RESET_9999 wrote: ↑Thu Jan 05, 2023 4:40 pm
I doubt this will ever be fixed and the different L1/L2 responses doesnt seem to affect actual content. Also, I don't think it's related to CMV2.9 vs CMV4.0 devices.
Here's my observations so far...
X800m2/x700:
L1 low contrast and L2 offset 0.75 doesnt work
no issue with "multi" pattern
Shield:
L2 offset 0.75 doesnt work
color issues (bug still there, checked recently)
issues with "multi" pattern (blue banding and clipping conversion issues)
C2 internal player:
all the L1/L2 metadata works
issues with "multi" pattern ( clipping conversion issues)
Movies with source max_pq 3696 (MDL 4000nits) look a lot darker than the x800/x700/shield:
https://www.youtube.com/watch?v=sl_wopsmdmc
C8 internal player:
all the L1/L2 metadata works
No issue with "multi" pattern
FireTV cube latest gen and probably firestick max:
all the L1/L2 metadata works
issues with "multi" pattern (tracking issue)
overall darker image
Chromecast GTV (cmv2.9 device):
L1 optimal doesnt work
L2 looks quite different in LLDV than in tv-led
the first test showed that it produced a brighter image but I still have more tests to do.
https://slow.pics/c/pEEQBJJD
no issue with "multi" pattern but if you FF/RW, the pattern start flickering.
I know @quietvoid found out that the "multi" pattern "may" have luma issues but I'm not sure about that because I checked the other versions of the pattern and they all behave exactly the same.
So unless Stacey Spears did the same mistake for all the patterns, I believe this is just how the pattern is supposed to work and it truly exposes issues on devices (shield and C2)
They also all show 10 000 nits but it's only because of a couple of pixels in the middle of the frame. The pattern are really 600/1000/4000 or 10 000 nits so I redid all the versions without trim passes and with proper source_max pq and maxcll/fall. You can find them here:
https://drive.google.com/drive/u/1/fold ... 52Ja3JEc7c