Dolby Vision now possible through MP4 Mux.

[rosavanwinkle]

Yeah, it's easy, paid media player. But this doesn't happen

Powerdvd is a paid media player and they don't have a license for it. Given that, I doubt we'll see it any time soon. I suspect we'll see TVs that negate the need for it before we see it on the pc.

Sent from my SM-G975U using Tapatalk

We need to wait until next month, for powerdvd20

If not come with DV, possibly we "never" have DV in Pc

From what i read. People doubt that it will come too powerdvd20 a lotpeople have given up on that program years ago :/

[shawnc22]

I can assure you that the tool works: the batch file (the one with the .bat extension) starts the program expecting to work with huge files (and not with 90MB files). To combine the two example HEVC raw streams you have to launch my tool using the command line:

ah got it, it was definitely the file size issue then. I've tried multiples full files and they've all worked beautifully.

i can get you a snippet of the doctor sleep tracks later today if nobody else has gotten to it yet by then

[mike admin]

I have rendered every single frame of a FEL (Terminator) and an MEL (Joker) stream.

Thanks for doing this! This is amazing discovery that sums it up nicely.

However, I believe it is appropriate to keep it.

100% Agree, MakeMKV will never do "dvd fab method" - the cost of keeping EL is negligible and benefits are many.

From what I see, the DV decoding process must be triggered externally. In m2ts such trigger is DV PSI descriptor, in mp4 - the presence of DV dsi box.
My current MKV samples store no DV metadata in codec private area - once I add the support I'll update the samples and hopefully this would give the start to DV support in MKV container.

[Angryunibrow]

what movies have you tried? I can maybe give it a try later tonight if I have it in my library

I tried Joker, Doctor Sleep, First Man, and Spectre.

yep, I think Doctor Sleep EL is not processed. It gives me an error with V1 tool and with V2 tool , EL analyzing stays at 0%.

Ok got Doctor Strange to process just fine, I always use EAC3TO to extract my BL.hevc and EL.hevc layers.
V2 of the tool processed the file to 100% just fine.

But when I mux it in TSMuxer this is what I get, and take a look at the Mediainfo output. There is no Dolby Vision (only HDR10+)

Code: Select all

Video
ID                                       : 4113 (0x1011)
Menu ID                                  : 1 (0x1)
Format                                   : HEVC
Format/Info                              : High Efficiency Video Coding
Format profile                           : Main 10@L5.1@High
HDR format                               : SMPTE ST 2094 App 4, Version 1, HDR10+ Profile B compatible
Codec ID                                 : 36
Duration                                 : 2 h 32 min
Bit rate                                 : 66.9 Mb/s
Width                                    : 3 840 pixels
Height                                   : 2 160 pixels
Display aspect ratio                     : 16:9
Frame rate                               : 23.976 (24000/1001) FPS
Color space                              : YUV
Chroma subsampling                       : 4:2:0 (Type 2)
Bit depth                                : 10 bits
Bits/(Pixel*Frame)                       : 0.337
Stream size                              : 71.1 GiB (94%)
Color range                              : Limited
Color primaries                          : BT.2020
Transfer characteristics                 : PQ
Matrix coefficients                      : BT.2020 non-constant
Mastering display color primaries        : Display P3
Mastering display luminance              : min: 0.0001 cd/m2, max: 1000 cd/m2
Maximum Content Light Level              : 745 cd/m2
Maximum Frame-Average Light Level        : 47 cd/m2

Audio
ID                                       : 4352 (0x1100)
Menu ID                                  : 1 (0x1)
Format                                   : AC-3 MLP FBA 16-ch
Format/Info                              : Audio Coding 3 + Meridian Lossless Packing FBA with 16-channel presentation
Commercial name                          : Dolby TrueHD with Dolby Atmos
Muxing mode                              : Stream extension
Codec ID                                 : 131
Duration                                 : 2 h 32 min
Bit rate mode                            : Variable
Bit rate                                 : 640 kb/s
Maximum bit rate                         : 7 347 kb/s
Channel(s)                               : 8 channels
Channel layout                           : L R C LFE Ls Rs Lb Rb
Sampling rate                            : 48.0 kHz
Frame rate                               : 31.250 FPS (1536 SPF)
Compression mode                         : Lossy
Stream size                              : 696 MiB (1%)
Language                                 : English
Service kind                             : Complete Main
Number of dynamic objects                : 15
Bed channel count                        : 1 channel
Bed channel configuration                : LFE

[SamuriHL]

I have rendered every single frame of a FEL (Terminator) and an MEL (Joker) stream.

