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 has been released! | Please read ffmpeg.org/developer.html#Code-of-conduct
dellas has quit [Remote host closed the connection]
navi has quit [Quit: WeeChat 4.0.4]
elastic_dog has quit [Ping timeout: 240 seconds]
elastic_dog has joined #ffmpeg-devel
thilo has quit [Ping timeout: 268 seconds]
MrZeus_ has quit [Ping timeout: 264 seconds]
thilo has joined #ffmpeg-devel
thilo has quit [Changing host]
thilo has joined #ffmpeg-devel
cone-006 has joined #ffmpeg-devel
<cone-006>
ffmpeg Cosmin Stejerean master:737ede405b11: avfilter/bwdif: account for chroma sub-sampling in min size calculation
<cone-006>
ffmpeg Haihao Xiang master:e5f8b5313e29: doc/ffmpeg: Update the description about d3d11va
<cone-006>
ffmpeg Haihao Xiang master:f89cff96d080: lavu/hwcontext_qsv: Make sure hardware vendor is Intel for qsv on d3d11va
lemourin has quit [Read error: Connection reset by peer]
lemourin has joined #ffmpeg-devel
jamrial has quit []
uau has quit [Remote host closed the connection]
uau has joined #ffmpeg-devel
cone-006 has quit [Quit: transmission timeout]
qeed has quit [Quit: Leaving]
Marth64 has joined #ffmpeg-devel
<Marth64>
Hello, here is the DVD demux patch after several months of on/off work (using dvdread/dvdnav). It's in good shape except for some small todos and an issue I can't seem to figure out how to solve cleanly (which I would like to resolve before submitting to ML). When DTS discontinuity occurs in DVD due to valid reasons (e.g. STC discont, or VOB reset), how to handle the scenario and wrap/adjust the DTS?
<philipl>
Marth64: if you are interested in getting that merged, you really need to open a PR on github. It is most likely to disappear into /dev/null otherwise.
<philipl>
Marth64: sorry. I thought this was the mpv channel. :-P
<Marth64>
philipl: NP :D I will, I just have some small details to fix up (actually this time, plan to submit in the next few days). I was hoping for any casual advice on dealing with the discontinuity here.
<Marth64>
Many non-fancily authored DVDs work out of the box. But need to fix this, and PCM before the weekend
<Marth64>
Usage is fairly simple ... `ffmpeg -f dvdvideo -i dvd_structure.iso|/dvd_structure|/dev/srN -title N` probing is planned for phase 2 maybe next year
<Marth64>
er ... `ffmpeg -f dvdvideo -title N -i dvd_structure.iso|/dvd_structure|/dev/srN`
<philipl>
Marth64: I'm deeply unqualified here but I know mpeg-ts has a specific packet to indicate a dicontinuity. Does mpeg-ps offer that? If it did, you could inject one.
<Marth64>
philip: That was my line of thinking as well...I couldn't find an obvious answer just reading the mpeg-ps demuxer (which is my sub-demuxer) :( I can get the discont signal from the DVD easily, but no place to put it in the sub-demuxer
<Marth64>
That said, good thought, maybe I should go check out the PS specs
<philipl>
Good luck!
<Marth64>
TY
<Lynne>
all this over some insignificant tests
rvalue has quit [Read error: Connection reset by peer]
rvalue has joined #ffmpeg-devel
<JEEB>
Marth64: I think H.222 2021 ed. is still free at ITU, I *think* that should contain both MPEG-2 systems (PS and TS)
AbleBacon has quit [Read error: Connection reset by peer]
<Marth64>
JEEB: nice, yes just got it. thank you that gives me a good start for tomorrow
<Marth64>
so close...just some minor but noticeable flaws
<Marth64>
at worst it's still giving me better luck than dvd2concat
Krowl has joined #ffmpeg-devel
lexano has quit [Ping timeout: 245 seconds]
Teukka` has quit [Ping timeout: 264 seconds]
lexano has joined #ffmpeg-devel
Teukka has joined #ffmpeg-devel
Teukka has quit [Changing host]
Teukka has joined #ffmpeg-devel
Krowl has quit [Read error: Connection reset by peer]
deus0ww has quit [Ping timeout: 252 seconds]
deus0ww has joined #ffmpeg-devel
dellas has joined #ffmpeg-devel
markh has quit [Ping timeout: 255 seconds]
Krowl has joined #ffmpeg-devel
markh has joined #ffmpeg-devel
tmm1 has quit [Ping timeout: 240 seconds]
ccawley2011 has joined #ffmpeg-devel
Marth64[m] has joined #ffmpeg-devel
Marth64 has quit [Ping timeout: 245 seconds]
elastic_dog has quit [Ping timeout: 240 seconds]
darkapex has quit [Remote host closed the connection]
elastic_dog has joined #ffmpeg-devel
tmm1 has joined #ffmpeg-devel
catfella__ has quit [Remote host closed the connection]
tmm1 has quit [Ping timeout: 246 seconds]
tmm1 has joined #ffmpeg-devel
mkver has joined #ffmpeg-devel
dellas has quit [Remote host closed the connection]
Krowl has quit [Read error: Connection reset by peer]
<Lynne>
nevermind, issue was on my end
dellas has joined #ffmpeg-devel
kurosu has joined #ffmpeg-devel
Krowl has joined #ffmpeg-devel
navi has joined #ffmpeg-devel
dellas has quit [Ping timeout: 256 seconds]
dellas has joined #ffmpeg-devel
mkver has quit [Ping timeout: 252 seconds]
zane has quit [Quit: WeeChat 4.1.1]
dellas has quit [Remote host closed the connection]
derpydoo has joined #ffmpeg-devel
catfella has joined #ffmpeg-devel
<BBB>
michaelni seems to be the only one actually asking sensible questions in the whole paul debacle. what actually happened? do I even want to know?
