www.makemkv.com

yes, this should work natively with the 2025 LG OLED TVs but the older TV will require the EDID hack to allow it. And I wouldn't be surprised if it was just LLDV.

Nice, WEBOS-25 on the G5 supports natively DV MKV, but still suffers from the colorspace conversion and YCbCr clipping issues

https://slow.pics/c/IBNzcYOA

The older models might support it soon too, once they get updated to WEBos 25

Where can I find new versions (at least a source) of madMeasureHDR.exe? I've looked for it everywhere on the internet but I haven't been able to find it anywhere.

@RESET_9999 Finally took your advice and got the AM6b+ . What a relief

A question though: On my ancient TV, LG E6, is it worth it to run CPM builds of CoreElec, and if so: Will the built in VS10 on-the-fly VS10 HDR10(+) to DV conversion, or the 3-2 processing from Dovi Scripts give me the best results? (I know the 3-1 processing is superior, but my hardware on my NAS/processing box is old, so I don't want to use 24 hours+, and god knows how much electricity, pr movie)

Epedemic wrote: Mon May 12, 2025 2:51 pm @RESET_9999 Finally took your advice and got the AM6b+ . What a relief

A question though: On my ancient TV, LG E6, is it worth it to run CPM builds of CoreElec, and if so: Will the built in VS10 on-the-fly VS10 HDR10(+) to DV conversion, or the 3-2 processing from Dovi Scripts give me the best results? (I know the 3-1 processing is superior, but my hardware on my NAS/processing box is old, so I don't want to use 24 hours+, and god knows how much electricity, pr movie)

You will love it

Yes, CPM build is a must for the true tv-led option: ''source metadata''. DS workflow 3-2 will be identical to CPM HDR10+ conversion as long as you use the histogram plus.
ps check your pm, I've sent you a link to my GD RPU collection. all you need to do is inject with workflow 1.

LV8HD wrote: Sun May 11, 2025 11:28 am Where can I find new versions (at least a source) of madMeasureHDR.exe? I've looked for it everywhere on the internet but I haven't been able to find it anywhere.

It's included in my tools pack and I don't know where I found it originally.

RESET_9999 wrote: Mon May 12, 2025 4:20 pm
You will love it

Yes, CPM build is a must for the true tv-led option: ''source metadata''. DS workflow 3-2 will be identical to CPM HDR10+ conversion as long as you use the histogram plus.
ps check your pm, I've sent you a link to my GD RPU collection. all you need to do is inject with workflow 1.

Amazing, will definitely give the injection procedure a shot.

Can you verify I set the right settings for my use case in CPM System -> CoreELEC -> Dolby Vision?
Mode: On demand
Type: Display Led (DV-Std)
Dolvy VSVDB: Off
Colimetry: disable/remode
Source levels metadata: On
Level 5 & 6: Source
Restrict overlays: off
Preferred HDR for dual Dolby Vision and HDR10+ content: Dolby Vision
Dynamic HDR10+ to Dolby Vision profile 8.1 conversion: on
Prefer conversion over original Dolby vision: Off
Peak Brightness Source: histogram-plus
For SDR8/10: Off
For HDR10: Dolby Vision
For HLG HDR: Dolby Vision
For Dolby Vision: Off
Convert profile 7: Off

Looks good but for the E6, you will have to use different Level 5 settings if you don't want to hide the OSD and/or subtitles.
For HDR10/HLG, I would not use the player to convert in DV, just inject true metadata or watch in their original form.

Hi Reset

When one bakes the FEL during the encoding process it takes the maxcll and maxfall metadata from the generated RPU.
Should it not take the values shown in the mediainfo file?

Thanks

No because when you bake FEL, it usually increases the brightness over 1000nits and if you keep the original MDL/maxcll, it will be clipped in HDR10 because the TV will use the 1000nits tone mapping curve with a higher roll-off point.

All of this depends on how the TV handles static tone mapping in HDR10 though but generally speaking, it's better to use the same MDL as DV.

RESET_9999 wrote: Tue May 13, 2025 3:06 pm No because when you bake FEL, it usually increases the brightness over 1000nits and if you keep the original MDL/maxcll, it will be clipped in HDR10 because the TV will use the 1000nits tone mapping curve with a higher roll-off point.

All of this depends on how the TV handles static tone mapping in HDR10 though but generally speaking, it's better to use the same MDL as DV.

Thanks for insight

If one watches the baked FEL encode on a non DV capable TV. Will using the maxcll and maxfall from the RPU be still correct?

yes that's what I just said.
The HDR10 metadata doesn't matter for DV TV and FYI, Samsung TVs ignore maxcll, they only use MDL

foxbat wrote: Fri May 02, 2025 2:28 am

RESET_9999 wrote: Thu May 01, 2025 11:14 pm

daffie wrote: Thu May 01, 2025 8:51 pm @RESET

You got that picture of the greyscale 10%?

sorry, the 5-10% grayscales are too dim. My camera cannot capture it... I have to raise ISO, which introduces noise that masks the issue.


EDIT: actually what I'm seeing is exactly like in this video:
https://youtu.be/czIxtnAh36w?si=--UggVf0iEz5VP7u&t=350

For what it's worth I've been watching movies in DV on my 83" G5 for a few weeks now and have not been able to see the dithering effect in any content in the darkest of scenes from my seating distance of 10' away. And believe me I looked.

FYI, the dithering lines are totally visible in real content (dark scenes) and they look horrible at 8-10ft. Check this out, when she closes the door and turns around.
https://mega.nz/folder/YXl32CTT#xvx028tSnyHhmbwqB2AdhA

Not visible on the C2 on this same scene but the C2 also has those moving/falling lines on the 5-10% grayscale.
https://drive.google.com/drive/folders/ ... drive_link

Is it normal for the DV P5 version of a given stream to be around 30% larger than the equivalent HDR10 version?
For example:
Andor.S02E04.2160p.DSNP.WEB-DL.DDPA5.1.DV.HEVC-PlayWEB.mkv: 6431381983
Andor.S02E04.2160p.DSNP.WEB-DL.DDPA5.1.HDR.HEVC-PlayWEB.mkv: 4305241410

I also notice that the hybrid P8 files people are posting are around the same size as the HDR10 versions. Is something being lost from the P5 files during that conversion?

DV was just encoded at a higher bitrate. 15mb/s vs 11mb/s
Get the higher bitrate version, P5 basically like FEL DV anyway. So yeah, there is a loss in the P5 to P8 conversion.

how is the DV performance of jellyfin? is it incorrectly?