Dolby Vision now possible through MP4 Mux.

[jayper]

You should also update to 1.7.6 in which you can now use (1-1) to inject both XML and json.

Is 1.7.6 official now? Or still in beta? Thanks again for all of your effort on this project.

[RESET_9999]

Is 1.7.6 official now? Or still in beta? Thanks again for all of your effort on this project.

yes it's the latest stable version but a new beta 1.7.6.1 coming up later today.

-- HDR10plus_tool and Dovi_tool must be updated to the latest version (hdr10plus to DV was totally wrong with the old dovi_tool)
I suggest you re-do any HDR10plus to DV conversion you did with the old dovi_tool
see: https://slow.pics/c/G6OH6k9p
-- HDR10plus plotter no longer requires python
-- new workflow that can batch remove CMV4.0 metadata block (2-1-6):
most of the CMV4.0 movies only have a L8 100nits trims and L2 has no effect on CMV4.0 device(shield,firestick,tv internal)
pattern to see what I'm talking about:
CMV4.0 original: https://drive.google.com/file/d/1lR1u9H ... share_link
cmv4.0 removed: https://drive.google.com/file/d/1ZqS9Ml ... share_link
-- bugfix: (2-1-5) batch edit L5 failed after the first file for mkv input
-- Code cleanup

[caelar900]

I have exported HDR 10+ and Dolby Vison metadata via DaVinci Resolve.

Hi,
you can use the script in my signature (workflow 1-1) to inject both metadata. Or use the Dovi_tool and HDR10plus_tool to inject manually, you can find these tools on quietvoid's GitHub.

Has anybody got a link to a decent tutorial for generating Dolby Vision using Davinci Resolve?

see my youtube channel (link in my signature), i made a video tutorial a while ago.

if you want to save some time, I already generated DV in resolve for most of the movies, here are over 600 RPU files ready to be injected

Code: Select all

https://mega.nz/folder/CjYWyaZR#tSlKPhdblAd8shSaYoiOWA

Looks like mega took this folder down. Anyway you could reupload?

[RESET_9999]

Looks like mega took this folder down. Anyway you could reupload?

yeah i should have mentioned that the link would be online for only 1 week.
I PMed you a new link.

[jayper]

Is 1.7.6 official now? Or still in beta? Thanks again for all of your effort on this project.

yes it's the latest stable version but a new beta 1.7.6.1 coming up later today.

-- HDR10plus_tool and Dovi_tool must be updated to the latest version (hdr10plus to DV was totally wrong with the old dovi_tool)
I suggest you re-do any HDR10plus to DV conversion you did with the old dovi_tool
see: https://slow.pics/c/G6OH6k9p
-- HDR10plus plotter no longer requires python
-- new workflow that can batch remove CMV4.0 metadata block (2-1-6):
most of the CMV4.0 movies only have a L8 100nits trims and L2 has no effect on CMV4.0 device(shield,firestick,tv internal)
pattern to see what I'm talking about:
CMV4.0 original: https://drive.google.com/file/d/1lR1u9H ... share_link
cmv4.0 removed: https://drive.google.com/file/d/1ZqS9Ml ... share_link
-- bugfix: (2-1-5) batch edit L5 failed after the first file for mkv input
-- Code cleanup

Thanks so much! Apologies, but I am not yet well-read on the CMV4.0 vs CMV2.9 discussion. Is there a reason one would prefer one versus the other? I'm guessing it is more future-proof, but at this time most hardware doesn't support it? Is there a downside to keeping it?

[dom61]

Good evening.
I apologize using the link Read Me https://github.com/quietvoid/hdr10plus_tool
I can download "hdr10plus_tool.exe" and not "hdr10plus_parser.exe"
Should I rename the file?
Thank you

[RESET_9999]

Thanks so much! Apologies, but I am not yet well-read on the CMV4.0 vs CMV2.9 discussion. Is there a reason one would prefer one versus the other? I'm guessing it is more future-proof, but at this time most hardware doesn't support it? Is there a downside to keeping it?

First, most of the movies are done in CMV4.0 but exported in CMV2.9.

CMV4.0 is backward compatible with cmv2.9. The trim passes (100-600-1000nits) are done in the level 8 for cmv4.0 and level 2 for cmv2.9.
When a device and the rpu are CMV4.0, the L2 trims are ignored because it will use the L8 trims which is the proper behavior.

Now the problem is that most of the CMv4.0 movies only have a 100nits L8 trim pass which is obviously not used in DV playback but they do have proper L2 trims that are ignored on cmv4.0 devices (shield/firestick/appletv/internal tv ). So you get inferior quality on those devices unless you remove the CMV4.0 block. The test files (made with original metadata from the movie Puss in boots) show exactly the problem and I bet this is the reason why we do not see a lot of CMV4.0 movies because the colorist has to do the trim passes two times (L8 + L2).

