Dolby Vision now possible through MP4 Mux.

[RESET_9999]

Hello @RESET_9999

My setup is the same as yours (x800m2 direct to the C2) but I am facing quite a strange scenario with two different DV movies.

Yesterday i was watching the new Wonka movie in the internal player of the C2 because the movie is cropped (3840x1604). Imediately i noticed that the movie was dimmer than normal levels. When i opened the same file in the x800m2, i noticed it was playing with a brighter image, albeit with grey bars.

Did some cmv4.0 test files for Wonka:

cmv4.0 removed P5: https://drive.google.com/file/d/1cbEXSO ... sp=sharing
cmv4.0 removed P8: https://drive.google.com/file/d/1t-xFto ... drive_link
cmv4.0 original P5: https://drive.google.com/file/d/1ZAUusI ... drive_link
cmv4.0 original P8: https://drive.google.com/file/d/1CRDuTM ... drive_link

I'm not home and can't test on my C2 but remotely with teamviewer, and using the Windows Energy player, I can see how the CMV4.0 without the L2 positive lift is noticeably darker.
I expect the same on my C2.
see this, CMV2.9 with trims vs CMV4.0: https://slow.pics/c/johVT1ao
and now CMV2.9 without trims vs CMV4.0: https://slow.pics/c/TKTvPkau

[valington]

I have added DoVi section to all the available settings docs.
VSVDB is the EDID value.

that's great info. thanks a lot man.

Reset i m doing BDMV DOLBY VISION from mkv 7profile dolby vision with your script , is it true dolby vision which i got in bdmv with your script ?

[azreil24]

So far I've only watched movies/TV with EAC3 audio and it is always in sync but last night I watched something with AC3 audio and the audio is indeed OOS 300-400MS. Tried other AC3 files and got the same problem then converted the audio to EAC3 and it fixed the issue.

Gotcha. I will test when I get home in a few days. Thanks for the find. Weird thing is that trying to re-sync the audio in Kodi doesn't help.

[RESET_9999]

Reset i m doing BDMV DOLBY VISION from mkv 7profile dolby vision with your script , is it true dolby vision which i got in bdmv with your script ?

If you demux ST P7 mkv and then mux to BDMV the BL and EL, yes.
If you just create a BDMV without demuxing the layer from a ST-P7, probably not?

[valington]

Reset i m doing BDMV DOLBY VISION from mkv 7profile dolby vision with your script , is it true dolby vision which i got in bdmv with your script ?

If you demux ST P7 mkv and then mux to BDMV the BL and EL, yes.
If you just create a BDMV without demuxing the layer from a ST-P7, probably not?

Yes i m demux st p7 mkv and then mux to bdmv the bl and el , thank you RESET

[RESET_9999]

Ok, I think I know what's going on with CMV4.0.

First, Dolby confirmed that L8 is not generated and must fully come from the colorist and only L2 is generated automatically and gets updated through L8 to L2 math. And 600-100nits L8 are not needed/requested most of the time.

Secondly, the 100nits L8 has an effect during DV playback and produce a similar/close effect and image to CMV2.9 with L2 trims. On Windows Energy player or movies and tv app, in HDR10 the active 100nits L8 trim always has an effect but on the C2, the 100nits L8 trim only has an effect if it's alone (without L8 600nits or 1000nits trim). Same behavior in cmv2.9. L2 100nits alone is used but when there are trims, it is not used on the C2.

So in Wonka, there is no active L8 100nit trim and the image is noticeably darker on cmv4.0 device than on 2.9 devices when there is L2 trims.
In Aquaman, there is an active L8 100nit trim and the image looks similar on both 4.0 and 2.9 devices.

So perhaps, Wonka and the ones without 100nits L8 were done in cmv2.9 and upconverted to 4.0 for streaming requirements? or maybe the colorist just was lazy and did not make trims?? idk

I've added the test files in my googledrive: https://drive.google.com/drive/u/1/fold ... fx6vRYDEfs

[speeddemon]

So perhaps, Wonka and the ones without 100nits L8 were done in cmv2.9 and upconverted to 4.0 for streaming requirements? or maybe the colorist just was lazy and did not make trims?? idk

Are you seeing a similarity in the studios that are leaving out the 100-nit L8 Trim?

I've been suspecting that it's not an accident that CMv4 produced movies only include an L8 Trim, and that Dolby is telling studios that CMv4 no longer requires separate trims at various luminance levels. I think that some of the "laziness" we're seeing may even be originating from Dolby themselves, or possibly through miscommunications between Dolby and the studios.

[chros]

I have added DoVi section to all the available settings docs.
VSVDB is the EDID value.

Were those values retrieved using bscpylgtv command or by extracting the default DV config files from the C2 TV firmware?
As discussed prior, I think @quietvoid commented that the EDID Tmax values transmitted to an hdmi external device are used for “detection/player led” and TV led uses those internal config values.

