posts in this 4K section of the forum! ) because the TV can only process SDR 709 4K signals, but this is what Sony built in anyway (measured from the service menu colour bars which bypasses the internal processing):

[James Freeman]

Thanks Geoff!

I am yet to receive a good answer of How, Where & Who does the range and color conversion, and how accurate they are.
Today every other TV has its own DCI-P3 % of gamut coverage as the technology continues to develop, how in the world the Player knows what the xyY coordinates to properly map the color to that specific TV?

Originally Posted by Geoff D

I think there's light at the end of the calibration tunnel for sure, but it's gonna take time.

I think we're eating a half baked pie here.
Well it's not better (much worse) than the pie we were eating back in 2006 with the first BD releases.

I highly doubt that a commercial product like a TV uses professional 3DLut to map DCI-P3 that the player sends to its own capabilities.

[Geoff D]

Until even basic calibration tools are in place, everyone's flying blind James. Makes me wonder whether something like Fury Road really *was* graded very differently for UHD (so that red becomes orange), or whether the primary/secondary colour points in people's displays are badly skewed from where they should be.

And I'm not blaming anyone here for that, as we've got the software and the metering hardware to be able to check for P3 coverage at the least, but without a set of properly mastered test signals that can be utilised with the UHD hardware then it's all for nought. Hopefully Spears and Munsil or DVE can come up with something sooner rather than later.

[edit] And not even the almighty UHDA 'Premium' certification will guarantee accuracy out of the box, all it will do is state that x TV can hit x target of their specs. Just like now, sets will still need specific fine-tuning in order to ensure that they're displaying the gamut correctly, amongst other things.