michaelni changed the topic of #ffmpeg-devel to: Welcome to the FFmpeg development channel | Questions about using FFmpeg or developing with libav* libs should be asked in #ffmpeg | This channel is publicly logged | FFmpeg 6.1.1 has been released! | Please read ffmpeg.org/developer.html#Code-of-conduct
rvalue has quit [Ping timeout: 260 seconds]
rvalue has joined #ffmpeg-devel
<BBB>
nobody ever needed more than yuv420p
<BBB>
the first yuv4mpeg1 was yuv420p only
<iive>
yuv422p was needed for interlaced stuff... I hope we'll never have to deal with interlaced stuff again.
dellas has joined #ffmpeg-devel
dellas has quit [Remote host closed the connection]
<nevcairiel>
unfortunately old content doesnt really disappear in a digital world
kasper93_ is now known as kasper93
Daemon404 has quit [Ping timeout: 276 seconds]
<Marth64>
it's still being broadcast ;-;
cone-975 has quit [Quit: transmission timeout]
cosimone has quit [Remote host closed the connection]
iive has quit [Ping timeout: 240 seconds]
<BBB>
let's just forget about broadcast, prtend it never existed
thilo has quit [Ping timeout: 256 seconds]
thilo has joined #ffmpeg-devel
Daemon404 has joined #ffmpeg-devel
philipl has quit [Remote host closed the connection]
philipl has joined #ffmpeg-devel
philipl has quit [Ping timeout: 245 seconds]
philipl has joined #ffmpeg-devel
mark4o has joined #ffmpeg-devel
markh has quit [Ping timeout: 245 seconds]
mark4o is now known as markh
navi has quit [Quit: WeeChat 4.0.4]
lemourin7 has joined #ffmpeg-devel
lemourin has quit [Killed (mercury.libera.chat (Nickname regained by services))]
kurosu has quit [Quit: Connection closed for inactivity]
jamrial has quit []
thilo has quit [Ping timeout: 260 seconds]
mkver has quit [Ping timeout: 256 seconds]
thilo has joined #ffmpeg-devel
thilo has quit [Changing host]
thilo has joined #ffmpeg-devel
cone-801 has joined #ffmpeg-devel
<cone-801>
ffmpeg Tong Wu master:d18ed2ddb59b: avcodec/d3d12va_vp9: fix vp9 max_num_refs value
<cone-801>
ffmpeg Tong Wu master:83e0fcbe037c: avcodec/d3d12va_decode: delete an empty line and fix a fuction alignment
<cone-801>
ffmpeg Tong Wu master:56c671c3b085: avcodec/d3d12va_h264: replace assert with av_assert0
<cone-801>
ffmpeg Tong Wu master:0f01581ccd75: avcodec/d3d12va_decode|dxva2: add a warning to replace assertion
<cone-801>
ffmpeg Wu Jianhua master:94949d4770d7: avcodec/d3d12va_decode: don't change the resource state if the referenced frame is the same as the current frame
<cone-801>
ffmpeg Tong Wu master:270cd14bbbd2: avcodec/dxva2(h264|mpeg2|vc1): use av_assert0 instead of assert
philipl has quit [Ping timeout: 268 seconds]
philipl has joined #ffmpeg-devel
MisterMinister has quit [Ping timeout: 255 seconds]
MisterMinister has joined #ffmpeg-devel
AbleBacon has quit [Read error: Connection reset by peer]
<Marth64>
I'm trying to add a fate subtitle test to tests/fate/subtitles.mak , but even after running configure make isn't finding the target. Is there some kind of "all" tests index I need add my test target to?
novaphoenix has quit [Ping timeout: 245 seconds]
novaphoenix has joined #ffmpeg-devel
<wbs>
Marth64: yes; it's set up like something of a tree; looking at subtitles.mak, right now, for most tests have something like this: FATE_SUBTITLES-$(call DEMDEC, SCC, CCAPTION) += fate-sub-scc
<wbs>
Marth64: which means, if the SCC demuxer and the CCAPTION decoder are available, the $(call) evaluates to "yes", which adds this to the variable FATE_SUBTITLES-yes, which then at the end gets folded into FATE_SUBTITLES, which are set up as dependency for fate-subtitles, which then later somewhere is added as a dependency of the main "fate" target
<wbs>
Marth64: so if your test fate-foo is runnable unconditionally (doubtful), do FATE_SUBTITLES += fate-foo, otherwise FATE_SUBTITLES-$(call your-dependencies-for-this-test) += fate-foo
<Marth64>
wbs: thank you, I was misunderstanding what $(call) did. it's working now
MisterMinister has quit [Ping timeout: 252 seconds]
dellas has joined #ffmpeg-devel
Mista_D has quit [Ping timeout: 256 seconds]
BetweenUs_ has quit [Quit: Leaving]
kurosu has joined #ffmpeg-devel
dellas has quit [Remote host closed the connection]
ccawley2011 has joined #ffmpeg-devel
<aaabbb>
about "avutil/eval: Use better PRNG", what are the performance and randomness requirements for the prng?