Thanks for doing this! This is amazing discovery that sums it up nicely.

However, I believe it is appropriate to keep it.

100% Agree, MakeMKV will never do "dvd fab method" - the cost of keeping EL is negligible and benefits are many.

From what I see, the DV decoding process must be triggered externally. In m2ts such trigger is DV PSI descriptor, in mp4 - the presence of DV dsi box.
My current MKV samples store no DV metadata in codec private area - once I add the support I'll update the samples and hopefully this would give the start to DV support in MKV container.

That would be FANTASTIC! Now I'm getting excited about DV backups. I have a ton of DV UHD's and usually I just use madvr tonemapping with HDR, but, if I can make an MKV that has both....WOA!

[jayper]

[DaMacFunkin]

The conversations I have had with Nvidia staff seem to sugest that Dolby Vision will never support MKV container ???

[SamuriHL]

Not super relevant. If we can take an MKV output with DV and convert it to TS then we have a viable backup solution. Maybe some day they'll decide MKV is good or maybe they won't. But for now, this would still be a usable solution.

[rosavanwinkle]

May i ask? why is everyone so focused on MKV? why not mp4 etc.? dumb question possibly

[SamuriHL]

MP4 doesn't technically support many of the things that MKV supports such as HD Audio and blu-ray type subtitles.

[Woodstock]

MP4 may be limited, but, for enough money, the spec will "magically" support additional things.

Graphic subtitles like VOBSUB and PGS aren't supported because the people who created them do not WANT them in MP4. That might encourage piracy, don'cha know!

[SamuriHL]

Righttttt cause I spent many thousands of dollars building a high end home theater system with 7.1.2 ATMOS sound, 65" OLED, UB820 UHD player, high end HTPC, 20 TB NAS, SHIELD, Firestick 4k....so I could pirate content. SMFH. (And no, I know you were being sarcastic). But yea, these formats can be a pain in the ass. Which is why I don't currently bother backing up the DV part of it and just use my high end HTPC to tone map with madvr.

[justinadams]

Welp, i'm stumped.
I can't seem to get Joker to process to single layer correctly.
I just did Alita and it worked without issue. Demux with tsmuxer (the one that has the folder name of "tsMuxer test"), yusesope's v2 tool, remux to ts with tsmuxer. I have tried a different copy of joker (just in case), demux with ffmpeg, remux without audio, adjusting the the load_balancing_factor of yuescope's tool (since it was showing either 99.6% or 99.7% of EL once it completed with no error).
only success I have had was with doing a 20 second clip with no audio.
anything suggestions would be much appreciated.

[shawnc22]

Welp, i'm stumped.
I can't seem to get Joker to process to single layer correctly.
I just did Alita and it worked without issue. Demux with tsmuxer (the one that has the folder name of "tsMuxer test"), yusesope's v2 tool, remux to ts with tsmuxer. I have tried a different copy of joker (just in case), demux with ffmpeg, remux without audio, adjusting the the load_balancing_factor of yuescope's tool (since it was showing either 99.6% or 99.7% of EL once it completed with no error).
only success I have had was with doing a 20 second clip with no audio.
anything suggestions would be much appreciated.