In my script, you can use 2-6 to export the trims to a text file and verify the raw data. So if you use a CMV4.0 devices, you might want to inspect your movies for cmv4.0 metadata (mostly amazon DV and movie anywhere web-dl ) .

[RESET_9999]

Good evening.
I apologize using the link Read Me https://github.com/quietvoid/hdr10plus_tool
I can download "hdr10plus_tool.exe" and not "hdr10plus_parser.exe"
Should I rename the file?
Thank you

yeah, in the next script version I'll keep the original name "hdr10plus_tool.exe" but for the current 1.7.6, you have to rename it "hdr10plus_parser.exe" or you can change the tool path name at line 93

set hdr10plus_parser_path=%~dp0tools\hdr10plus_parser.exe ====> set hdr10plus_parser_path=%~dp0tools\hdr10plus_tool.exe

[dom61]

Good evening.
I apologize using the link Read Me https://github.com/quietvoid/hdr10plus_tool
I can download "hdr10plus_tool.exe" and not "hdr10plus_parser.exe"
Should I rename the file?
Thank you

yeah, in the next script version I'll keep the original name "hdr10plus_tool.exe" but for the current 1.7.6, you have to rename it "hdr10plus_parser.exe" or you can change the tool path name at line 93

set hdr10plus_parser_path=%~dp0tools\hdr10plus_parser.exe ====> set hdr10plus_parser_path=%~dp0tools\hdr10plus_tool.exe

Thank you for the confirmation
I went ahead in the procedure and when I am processing the file "DV.xml", it is prompted to enter a value of enter a value of CANVAS_WIDTH
How to do this?
Thank you very much

[RESET_9999]

Thank you for the confirmation
I went ahead in the procedure and when I am processing the file "DV.xml", it is prompted to enter a value of enter a value of CANVAS_WIDTH

mediainfo is used to read and set the input resolution. So your path to mediainfo might not be right??

[quietvoid]

I bet this is the reason why we do not see a lot of CMV4.0 movies because the colorist has to do the trim passes two times (L8 + L2).

That's not really how the workflow works.
The grading is done in CM v4.0 and the L2 metadata is generated automatically by software.
So the trim passes are still done only once.

Anyways, there has to be more experimenting on how CM v4.0 and L8 behave.
I don't agree that it should be removed.

[RESET_9999]

That's not really how the workflow works.

yes it is how it works... Each trims are edited manually after the analysis to make sure the desired look is preserved.
Dolby vision is not just an algorithm.
Just take the Spears and Munsill clip and generate DV in resolve, the automatic trims are a LOT darker than the original as shown here https://drive.google.com/file/d/1_PkGF9 ... share_link

Anyways, there has to be more experimenting on how CM v4.0 and L8 behave.
I don't agree that it should be removed.

As I told you in PM, take the spears and munsil, generate cmv4.0 metadata and compare on cmv2.9 and 4.0 devices. I'm sure you will agree.

[quietvoid]

yes it is how it works... Each trims are edited manually after the analysis to make sure the desired look is preserved.
Dolby vision is not just an algorithm.

I'm only clarifying your statement that the trim passes are done two times (L8 + L2), because it's not true.

If you even did a little research, you would've found that L2 is always autogenerated.
So the trim passes are what creates the L8, not the L2. At least in a CM v4.0 workflow.

For CM v2.9, of course the trims are done once and they result in the L2.

As I told you in PM, take the spears and munsil, generate cmv4.0 metadata and compare on cmv2.9 and 4.0 devices. I'm sure you will agree.

That's not at all the same situation. It would only be comparing different algorithms.

[RESET_9999]

If you even did a little research, you would've found that L2 is always autogenerated.

LOL I know very well that the dv analysis generates L2/L8 automatically, I literally asked you by pm how to overwrite them in my script because they have a too strong effect....
My point is that the trims are edited manually after the analysis. They are NOT just the result of an algorithm.

I'm only clarifying your statement that the trim passes are done two times (L8 + L2), because it's not true.

The metadata in L2 and L8 works quite differently as you know, so i disagree with you here.
If they do L2 and L8 trims they must go through each of them but then we learned that the studios can be very lazy so who knows how much work they really put in the trims.

That's not at all the same situation. It would only be comparing different algorithms.

huh? please do some tests and then come at me again

[dom61]

Thank you for the confirmation
I went ahead in the procedure and when I am processing the file "DV.xml", it is prompted to enter a value of enter a value of CANVAS_WIDTH

mediainfo is used to read and set the input resolution. So your path to mediainfo might not be right??

I solved it by downloading the correct version

After processing a movie with DaVinci, I performed the export of Dolby Vision metadata (DV.xml), following the tutorial on Youtube as reference.
I ask if it is possible to inject only Dolby Vision metadata (DV.xml)
to get Mediainfo similar to this.
Thank you very much