HDR heatmaps and torch mode UHDs

[Amano]

Well, the European SC disc of Cliffhanger would be a welcome comparison (real 4k from the OCN and not so hot graded), but I doubt that dorian has that available.

Big thanks for your work though, Dorian, you are my hero. It was soo sad that pyoko went AWOL.

[Macatouille]

Thelma and Louise (1991)

Mastering display luminance              : min: 0.0001 cd/m2, max: 1000 cd/m2
Maximum Content Light Level              : 841 cd/m2
Maximum Frame-Average Light Level        : 311 cd/m2

Heatmaps

[Show spoiler]

Gamut Visualizations

[Show spoiler]

HDR10 Plot

[Macatouille]

Ignoring the technical elements, I just genuinely like how this shot looks in heatmap form:

[Modren]

Originally Posted by Macatouille

Ignoring the technical elements, I just genuinely like how this shot looks in heatmap form:

Some strong Fury Road vibes here

[Macatouille]

Road House (1989)

Mastering display luminance              : min: 0.0001 cd/m2, max: 1000 cd/m2

Heatmaps

[Show spoiler]

Gamut Visualizations

[Show spoiler]

HDR10 Plot

[Fjodor2000]

@Macatoullie & @dorian:

Pyoko used to have quotes like this after his analyses:

With a similar highlight roll-off the HDR version would on average be equivalent to the SDR version with peak white at 150 nits (Gamma at 2.40), based on 20 samples. (Min: 41, Max: 513, Median: 126)

Example above from:

https://forum.blu-ray.com/showpost.p...91&postcount=3

Any idea how those values were calculated and if you could add it?

[dorian]

SDR Heatmap Colab Notebook

Instructions are the same as for the HDR Heatmap notebook. Expects 1080p screenshots and uses a nominal 100 nit peak. Honestly not sure of how much use this will be but it's here for anyone interested.

Here's an example comparing the 2017 blu-ray of Heat to the UHD:

[Show spoiler]
Blu-ray


UHD


Blu-ray


UHD


Blu-ray


UHD

[Macatouille]

Apocalypse Now (1979)
Final Cut

Mastering display luminance              : min: 0.0010 cd/m2, max: 1000 cd/m2
Maximum Content Light Level              : 1000 cd/m2
Maximum Frame-Average Light Level        : 462 cd/m2

Heatmaps

[Show spoiler]

Gamut Visualizations

[Show spoiler]

HDR10 Plot

[Fjodor2000]

Thanks a lot! One of my favorite movies of all time.

Would it be possible to get some WCG analysis too? I'd like to see if it's within DCI-P3 only, or also stretches into Rec.2020.

[Mierzwiak]

It's funny how night scenes in Apocalypse Now are brighter than daylight scenes in Heat.

[Macatouille]

Back to the Future (1985)

Mastering display luminance              : min: 0.0050 cd/m2, max: 4000 cd/m2
Maximum Content Light Level              : 1000 cd/m2
Maximum Frame-Average Light Level        : 755 cd/m2

Heatmaps

[Show spoiler]

Gamut Visualizations

[Show spoiler]

HDR10 Plot

[Macatouille]

Originally Posted by Fjodor2000

Would it be possible to get some WCG analysis too? I'd like to see if it's within DCI-P3 only, or also stretches into Rec.2020.

Talking to dorian, it sounds like it's not possible to do in workbook form yet. But if that changes or if I get a way to do it locally, I'll gladly add them to the posts.

[Fjodor2000]

Originally Posted by Macatouille

Talking to dorian, it sounds like it's not possible to do in workbook form yet. But if that changes or if I get a way to do it locally, I'll gladly add them to the posts.

But isn't it possible to do in the same way as dorian did previously? I.e. like this:

https://forum.blu-ray.com/showpost.p...&postcount=493

and also here:

https://forum.blu-ray.com/showpost.p...&postcount=495

[Macatouille]

Originally Posted by Fjodor2000

But isn't it possible to do in the same way as dorian did previously? I.e. like this:

https://forum.blu-ray.com/showpost.p...&postcount=493

and also here:

https://forum.blu-ray.com/showpost.p...&postcount=495

Possible to do locally with the right software (which I don't currently have). Not possible, at the moment, to set it up in a Google workbook so that anyone can do it with a couple clicks. But I'm keeping all my raw screenshots for posterity.

[Macatouille]

Back to the Future Part II (1989)

Mastering display luminance              : min: 0.0050 cd/m2, max: 4000 cd/m2
Maximum Content Light Level              : 1000 cd/m2
Maximum Frame-Average Light Level        : 606 cd/m2

Heatmaps

[Show spoiler]

Gamut Visualizations

[Show spoiler]

HDR10 Plot

[dorian]

Finally found a workaround for the gamut visualisation notebook. The instructions are as per the HDR & SDR heatmap notebooks. Works with both 8-bit and 16-bit ST 2084 encoded screenshots that haven't been tone mapped. The script isn't very fast but should function as intended.

Gamut Visualisation Notebook

HDR Heatmap Notebook

SDR Heatmap Notebook

[Macatouille]

Originally Posted by dorian

Finally found a workaround for the gamut visualisation notebook. The instructions are as per the HDR & SDR heatmap notebooks. Works with both 8-bit and 16-bit ST 2084 encoded screenshots that haven't been tone mapped. The script isn't very fast but should function as intended.

Gamut Visualisation Notebook

HDR Heatmap Notebook

SDR Heatmap Notebook

Submitted my Back to the Future Part II screenshots through the new notebook. Results were pretty interesting:









Will add a few to my last post. In the future I might just post any shots that breach 709, given how big the images are. Might be worth outputting these at 1080p as well?

[Macatouille]

Originally Posted by Fjodor2000

Thanks a lot! One of my favorite movies of all time.

Would it be possible to get some WCG analysis too? I'd like to see if it's within DCI-P3 only, or also stretches into Rec.2020.

Added some gamut visualizations to my Apocalyse Now post above.

[Fjodor2000]

Originally Posted by Macatouille

Added some gamut visualizations to my Apocalyse Now post above.

Thanks a lot! Looking at those gamut visualizations, quite a lot of them stretch into DCI-P3, so an improvement over BD in that regard. Some also stretch into Rec.2020, but a bit less.

[Fjodor2000]

Just wondering, is the DCI-P3 triangle defined in the same way as in Pyoko's analyses? The area towards the border of the Rec.2020 triangle looks very narrow in the area marked by the red rectangle here:



Compared to the triangles from Pyoko: