Dolby Vision now possible through MP4 Mux.

[quietvoid]

Sorry no. All my P7.ts are from Dovi Fel on purpose. None is from a Mel. And I can assure you that they worked and played perfectly. Unfortunately I did not keep that version.

I don't know but it doesn't add up. I tested a version from the day before and it's also crashing with my test files.
You're welcome to try it: https://mega.nz/file/pZln1IhT#QlMIyPOhW ... 6LidHO1Xqg
This one is the old org.xbmc.kodiDV package.

I don't have the 2023-01-23 build anymore.

Anyways, I have to rework the code so that it works for all files.
This is just temporary.

[dwalme]

What is the best way to convert .TS files containing Profile 8 Dolby Vision (converted from Profile 7 MEL) to MKV?

I thought I'd just use MKVToolNix and batch up a bunch of them, but the resulting MKVs only show HDR10 with mediainfo.

If it use tsMuxeR to demux the .TS file first and then use MKVToolNix the resulting MKV files preserve Dolby Vision.

Is this the fastest way to go from a .TS container to an MKV container?

[ragico]

Sorry no. All my P7.ts are from Dovi Fel on purpose. None is from a Mel. And I can assure you that they worked and played perfectly. Unfortunately I did not keep that version.

I don't know but it doesn't add up. I tested a version from the day before and it's also crashing with my test files.
You're welcome to try it: https://mega.nz/file/pZln1IhT#QlMIyPOhW ... 6LidHO1Xqg
This one is the old org.xbmc.kodiDV package.

I don't have the 2023-01-23 build anymore.

Anyways, I have to rework the code so that it works for all files.
This is just temporary.

Tried and it crashes as you said.
Will wait for the code reworked version.
Thanks a lot.

[quietvoid]

What is the best way to convert .TS files containing Profile 8 Dolby Vision (converted from Profile 7 MEL) to MKV?

I thought I'd just use MKVToolNix and batch up a bunch of them, but the resulting MKVs only show HDR10 with mediainfo.

If it use tsMuxeR to demux the .TS file first and then use MKVToolNix the resulting MKV files preserve Dolby Vision.

Is this the fastest way to go from a .TS container to an MKV container?

MKVToolNix only supports muxing the Dolby Vision config for MP4/MKV (copied from source) and ES (raw bitstream) which is parsed to create the config.
.TS is not supported, you could always make a feature request at the GitLab.

Alternatively, you can try FFmpeg as well.

[dwalme]

MKVToolNix only supports muxing the Dolby Vision config for MP4/MKV (copied from source) and ES (raw bitstream) which is parsed to create the config.
.TS is not supported, you could always make a feature request at the GitLab.

Alternatively, you can try FFmpeg as well.

FFmpeg did the trick. Thank you.

EDIT: Spoke too soon. MediaInfo reports the correct information, but the files only play in HDR.

[DylanValenti]

Forgive me if this is a dumb question but what does MDL mismatch mean? Is there a way of fixing it? I noticed a lot of the incompatible web-dls on your spreadsheet cite “MDL mismatch”, but a lot of the compatible ones do as well @RESET_9999

[RESET_9999]

Forgive me if this is a dumb question but what does MDL mismatch mean? Is there a way of fixing it? I noticed a lot of the incompatible web-dls on your spreadsheet cite “MDL mismatch”, but a lot of the compatible ones do as well @RESET_9999

MDL = mastering display luminance, in other words, this is the monitor brightness they used to grade the movie.

Some are in the "green" section because of the L1 plot comparison that clearly show that they are the same brightness. Why is there MDL mismatch? I dont know but studios are probably stupid, i doubt they did exactly the same grade on two different monitors so one value must be an error. Same goes for Bt2020 vs P3 of the MD, sometimes the same movie have different value depending on the streaming services but they are exactly the same grade.

But the thing is that MDL is defined in L6 metadata which has no effect on the image in DV but its also defined in the "source max PQ" which has a direct impact on the brightness in DV, see:
https://www.youtube.com/watch?v=sl_wopsmdmc
https://www.youtube.com/watch?v=56y0svpsTMw

so three options:
1- be 100% safe and never mux web and bd that have a MDL mismatch
2- if L1 plot is close enough, edit "source max pq" to what makes more sense(by looking at the grade brightness) and hope it's accurate.
3- if L1 plot is close enough, leave "source max pq" untouched and hope it's accurate.

I use option 2 FYI.