<aaabbb>
i can think of some prngs that are faster but just as random as kiss, or which are slightly slower but much more random
<jkqxz>
Does anyone ever use RGB in DNx?
<jkqxz>
The codec has a wacky macroblock-adaptive RGB feature where you can set a single bit to code a block as RGB instead of YUV (a bit like MBAFF), but I wonder whether anyone ever uses it.
<jkqxz>
The lavc implementation doesn't bother doing it adaptively, instead it checks whether all blocks in the frame are the same and sets output as YUV or RGB based on that. If they aren't all the same then it errors out, suggesting that maybe noone does that.
<JEEB>
sounds like a rather niche feature
<aaabbb>
i've used dnxhd (mostly ffv1 and prores recently tho), never used that feature either
<JEEB>
I think the question is more about "is this actually implemented by any other vendor?"
<aaabbb>
i know, i've never seen it in any other vendor is what i mean
<aaabbb>
since using ffmpeg i haven't used dnxhd though, but i've never seen that feature when i did use it
<elenril>
we should have a news entry for vvc
<kurosu>
jkqxz: I think I touched that code. My test streams never seemed to be adaptive. That's indeed an expensive way of signalling RGB
<jkqxz>
kurosu: Yes. Even more absurdly, the bit is still present and must be zero in profiles where you aren't allowed to set it.
<kurosu>
ok that was 8 years ago (71199ee9077da7d92a8728e2a694fb1ab31488a7), so I don't think I will be able to remember much beyond that
<kurosu>
if you want another weirdness, the log reminded me that a qstep once coded on 11 bits, became 10 bits to make room for an interlaced flags. profile-dependent of course
epony has quit [Remote host closed the connection]
<aaabbb>
Marth64: since you seem to have submitted some documentation typo fixes, there are a few others that i found that i could relay to you. only one i remember right off the top of my head is https://ffmpeg.org/ffmpeg-all.html#opus ("Currently its in development" should be "Currently, it's in development")
nevcairiel has quit [Remote host closed the connection]
<cone-467>
ffmpeg Martin Storsjö master:102045028ee5: configure: Disable inline assembly with nonlocal labels with LTO on Clang on Windows
<cone-467>
ffmpeg Martin Storsjö master:de9bc4f2057b: configure: Don't pass -mfp16-format to MSVC on ARM
jamrial has joined #ffmpeg-devel
epony has joined #ffmpeg-devel
tmm1_ has joined #ffmpeg-devel
navi has joined #ffmpeg-devel
tmm1 has quit [Ping timeout: 256 seconds]
novaphoenix has quit [Quit: i quit]
novaphoenix has joined #ffmpeg-devel
dellas has quit [Remote host closed the connection]
Krowl has joined #ffmpeg-devel
deus0ww has quit [Ping timeout: 264 seconds]
deus0ww has joined #ffmpeg-devel
Krowl has quit [Quit: Krowl]
epony has quit [Remote host closed the connection]
Krowl has joined #ffmpeg-devel
epony has joined #ffmpeg-devel
zsoltiv has joined #ffmpeg-devel
Krowl has quit [Read error: Connection reset by peer]
cone-467 has quit [Quit: transmission timeout]
tmm1_ has quit [Ping timeout: 256 seconds]
tmm1 has joined #ffmpeg-devel
Krowl has joined #ffmpeg-devel
zsoltiv has quit [Remote host closed the connection]
zsoltiv has joined #ffmpeg-devel
MisterMinister has joined #ffmpeg-devel
<BBB>
kierank: the CC considers your conduct on IRC to have been needlessly inflammatory and derogatory to a fellow developer ("forget what Lynne says", "Vulkan is not important", "meltdown", "farce").
<BBB>
kierank: the CC asks that you choose your words more carefully and avoid turning civil discussion into fights.