dellas has joined #ffmpeg-devel
<jamrial>
BBB: i don't think anyone knows
<BBB>
I grow rather tired of children's temper tantrums
<BBB>
you know, having 2 teenagers myself
<BBB>
ohwell. if paul can't be bothered to communicate like an adult, then whatever...
dellas has quit [Remote host closed the connection]
<BBB>
unfortunate
<kierank>
I don't blame paul for leaving
<kierank>
the whole nicolas situation is awful
<jamrial>
but that's not with him
<ePirat>
BBB, wasnt it bc elenril wanted to remove QOA because lack of tests?
<BBB>
that was a bit overly aggressive, yes
* BBB
frowns @ elenril
<aaabbb>
he left for good?
<courmisch>
as he wrote, that was just the drop to break the back of the proverbial camel
<Lynne>
I hope he doesn't leave, without him, we'd get one native decoder and filter a year
<Lynne>
if at all
* elenril
stabs BBB
dellas has joined #ffmpeg-devel
<BBB>
this is how wars start. someone sends an innocent patch trying to improve, someone frowns, someone's stabbed, and next thing you know a war breaks out
* elenril
throws a rotting trout at BBB
<BBB>
<3
<another|>
FFmpeg: Civil War
<elenril>
I don't know why everybody is being all panicky
Krowl has quit [Read error: Connection reset by peer]
<another|>
I'm not.
<elenril>
paul throws such tantrums regularly
<ePirat>
well not in this way usually
<BBB>
I know. but he's doing it to get a reaction, so we would do well to respond to it a little bit
<another|>
I was actually kind of relieved when I read his mail. He's been very loud, annoying and rude IMHO.
<elenril>
^
<ePirat>
if there is someone I would want to leave, some other name comes to my mind first though
<elenril>
people who cannot play nice with others are a liability, not an asset
<BBB>
paul v nicolas is indeed something that makes me seriously reconsider whether I want to be part of this project
<BBB>
or have my kids know I'm somehow part of this
<courmisch>
I think that there is a way to keep people like him - have a more automated system where you can't simply merge without the code having been sent for review for a while
<courmisch>
but of course that means moving on from the ML workflow
<Lynne>
I wouldn't accept such self-limits
<Lynne>
an understanding is far better
<Lynne>
and either way, what happened here was not due to that
<Lynne>
the project will die if it goes back to libav's requirement of at least 1 review of a long-time developer
<JEEB>
that's not what he said, also not what vlc is doing
<Lynne>
I think that simply switching away from a mailing list will immediately improve everything, right through to the community of developers
<JEEB>
vlc just has that if X days go on without anyone poking a thing, then it can go in or so
<courmisch>
3 days
<JEEB>
it just needs to go through MR/CI
<courmisch>
1 day *if* it was reviewed
<Lynne>
JEEB: an automated system where you can't merge *is* what I just said
<Lynne>
time limits like these are idiotic and I won't have them
<JEEB>
> at least 1 review of a long-time developer
mkver has joined #ffmpeg-devel
<JEEB>
that is not what is the case, you can merge w/o it
<Lynne>
libav, learn to read
<JEEB>
ok
<JEEB>
I just took the "if it goes back to libav's requirement"
<JEEB>
but sure, just don't get aggressive here. no need for that
<ePirat>
Lynne, I misunderstood what you meant as well…
<Lynne>
time limits like these would be like a parent buying their kid a computer but only allowing one hour a week of usage
tmm1 has quit [Read error: Connection reset by peer]
tmm1 has joined #ffmpeg-devel
<Lynne>
I'd agree with CI passing
<Lynne>
that works just fine for mesa
<Lynne>
even though it takes an hour to pass and frequently crashes
<courmisch>
no, it's admitting that the relevant reviewer is not necessarily the fastest person to write LGTM
<kierank>
BBB: happy birthday!!!!one
Krowl has joined #ffmpeg-devel
<BBB>
<3
<BBB>
ty
<courmisch>
my hot take is that patches should have at least one week time to review unless they're critical fixes
<courmisch>
it's all about being inclusive for people who can't review code on working days
<courmisch>
some of my VLC features have been waiting for a releasse for 4 years now. not like a week makes a difference
dellas has quit [Remote host closed the connection]
<cosminaught>
waiting a week on every change sounds painful, if it's not obviously broken it can be improved after being merged.
kurosu has quit [Quit: Connection closed for inactivity]
<JEEB>
currently it's something like that anyways :P since people have previously noted that you ping your patch set after a ~week
<JEEB>
which reminds me I need to ping one of my sets .-.
