Dolby Vision now possible through MP4 Mux.

[HotFudge]

if L8 is active, it will use L1+L3+L8. Check out these metadata video comparisons: https://drive.google.com/drive/folders/ ... dMiy2ka0mf

RESET you are MVP

Thanks again

[HotFudge]

L6 is useless

why is this metadata useless?

[RESET_9999]

it's not used in DV

[HotFudge]

it's not used in DV

Interesting. Thanks

[deadchip12]

it's not used in DV

Hmm then what's the point of making it a setting in cpm build?

[RESET_9999]

For the same reason my scripts automatically edit it when injecting XML even though it's useless, why not?
Just enable it if it makes you feel better...

[Epedemic]

So, google decided to finally update the CCwGTV to Android 14. However, other than it being a totally botched update, it seems FEL support might be permanently gone, "thanks to" an upgrade to kernel 5.15. I can live with that, as I have a X700 as well, however I wonder if anyone can confirm wether MEL is still played correctly? (It looks like it to me, but I cannot verify it as well as the experts here can)

[RESET_9999]

It doesnt matter because the GTV FEL playback is out of sync anyway and it does fake tv-led.
The x700 alters the metadata(not true TV-LED) and ignores any positive lift so get the ugoos if you want the best bug-free DV quality.

[Epedemic]

It doesnt matter because the GTV FEL playback is out of sync anyway and it does fake tv-led.
The x700 alters the metadata(not true TV-LED) and ignores any positive lift so get the ugoos if you want the best bug-free DV quality.

Well, I have an ancient TV (LG E6), so TV-LED is not for me yet.

[RESET_9999]

pretty sure all the LG TV supports TV-LED and only a few early SONY models didn't.
Maybe you're confusing it with cmv4.0?

[RESET_9999]

Since my next TV (G5) will be able to play most content without tone mapping, I thought it would be useful to list all the 4000-nit movie grades. So, I added two new tabs to my Google Drive Excel sheet:

1- movies brighter than 1000nits: https://docs.google.com/spreadsheets/d/ ... 1557834794
2- movies that are practically SDR in HDR container: https://docs.google.com/spreadsheets/d/ ... 1488998734

The dark grades list will be updated frequently over the next few days, as many movies are graded under 300 nits, and I'm still going through my library. For the bright grades list, I think I found most of them (about 200 titles)

[powdeau]

If I want to check whether a x700 would alter metadata or not, I just just check if L1 plot values are lower than MDL value?
E.g. if all L1 values would be under 1000 nits and MDL is 1000, then it is OK?

So the only problem would be L1 over 1000 nits with MDL 1000 and L1 over 4000 with MDL 4000? And if that happens I just edit the MDL to higher value (e.g. edit Mad Max Fury Road to MDL 10 000)

[RESET_9999]

Yes, any value over the MDL is ignored, and you cannot change the RPU MDL. Modifying it would make everything else inaccurate since Dolby calculates metadata differently based on the MDL.
Also, it ignores any positive lift offset (values over 2048). Used in almost all movies...

[powdeau]

So if the L1 values are higher than MDL, it is better to generate new DV metadata with higher MDL, right?
But what about the Mad Max Fury Road, since I cannot set MDL higher than 4000 in Davinci? Is it than better to watch it in HDR10?

Regarding the positive offset – isn’t it mostly used in 100-nit plots and thus doesn’t have much effect?

BTW Haven’t you changed something in your "Chroma Weight Bug" sheet, since I cannot scroll there anymore?

[RESET_9999]

it is better to generate new DV metadata with higher MDL, right?
But what about the Mad Max Fury Road, since I cannot set MDL higher than 4000 in Davinci? Is it than better to watch it in HDR10?

if the movie constantly goes over 1000nits with a 1000nits MDL, yes.
if it goes over 4000nits, there's nothing you can do about it, it will clip.

Regarding the positive offset – isn’t it mostly used in 100-nit plots and thus doesn’t have much effect?

No, the positive lift is often found in the 600-1000-nit trim, and the example I just posted is a 600-nit trim.
Even if the positive lift were only in the 100-nit trim, the problem would still exist because the metadata is always extrapolated to any brightness target.

BTW Haven’t you changed something in your "Chroma Weight Bug" sheet, since I cannot scroll there anymore?

should be fixed now