Maarten Vanraes
Maarten Vanraes
Still the same issue with python3-requests-2.25
I guess the negative is about the s64 vs u64 ? but, I assume this means that the firmware does not actually do anything with the timestamps (for all the...
ok, good news, I'll definately give it a try; also, wont both of these frames have an identicaly timestamp then? would that be problematic? maybe if someone selects deinterlace on...
So, i found this old commit https://github.com/raspberrypi/linux/commit/17fa4692615e5aea379af8764342f21198dd20f5 where it introduces the nsec -> usec conversion due to encoding rate control (not sure decoder has a similar issue?) But, how would...
that's awesome news, i had already reverted it on my version, and it seemed to work, but i didn't test any rate control yet. PS: a bit off-topic, but i...
thx! btw: should we close this? or are you planning on reverting this on the raspberry pi kernel too? I'll have this reverted when i do my next RFC on...
> FWIW I'd suggest renaming it to bcm2835-m2m when you next upstream it to make Nicolas happier. Yes only 2 use cases are codecs, but they all follow the memory-to-memory...
i second that the steps are gone when you press accept... and can't get them back.
> To me the request looks like overkill for an edge case. > > The request even start with false premises, which hint (at least to me) that the requester...
yeah, i noticed that i checked 3.3.3 and not 3.5.3, so it was older, i mislooked.