Page 549 of 756

Re: Dolby Vision now possible through MP4 Mux.

Posted: Wed Apr 26, 2023 9:01 am
by qq1649536801
I got a problem with script to mux a ts file.
this ts file
HDR Format : Dolby Vision, Version 1.0, dvhe.08.06, BL+RPU, HDR10 compatible / SMPTE ST 2086, HDR10 compatible
Code ID : 36
the code id 36 , won`t get work in Windows computer. only the code id dvhe, dvh1 and some else.

to get id dvhe, I use the mp4muxerNEW with
./mp4muxerNEW -i D:\DV\20000_P8.hevc --dv-profile 8 --dv-bl-compatible-id 1 -o D:\DV\20000_P8.mp4
so, if there is another way to get the dvhe flag?
cause I found some DoVi demo videos are ts file with dvhe, such as LG Amaze Dolby Vision UHD 4K Demo.ts
HDR Format : Dolby Vision, Version 1.0, dvhe.05.09, BL+RPU
Code ID : dvhe
Code ID/info : High Efficiency Video Coding with Dolby Vision

Re: Dolby Vision now possible through MP4 Mux.

Posted: Wed Apr 26, 2023 2:09 pm
by RESET_9999
qq1649536801 wrote: Wed Apr 26, 2023 9:01 am I got a problem with script to mux a ts file.
If you have a file that can't be muxed in tsmuxer, you should open a ticket here: https://github.com/justdan96/tsMuxer/issues

Re: Dolby Vision now possible through MP4 Mux.

Posted: Wed Apr 26, 2023 2:29 pm
by qq1649536801
RESET_9999 wrote: Wed Apr 26, 2023 2:09 pm
qq1649536801 wrote: Wed Apr 26, 2023 9:01 am I got a problem with script to mux a ts file.
If you have a file that can't be muxed in tsmuxer, you should open a ticket here: https://github.com/justdan96/tsMuxer/issues
sorry to bother you, I found it`s my mistake. only mp4 file can make the codec id "dvhe".
just like the files in your "Dolby Vision Stuff" and "RPU and FEL test", some test video file are marked "NEWER TV" and "OLDER TV", both of them are mp4 files.
the codec ID of "NEWER TV“ mp4 file is dvh1, and the older one is dvhe.

so, I want to ask you : how did you mux those newer and older mp4 files.
Thanks for your work and everything you did in this posting!!! your scripts make me know a lot things od dolby vision.

Re: Dolby Vision now possible through MP4 Mux.

Posted: Wed Apr 26, 2023 5:56 pm
by RESET_9999
qq1649536801 wrote: Wed Apr 26, 2023 2:29 pm
sorry to bother you, I found it`s my mistake. only mp4 file can make the codec id "dvhe".
just like the files in your "Dolby Vision Stuff" and "RPU and FEL test", some test video file are marked "NEWER TV" and "OLDER TV", both of them are mp4 files.
the codec ID of "NEWER TV“ mp4 file is dvh1, and the older one is dvhe.

so, I want to ask you : how did you mux those newer and older mp4 files.
Thanks for your work and everything you did in this posting!!! your scripts make me know a lot things od dolby vision.

Ah ok I understand. in my script the ''OLDER TV'' files are muxed with the old mp4muxer.exe (7-2-1) and the ''NEWER TV'' with the new mp4muxerNEW.exe (7-1-1)

Re: Dolby Vision now possible through MP4 Mux.

Posted: Thu Apr 27, 2023 7:22 pm
by legosoff
Good evening! I'm shocked LG 1C audio ALAC to appear 8 channels with a working.

Re: Dolby Vision now possible through MP4 Mux.

Posted: Fri Apr 28, 2023 12:29 am
by RESET_9999
Next version of the DoVi_Scripts will be able to plot L2 :D
tomorrow probably...

examples:
https://slow.pics/c/OIKGsjr2
https://slow.pics/c/0ECUzXfq
https://slow.pics/c/9fiJz8tM
https://slow.pics/c/SWYYBnjD

Re: Dolby Vision now possible through MP4 Mux.

Posted: Fri Apr 28, 2023 5:19 am
by NiCE77
@RESET_9999

I found an interesting topic: http://forum.zidoo.tv/index.php?threads ... 203.95534/
One could argue that some scenes in No Time To Die are brighter and more colourful on Oppo than Zidoo Z9X.
I made an L1 plot of the RPU: https://slow.pics/J7uB5NCt/
Is it because of the FEL or because of the faulty L6 data (at least as far as I can see)?

Re: Dolby Vision now possible through MP4 Mux.

Posted: Fri Apr 28, 2023 11:03 am
by RESET_9999
NiCE77 wrote: Fri Apr 28, 2023 5:19 am @RESET_9999

I found an interesting topic: http://forum.zidoo.tv/index.php?threads ... 203.95534/
One could argue that some scenes in No Time To Die are brighter and more colourful on Oppo than Zidoo Z9X.
I made an L1 plot of the RPU: https://slow.pics/J7uB5NCt/
Is it because of the FEL or because of the faulty L6 data (at least as far as I can see)?
probably because the zidoo wrongly send a bt2020 flag with the TV-LED signal.
When I tested this device a very long time ago, it was slightly affecting the colors: Green, yellow and blue were paler and red darker.

Re: Dolby Vision now possible through MP4 Mux.