There are also some MA web-dl that have the first two frames MDL 1000nits but all the other frames have MDL 4000nits (Puss in boots 2 / Tar)

[powdeau]

So what's your latest opinion on L1 in real world content on the x700/x800m2. I can see the issues on the test patterns but want to know how much it really affects actual films.I personally have a 5.1.4 atmos setup so dont really wanna sacrifice audio and use the internal player on my cx but if T1 issues are truly causing perceptible issues with DV then maybe the trade off is worthwhile. Are you still personally exclusively using your c8/c2 over the x800m2? Thanks.

since most movies have l2 trim passes, not much difference. The problem seem to be with L1 high max_pq values that have no effect.

What if the movie has only 100 nits or only 600 nits trim pass, would that be a problem?

[RESET_9999]

What if the movie has only 100 nits or only 600 nits trim pass, would that be a problem?

100nits = AFAIK it's not used in DV playback but it is used with Windows 10 Movies&TV app when tone mapped to SDR. So when the movies only has 100nits trim, its the same has no trims for DV playback.
600nits= definitely has an effect on my C2, in fact, all the trims work together at the same time. I used to think that only one trim was selected but it's not the case at all.
pattern to see what I'm talking about:
https://drive.google.com/file/d/1wgph-I ... share_link

[powdeau]

What if the movie has only 100 nits or only 600 nits trim pass, would that be a problem?

100nits = AFAIK it's not used in DV playback but it is used with Windows 10 Movies&TV app when tone mapped to SDR. So when the movies only has 100nits trim, its the same has no trims for DV playback.
600nits= definitely has an effect on my C2, in fact, all the trims work together at the same time. I used to think that only one trim was selected but it's not the case at all.
pattern to see what I'm talking about:
https://drive.google.com/file/d/1wgph-I ... share_link

Thank you, you were right about 600nits trim pass taking an effect.

So I guess if a movie has only 100nits trim pass, it's better to generate new DV metadata with Resolve (if using C1 + x700).

If I convert a HDR10+ to DV, is it possible to additionally add trim passes?

[RESET_9999]

So I guess if a movie has only 100nits trim pass, it's better to generate new DV metadata with Resolve (if using C1 + x700).

No, the only reason you should generate DV instead of using the official dv source is if the L1 plot and PQ comparison clearly show that they are not the same grade. (speaking of hybrid p8 remux of course)

now if the content has high L1 max_pq values and is "L1 no trim" or "L1 + 100nits trim only", there may be an issue with the x700/x800m2/oppo which seem to have no reaction to higher values but this is has nothing to do with generated DV.

If I convert a HDR10+ to DV, is it possible to additionally add trim passes?

Well not sure why you would want to do that but the dovi_tool has a function that can copy the metadata from one rpu to another but I believe the shots must be the same.

Anyway, i dont recommend converting HDR10plus anymore see my response here:
viewtopic.php?p=131307#p131307

[powdeau]

No, the only reason you should generate DV instead of using the official dv source is if the L1 plot and PQ comparison clearly show that they are not the same grade.

now if the content has high L1 max_pq values and is "L1 no trim" or "L1 + 100nits trim only", there may be an issue with the x700/x800m2/oppo which seem to have no reaction to higher values but this is has nothing to do with generated DV.

So to avoid that issue with x700, isn't it better to generate new DV that will have trim passes instead of using the official one without any trim passes?

[RESET_9999]

So to avoid that issue with x700, isn't it better to generate new DV that will have trim passes instead of using the official one without any trim passes?

No, the default resolve trim passes have a too strong/undesired effect when the content is very bright. When the studios grade DV, the L2 trims are edited manually shot by shot after the algo generated the default values.

This is why I overwrite them if you use my script to inject an XML file.If you go back a couple of pages, I posted a pattern that shows exactly that.

[ben_uk]

Hi, bit of a newbie here still, so forgive the questions...

Up until this point had just been using MakeMKV in this way;

Insert disc > Open MakeMKV > File > Open Disc > Then ticking the appropriate film title(s) > Then hitting makemkv

I understood this method to backup everything including the DV layer. I don't have a capable player currently of playing DV, but I thought that when I did get one (soon) all my film rips would be useable with DV, having read through some of this thread, I don't think that is possible?

Is the above method the correct way of capturing a film? I thought I read somewhere that the m2ts files are a better approach for displaying DV content, but that means you have to do a complete backup of every disc?

[RESET_9999]

@quietvoid.

Wow the new plot function in your dovi_tool is amazing. Soooo much faster than the python script.
thanks a lot