Re: Dolby Vision now possible through MP4 Mux.
Posted: Sat Feb 29, 2020 7:03 pm
Interesting, so profile 04 doesnt’t have a different colorspace to UHD bluray, it seems in effect DL rolled in to a single layer.
MakeMKV support forum
https://forum.makemkv.com/forum/
For my part I will not try more test and rounds, if he release the tool I will test it and publish resultsATD555 wrote: Sat Feb 29, 2020 6:36 pmThis information is great but I think we are all interested in knowing how you created the Single Track Double Layer MP4 files so we can recreate and test with our own videos?yusesope wrote: Sat Feb 29, 2020 6:26 pm I try to answer the questions of some of you.
These are my personal considerations.
There are several "types" of Dolby Vision:
Double Track Double Layer: BL and EL + RPU (this is the case of Bluray discs where the Base Layer is represented by the first video track and the rest is placed in a second video track)
Single Track Single Layer: BL + RPU (this is the case of DVDFab)
Single Track Double Layer: BL + EL + RPU (my attempt)
There are also two types of EL:
The full enhancement layer (FEL) is a HEVC video stream with embedded Dolby Vision metadata: it requires a SoC to be capable of simultaneously decoding two video streams (base layer video stream and FEL video stream).
The minimal enhancement layer MEL, a constrained FEL, only consists of metadata: it requires a SoC to be capable of only decoding one single base layer video stream.
My mp4 files meet the characteristics listed in section 3.2 (in particular, subsection 3.2.2) of THIS document and have been generated from a Bluray disc with a Dolby Vision Double Track Double Layer (BL and EL + RPU) without alter any of the components.
It is logical to think that the best profiles for decoding my files are dvhe.04 and dvhe.07 (refer to THIS document) as they allow to improve the Base Layer using EL and the instructions contained in the RPU metadata.
I do not dwell on profile 07.
I take into account profile 04 (surely those interested in this type of file have a device, such as the new Shield, which does not support profile 07).
In official Dolby documents it is stated that a new Dolby Vision certified device is able to decode a profile 4 MEL bitstream without instantiating a secondary HEVC decoder for the enhancement layer.
A new Dolby Vision device that chooses not to instantiate a second HEVC decoder and supports
profile 4 must distinguish the original profile 4 bitstream from the profile 4 MEL bitstream. When
receiving an original profile 4 bitstream, such a device:
• Exits the Dolby Vision video pipeline
• Uses its normal video pipeline for video processing, and displays a standard dynamic range video
signal only
• Does not display the Dolby Vision logo
So follow me:Code: Select all
Single Track Dual Layer dvhe.04.06 (BL + EL + RPU) | | | | My device receives a Dolby Vision Track whith a Base Layer and a Full ------------- My device is able to decode a profile 4 MEL bitstream enhancement Layer (FEL) without instantiating a secondary HEVC decoder but is ------------- Exits the Dolby Vision video pipeline, | receiving a FEL bitstream NO Dolby Vision logo, SDR | | | My device can initiate a secondary HEVC decoder | | | | Dolby Vision (Full enhanced)
My files (dvhe.04) are slightly out of standard: the EL layer should be represented by a HEVC main10 level 4.1 stream instead the one present in the Bluray disk is a High Tier level 5.1.
On some devices, the file may start (with its Dolby Vision Logo) but the file would not be reproduced because the decoder buffer would quickly become saturated.
If instead I choose to play my files with the profiles dvhe.05 and dvhe.08 (useful for processing Single Track Single Layer files (BL + RPU)), the hardware decoder will try to process the Base Layer (BL) and to improve it through RPU metadata by discarding the EL layer.
however, the Base Layer is non-standard (it has a different color space than the IPTPQc2/IPT one) as extracted from a Bluray disk.
It should appear better than an HDR stream (since metadata is dynamic and non-static) but will not reach its maximum splendor due to the lack of data contained in the EL layer.
The lack of this information, however, could lead, in some cases, to evident chromatic aberrations.
It looks just like that.mattmarsden wrote: Sat Feb 29, 2020 7:03 pm Interesting, so profile 04 doesnt’t have a different colorspace to UHD bluray, it seems in effect DL rolled in to a single layer.
yusesope wrote: Fri Feb 28, 2020 8:04 pm I think I'm onto something: I am also able to demux Single Layer DV files and transform them into Dual Layer DV files.
I made an attempt with the file you find HERE.