Same issue with Doctor Sleep. Going through the steps with the whole movie resulted in a 99.2% EL analysis in yusecope's tool and then no DV layer once remuxed back into a TS container. Using a 20s clip, the files go through yusecope's at 100% on everything and the resulting TS with no audio contains both DV and HDR 10+:

Code: Select all

Video
ID                                       : 4117 (0x1015)
Menu ID                                  : 1 (0x1)
Format                                   : HEVC
Format/Info                              : High Efficiency Video Coding
Format profile                           : Main 10@L5.1@High
HDR format                               : Dolby Vision, Version 1.0, dvhe.08.06, BL+EL+RPU / SMPTE ST 2094 App 4, Version 1, HDR10+ Profile B compatible
Codec ID                                 : 36
Duration                                 : 19 s 19 ms
Bit rate                                 : 46.1 Mb/s
Width                                    : 3 840 pixels
Height                                   : 2 160 pixels
Display aspect ratio                     : 16:9
Frame rate                               : 23.976 (24000/1001) FPS
Color space                              : YUV
Chroma subsampling                       : 4:2:0 (Type 2)
Bit depth                                : 10 bits
Bits/(Pixel*Frame)                       : 0.232
Stream size                              : 104 MiB (95%)
Color range                              : Limited
Color primaries                          : BT.2020
Transfer characteristics                 : PQ
Matrix coefficients                      : BT.2020 non-constant
Mastering display color primaries        : Display P3
Mastering display luminance              : min: 0.0001 cd/m2, max: 1000 cd/m2
Maximum Content Light Level              : 745 cd/m2
Maximum Frame-Average Light Level        : 47 cd/m2

[yusesope]

My current MKV samples store no DV metadata in codec private area - once I add the support I'll update the samples and hopefully this would give the start to DV support in MKV container.

Wonderful

Welp, i'm stumped.
I can't seem to get Joker to process to single layer correctly.
I just did Alita and it worked without issue. Demux with tsmuxer (the one that has the folder name of "tsMuxer test"), yusesope's v2 tool, remux to ts with tsmuxer. I have tried a different copy of joker (just in case), demux with ffmpeg, remux without audio, adjusting the the load_balancing_factor of yuescope's tool (since it was showing either 99.6% or 99.7% of EL once it completed with no error).
only success I have had was with doing a 20 second clip with no audio.
anything suggestions would be much appreciated.

Same issue with Doctor Sleep. Going through the steps with the whole movie resulted in a 99.2% EL analysis in yusecope's tool and then no DV layer once remuxed back into a TS container. Using a 20s clip, the files go through yusecope's at 100% on everything and the resulting TS with no audio contains both DV and HDR 10+:

Code: Select all

Video
ID                                       : 4117 (0x1015)
Menu ID                                  : 1 (0x1)
Format                                   : HEVC
Format/Info                              : High Efficiency Video Coding
Format profile                           : Main 10@L5.1@High
HDR format                               : Dolby Vision, Version 1.0, dvhe.08.06, BL+EL+RPU / SMPTE ST 2094 App 4, Version 1, HDR10+ Profile B compatible
Codec ID                                 : 36
Duration                                 : 19 s 19 ms
Bit rate                                 : 46.1 Mb/s
Width                                    : 3 840 pixels
Height                                   : 2 160 pixels
Display aspect ratio                     : 16:9
Frame rate                               : 23.976 (24000/1001) FPS
Color space                              : YUV
Chroma subsampling                       : 4:2:0 (Type 2)
Bit depth                                : 10 bits
Bits/(Pixel*Frame)                       : 0.232
Stream size                              : 104 MiB (95%)
Color range                              : Limited
Color primaries                          : BT.2020
Transfer characteristics                 : PQ
Matrix coefficients                      : BT.2020 non-constant
Mastering display color primaries        : Display P3
Mastering display luminance              : min: 0.0001 cd/m2, max: 1000 cd/m2
Maximum Content Light Level              : 745 cd/m2
Maximum Frame-Average Light Level        : 47 cd/m2

Can you both try THIS temporary version (only source code)?

Code: Select all

python-3.7.6.amd64\python \path\to\bl_el_rpu_builder_temp_version.py -bl \path\to\BL.hevc -el \path\to\EL.hevc -of \path\to\output.hevc 

I imposed a more rigid alignment between the reading and writing processes (in order to avoid that the two layers, BL and EL, are staggered in the output file).

In the event that the counters reach all 100%, wait for the message "Press a button to exit".