The C2 default config file entry is same to what you’ve captured in the docs. It may not or is not the same as the EDID VSVDB value transmitted over the wire. Here’s my findings for my older E7, where the EDID VSVDB Dolby Tmax is way lower (350) than my TV’s internal DV Tmax (~750). I don’t have any Vertex to confirm this, I used an rpi5 connected to the TV & a command to get the EDID.

Those values come from the extracted firmware.
quietvoid is right, vsvdb values are used by lldv external devices.
And internal and vsvdb values can be different, eg with C8 (560 in Cinema vs 700 vsvdb).

Maybe I will add more models later as a sub section (Bx, Ax, and diferent sizes with newer models).

[valington]

Ok, I think I know what's going on with CMV4.0.

First, Dolby confirmed that L8 is not generated and must fully come from the colorist and only L2 is generated automatically and gets updated through L8 to L2 math. And 600-100nits L8 are not needed/requested most of the time.

Secondly, the 100nits L8 has an effect during DV playback and produce a similar/close effect and image to CMV2.9 with L2 trims. On Windows Energy player or movies and tv app, in HDR10 the active 100nits L8 trim always has an effect but on the C2, the 100nits L8 trim only has an effect if it's alone (without L8 600nits or 1000nits trim). Same behavior in cmv2.9. L2 100nits alone is used but when there are trims, it is not used on the C2.

So in Wonka, there is no active L8 100nit trim and the image is noticeably darker on cmv4.0 device than on 2.9 devices when there is L2 trims.
In Aquaman, there is an active L8 100nit trim and the image looks similar on both 4.0 and 2.9 devices.

So perhaps, Wonka and the ones without 100nits L8 were done in cmv2.9 and upconverted to 4.0 for streaming requirements? or maybe the colorist just was lazy and did not make trims?? idk

I've added the test files in my googledrive: https://drive.google.com/drive/u/1/fold ... fx6vRYDEfs

RESET which player is better for proper dolby vision 7 profile bdmv and mp4 ,SONY X700-X800 OR CLON OPPO ?

[RESET_9999]

Are you seeing a similarity in the studios that are leaving out the 100-nit L8 Trim?

I've seen a couple of CM4.0 without an active L8.
I don't know the % but I'll check my library for sure.

It is strange that it only works when L2 100nits is alone without any other trims on the C2.. It might be different on other TVs because as I mentioned, on windows Energy player, the 100nits trim is always used regardless of the other trims.

[RESET_9999]

RESET which player is better for proper dolby vision 7 profile bdmv and mp4 ,SONY X700-X800 OR CLON OPPO ?

the oppo is more practical because it supports iso and bdmv via the network. In term of quality, there shouldn't be much difference.
x800m2 as fewer bug than the oppo/x700 though.

[kaledi]

I've been using MakeMKV for some time and successfully ripping my 4K BR including those with DV HDR to mkv files and playing these back via a Shield TV pro (2019).
I've just purchased a Fire TV Stick 4K mkv and the same files don't play (via Emby), and I've now discovered the complexity of DV profiles and playback.
There is a lot of info in these forums, but I struggling to find concise and up to date info, so forgive me if this has been answered elsewhere.
1. Does MakeMkv rip DV tracks as dual track dual layer, or single track dual layer profile 7?
2. Related to above, are there options for how MakeMKV rips DV layers I should be aware of?
3. I believe that to improve compatibility of DV files, I need to convert profile 8 and there are tools available to do this (dovi_tool)
4. I've read that removing FEL in that process is a good thing as it isn't used by players?

[susanstone2022]

It is strange that it only works when L2 100nits is alone without any other trims on the C2.. It might be different on other TVs because as I mentioned, on windows Energy player, the 100nits trim is always used regardless of the other trims.

isn't 100nits trim intended for SDR target display?

[skull88]

Yes, it is primarily intended for that according to Dolby documentation and spec, as colourist should be monitoring on a 100-nit SDR display when making that trim, however, it seems that if it's the only L2 level trim present, it's still used to adjust tonemapping on at least some TV models and playback devices that are not SDR or low-nits. If you think about it, it makes sense, and explains why Dolby says it is mandatory to do the 100-nit trim first, whether any SDR deliverable is going to be exported or not, as 600-nit and 1000-nit level L2 is just interpolated from the 100-nit SDR trim, if no further edits are made. So, if the 600/1000-nit L2 trims are missing/removed, I think the Dolby Vision engine must be doing its own interpretation of the 100-nit trim - we just assumed it was ignored unless screen was 100-nit SDR display, but that doesn't seem to be the case, at least in some situations. I do wonder if EDID plays a role here and/or to what extent? How much of an impact it actually has in real world playback is something RESET_9999 is testing currently, I believe, perhaps he will share findings.

[susanstone2022]

Let's say one makes a content with 1000nits master display and does 100nits trim for SDR. When the content gets playback on a 1000nits display, it will need to use 100nits trim? This does not make sense to me.