The result is this (HERE the DL file):Code: Select all
Video #1 ID : 1 Formato : HEVC Formato/Informazioni : High Efficiency Video Coding Profilo formato : Main 10@L5@Main ID codec : hev1 ID codec/Informazioni : High Efficiency Video Coding Durata : 1 min 14s Bitrate : 15,4 Mb/s Larghezza : 3.840 pixel Altezza : 2.160 pixel Rapporto aspetto visualizzazione : 16:9 Modalità frame rate : Costante Frame rate : 23,976 (24000/1001) FPS Spazio colore : YUV Croma subsampling : 4:2:0 Profondità bit : 10 bit Bit/(pixel*frame) : 0.077 Dimensione della traccia : 137MiB (74%) Default : Si AlternateGroup/String : 1 Data codifica : UTC 2020-02-28 19:23:55 Data : UTC 2020-02-28 19:23:55 Codec configuration box : hvcC Video #2 ID : 2 Formato : HEVC Formato/Informazioni : High Efficiency Video Coding Profilo formato : Main 10@L4@Main Dolby Vision : 1.0, dvhe.04.06, EL+RPU ID codec : hev1 ID codec/Informazioni : High Efficiency Video Coding Durata : 1 min 14s Bitrate : 5.519 kb/s Larghezza : 1.920 pixel Altezza : 1.080 pixel Rapporto aspetto visualizzazione : 16:9 Modalità frame rate : Costante Frame rate : 23,976 (24000/1001) FPS Spazio colore : YUV Croma subsampling : 4:2:0 Profondità bit : 10 bit Bit/(pixel*frame) : 0.111 Dimensione della traccia : 49,0MiB (26%) Default : No AlternateGroup/String : 1 Data codifica : UTC 2020-02-28 19:23:55 Data : UTC 2020-02-28 19:23:55 Codec configuration box : hvcC+dvcC
ATD555 wrote: Fri Feb 28, 2020 10:48 pm I will be testing the SL DV files on my Hisense H8F when I get home and post the results. Also curious to know how to make them myself.
if you repeat it a fourth time, in front of a mirror, I could appear behind you.ATD555 wrote: Sat Feb 29, 2020 6:36 pm This information is great but I think we are all interested in knowing how you created the Single Track Double Layer MP4 files so we can recreate and test with our own videos?
Code: Select all
for example a hypothetical NAL AUD
0x00000001460110
will become
0x000000017e01460110
your .tsDV files are single or double layer?bgalakazam wrote: Sun Mar 01, 2020 3:01 pm .ts file with DV and TrueHD Atmos playing back on X700![]()
Tested and works properly with PGS and fast forwarding. This is amazing. As far as I am concerned my DV needs are fully met.

update:superfans124 wrote: Sun Mar 01, 2020 5:46 pm I can confirm no more lipsync issues with DL TS on x700.
but, I'm hard crashing x700 with fast forwarding and rewinding, once again...first experienced a hard crash last night trying to rewind a section I missed when taking a leak, x700 froze beyond maybe 10 seconds of rewinding then I had to unplug the power cord. Tried the Deer Hunter this morning, glad to report no lipsync, so I fast forwarded to test, hard crashed again beyond a few seconds of ff.
any tips ? yeah, remux the original .mpls from the disc directly with tsMuxer, no eac3to.built_to_chill wrote: Sun Mar 01, 2020 6:36 pm
- I’m finding the Atmos tracks on some of my Disney movies always cause the eac3 to crash partway through, meaning I can’t add Atmos to
these .ts files. Reading around it seems because eac3 uses some out of date libraries. There was talk of recompiling the source code with newer libraries. If there is a known easy solution here any tips are welcome.
Cheers
Yes, I have an HDfury Vertex and I confirm that the x700 output proper metadata with any HDR10 content. Of course as you know, DV must be turned off which is annoying.built_to_chill wrote: Sun Mar 01, 2020 6:36 pm - Something id meant to ask: does watching .ts of HDR10 on the X700 pass through accurate maxfall etc data?
Cheers
...what's the purpose of all of that work? I'm confused. X700 can always play HDR10 and atmos off of the native m2ts file ripped from BDMV, what's all the conversion for?built_to_chill wrote: Sun Mar 01, 2020 6:36 pm I was relieved to see others experience the crash issue, and needing to turn it on/add after playing any DV ts file. I nearly went into town to have my X700 swapped assuming it was a fault. I’d also spent a while trying different HDMI cables and HDMI paths to no avail.
At the end of the day not a big deal, since I typically will watch an entire movie at once then turn it off anyway.
Some observations:
- Continuing to see lip sync is fine now having gotten away from LPCM. As mentioned above this was my experience a year or so ago with the X700 when I was just using it for HDR10.
- The crashing isn’t limited to DTS-HD in my experience, any DL DV ts file will cause the X700 to output a green screen until its turned off and on. Doesn’t happen with HDR10
- I’m finding the Atmos tracks on some of my Disney movies always cause the eac3 to crash partway through, meaning I can’t add Atmos to
these .ts files. Reading around it seems because eac3 uses some out of date libraries. There was talk of recompiling the source code with newer libraries. If there is a known easy solution here any tips are welcome.
- I’m finding fast forward (right up to x3) works fine. I think I’ve observed the crashing linked to the use of HDD vs SSD.
- Something id meant to ask: does watching .ts of HDR10 on the X700 pass through accurate maxfall etc data?
Cheers