Posted: Fri Apr 28, 2023 11:21 am
by NiCE77
RESET_9999 wrote: Fri Apr 28, 2023 11:03 am
NiCE77 wrote: Fri Apr 28, 2023 5:19 am @RESET_9999

I found an interesting topic: http://forum.zidoo.tv/index.php?threads ... 203.95534/
One could argue that some scenes in No Time To Die are brighter and more colourful on Oppo than Zidoo Z9X.
I made an L1 plot of the RPU: https://slow.pics/J7uB5NCt/
Is it because of the FEL or because of the faulty L6 data (at least as far as I can see)?
probably because the zidoo wrongly send a bt2020 flag with the TV-LED signal.
When I tested this device a very long time ago, it was slightly affecting the colors: Green, yellow and blue were paler and red darker.
Thanks, in principle, it only makes a difference for this film. Anyway, is the L6 metadata good? MaxCLL 4507 seems to me to be a bad value.

Re: Dolby Vision now possible through MP4 Mux.

Posted: Fri Apr 28, 2023 11:26 am
by RESET_9999
NiCE77 wrote: Fri Apr 28, 2023 11:21 am
RESET_9999 wrote: Fri Apr 28, 2023 11:03 am
NiCE77 wrote: Fri Apr 28, 2023 5:19 am @RESET_9999

I found an interesting topic: http://forum.zidoo.tv/index.php?threads ... 203.95534/
One could argue that some scenes in No Time To Die are brighter and more colourful on Oppo than Zidoo Z9X.
I made an L1 plot of the RPU: https://slow.pics/J7uB5NCt/
Is it because of the FEL or because of the faulty L6 data (at least as far as I can see)?
probably because the zidoo wrongly send a bt2020 flag with the TV-LED signal.
When I tested this device a very long time ago, it was slightly affecting the colors: Green, yellow and blue were paler and red darker.
Thanks, in principle, it only makes a difference for this film. Anyway, is the L6 metadata good? MaxCLL 4507 seems to me to be a bad value.
L6 has no effect in DV.

There might be some parts that get slightly expanded via fel though. https://slow.pics/c/NFuzhfkV

The OP in this post says that he doesn't see a difference in other titles which is unlikely to me since a lot of FEL movies will look darker on a non-fel device.
https://docs.google.com/spreadsheets/d/ ... edit#gid=0

Re: Dolby Vision now possible through MP4 Mux.

Posted: Fri Apr 28, 2023 1:50 pm
by quietvoid
I don't remember the FEL doing much for No Time to Die.
If someone can figure out what timestamp they're talking about, maybe I'll take a look.

It's probably around 00:44:19.
Anyways, whatever difference they're seeing has nothing to do with FEL or the Dolby Vision metadata.
There's no difference with FEL.

Re: Dolby Vision now possible through MP4 Mux.

Posted: Fri Apr 28, 2023 3:50 pm
by RESET_9999
This is why my script removes the default trims for the Resolve generated DV.

https://slow.pics/c/wxH1FjWA

Re: Dolby Vision now possible through MP4 Mux.

Posted: Fri Apr 28, 2023 6:50 pm
by powdeau
@RESET_9999

Is there a way to disable all subtitles extraction in your script (even forced and full)?

Re: Dolby Vision now possible through MP4 Mux.

Posted: Fri Apr 28, 2023 7:17 pm
by RESET_9999
powdeau wrote: Fri Apr 28, 2023 6:50 pm @RESET_9999

Is there a way to disable all subtitles extraction in your script (even forced and full)?
By default, the script doesn't export subs unless you changed line 40
set export.subs=YES

TS version of the script may still extract them but it doesn't keep them in your output path if line40 is set to NO

Re: Dolby Vision now possible through MP4 Mux.

Posted: Sat Apr 29, 2023 3:09 pm
by RESET_9999
So I'm trying to make sense of L2 now that I can plot it...
From the movies I plotted L2 so far, it seems the most important and used trims are ''slope'' and ''power'' and if we look at the Eagle Beak L2 pattern: https://drive.google.com/file/d/1qdHux6 ... share_link

power: values lower than 2048 make the image brighter.
slope: values lower than 2048 make the image darker.
offset: values lower than 2048 make the image darker.
chroma weight: affect the colors
trim_saturation_gain: no effect on all the devices I tested
ms_weight: no effect on all the devices I tested

M3GAN: The L2 trims on this movie make the image noticeably brighter than HDR10 or DV without trims even though the grade L1 maxcll is only 485nits (avg 284nits).
https://slow.pics/c/KzTUSbmB

eagle beak L2 trims in action video: https://www.youtube.com/watch?v=qSo-h-Rl-ps
original 600 and 1000nits values of the Eagle beak pattern:

Code: Select all

          "Level2": {
            "target_max_pq": 2851,
            "trim_slope": 1594,
            "trim_offset": 1935,
            "trim_power": 1515,
            "trim_chroma_weight": 2474,
            "trim_saturation_gain": 2036,
            "ms_weight": 4095
          }
        },
        {
          "Level2": {
            "target_max_pq": 3079,
            "trim_slope": 1413,
            "trim_offset": 1923,
            "trim_power": 1353,
            "trim_chroma_weight": 2720,
            "trim_saturation_gain": 2308,
            "ms_weight": 4095
          }