<kierank>
ok
epony has quit [Remote host closed the connection]
<Lynne>
I'd like to add that ffmpeg is not a consensus-based project unless the TC is involved, and that valid review comments must never be ignored
<kierank>
Threatening a revert is not acceptable
<kierank>
end of story
<Lynne>
threatening to push while there are comments is not acceptable
<kierank>
The comments were addressed, the patches were already written
<Lynne>
while in the past there may have been elements where it was necessary, those are not active
<Lynne>
the comments were still discussed
<kierank>
One person cannot block years of work
<Lynne>
actively, between me and nuomi, and publicly here
<kierank>
Just like the way FFmpeg 6.1 was blocked for a whole month beause "vulcan was going to be done in a week"
<kierank>
and eventually vulcan was merged 3 months later
<elenril>
sigh
<Lynne>
it was not blocked
<elenril>
don't start again
<Lynne>
developers can block patches, this is how development works, there are comments that have to be discussed, and they were
<kierank>
Yes, by shouting at everyone at FOSDEM
<kierank>
That's exactly how it works
<elenril>
kierank: stop that right now
<BBB>
oh boy... yeah, let's not do this again please
kierank has left #ffmpeg-devel [#ffmpeg-devel]
<BBB>
:(
<elenril>
sigh
<BBB>
elenril: so... I think the lack of clarity regarding dev rules could use some clarification
* BBB
summons TC
<elenril>
I don't think that's a TC question really
<BBB>
because what Lynne/kierank are ... "discussing"? ... is to my knowledge not codified
<BBB>
possibly not. let's vote? :)
<jamrial>
BBB: you need to sacrifice a git branch to summon the TC
<elenril>
yeah, people love voting
<BBB>
my point was more that I don't think it's settled law
<elenril>
jamrial: lol
<BBB>
I don't have a solution, just a problem
<BBB>
I'm the dummy user filing github issues with unclear descriptions
<elenril>
BBB: I entirely agree, but it's a part of a wider problem
<BBB>
"I pressed three buttons and it died!"
<elenril>
development rules in general are not cofified
<BBB>
(whch three? dunno)
<elenril>
s/f/d
<BBB>
right
<BBB>
my personal opinion is that it's not ok for one developer to be able to block patches. but if that happened we already have TC to solve that, so I don't see the need to fight over it
<BBB>
but maybe ideally we would all agree that generally that's not how things work
<elenril>
yeah, but you'd need to propose some way how they should work
<elenril>
I imagine making that precise enough would be nontrivial
<elenril>
Lynne: that said, threatening a revert is not very nice, and should only be done as a last restort, if ever
<wbs>
admittedly, this time, it would have been considerate to have waited a couple hours before pushing. lynne was considering whether she was ok with deferring her request, and her last comment before it was pushed was asking for opinions from others
<wbs>
if that feedback had been received and lynne had replied "ok, fine, go ahead", the whole interaction would have felt less overriding an individual reviewer
Mista_D has joined #ffmpeg-devel
MisterMinister has quit [Ping timeout: 255 seconds]
Mista_D has quit [Ping timeout: 260 seconds]
MisterMinister has joined #ffmpeg-devel
AbleBacon has joined #ffmpeg-devel
zsoltiv__ has joined #ffmpeg-devel
zsoltiv_ has quit [Ping timeout: 256 seconds]
dellas has joined #ffmpeg-devel
\\Mr_C\\ has joined #ffmpeg-devel
epony has joined #ffmpeg-devel
Krowl has quit [Read error: Connection reset by peer]
rvalue has joined #ffmpeg-devel
Mista_D has joined #ffmpeg-devel
MisterMinister has quit [Ping timeout: 260 seconds]
dellas has quit [Remote host closed the connection]
dellas has joined #ffmpeg-devel
epony has quit [Quit: QUIT]
Mista_D has quit [Ping timeout: 264 seconds]
<Lynne>
I agree, at the time though, I felt it was the last resort, as NuoMi was unsure whether to push or not, and kierank telling him to do it immediately
<Lynne>
in hindsight, the patches were going in regardless of when, and if there were bugs, they will be fixed
<wbs>
yeah. it's great that we have it in now, sad that the celebratory moment was somewhat soured by this
Krowl has joined #ffmpeg-devel
epony has joined #ffmpeg-devel
epony has quit [Max SendQ exceeded]
epony has joined #ffmpeg-devel
<Marth64>
aaabbb: thanks!
epony has quit [Remote host closed the connection]
epony has joined #ffmpeg-devel
elastic_dog has quit [Ping timeout: 268 seconds]
zwayn has quit [Quit: Connection closed for inactivity]
dellas has quit [Remote host closed the connection]