<Lynne>
courmisch: that's not a very good example
<courmisch>
how so? I can review patches almost everyday but that's because I don't have a family and I have decent working hours
<courmisch>
when I was in undergrad, I could only review patches on saturday afternoon or sunday morning
<BBB>
I don't think we need such fixed rules for everything. for stuff I myself maintain, 1-2 days is more than enough. if the maintainer asks for more time on specific changes or modules, that should be within the realm of possible also
<BBB>
we can be a bit dynamic guys
<BBB>
we're not 100yrs old yet
<Lynne>
courmisch: if you have an interest in a given piece of code, whether it's through being a maintainer of it, or frequent contributor to it, people would tag you, or you could set up an alert
<courmisch>
in an ideal world, yes.
<courmisch>
you don't have to agree with me, but you can't have the cake and eat it.
kurosu has joined #ffmpeg-devel
<Lynne>
nonsense, an antique expression that may have been valid in the bronze age, but not now
<Lynne>
we have computers, we can have the cake, eat it, and order more, for free
<courmisch>
in this computer age, the cake is a lie anyway, so you can't eat it
<another|>
cp -a cake cake2
<elenril>
cake is for traffic shaping, not eating
<BBB>
who pays for the delivery dude?
<Marth64[m]>
Can a demuxer "influence" a default encoder for a stream? I know it sounds strange. I have got DVD PCM working. The example here is when doing stream copy, Matroska won't take the DVD LPCM format, so user needs to manually specified to transcode (e.g. FLAC)
<Marth64[m]>
Not a big deal but I can see confusion
<elenril>
no
<BBB>
the "default" encoder (or rather, codec) is set by muxer, not demuxer
<Marth64[m]>
Understood. Not a serious issue.Thx!
<JEEB>
yea, and if you set codec copy
<JEEB>
then you are explicitly defining it
<BBB>
set vcodec copy and acodec will default to something useful and you can manually override it :)
<JEEB>
-c:v copy
<BBB>
or better, use av1. mpeg2 sucks /me runs
<Marth64[m]>
Ha yeah good point. As the user I already declared what I wanted, so it's on me
<elenril>
BBB: av1 has no good encoders
<elenril>
h264 is best
<BBB>
h264 sucks, x264 is pretty good though
<BBB>
h264 is only good because of x264
<Marth64[m]>
With PCM working I will send the patch to ML now and I had some questions maybe answered best over email. There's still kinks to work out but isn't there always.
<JEEB>
btw, what was a way to test mailmap?
<jamrial>
elenril: av1 has good encoders, just not open source :p
<elenril>
that makes them not good by definition =p
Krowl has quit [Read error: Connection reset by peer]
<Traneptora>
Marth64[m]: this is already the case when trying to remux blurays to matroska
<Traneptora>
if you just run -map 0 - copy -f matroska out.mkv, it'll fail. the user will need to run -c copy -c:a flac, or something
<Traneptora>
but that's not the responsibility of the demuxer, it's the responsibility of the user of ffmpeg
<Marth64[m]>
makes sense
<Traneptora>
especially considering that if you try the first one, it'll say "can't put pcm inside matroska" and then the user goes "I guess I'll try flac"
<Traneptora>
at least that was my thought process
<Marth64[m]>
It took me 1-2min to realize it but once I did it was obvious. I got spoiled by stream copy working for everything else.
<courmisch>
elenril: meaning hardware encoders are bad or they have to be open-source? open-source hardware is a lie
<Marth64[m]>
I've got to wait 45 mins to test a PAL DVD, then this is getting sent in
<elenril>
courmisch: of course they are bad
<courmisch>
they are more energy efficient
<courmisch>
and require fewer gates for the same speed
<elenril>
orphans may make a good source of organs, but it's still evil to use them that way
<courmisch>
but what about global warming?
<elenril>
I would assume the difference to be far below measurable
<courmisch>
elenril: and how is that worse than proprietary CPUs?
<elenril>
proprietary CPUs are evil too
<elenril>
obviously
<courmisch>
so where's that open-source RV64GQV design of yours?
<courmisch>
or was it supposed to be RV128GQ?
<Traneptora>
is there actually a measurable difference in the electricity used by hardware vs software encoding video
<Traneptora>
even for a data center it's gotta be a small fraction. it's not like gpus cryptomining
<elenril>
for phones it is massive AFAIK
<courmisch>
Traneptora: why do you think mobile devices have hardware encoders and decoders?
<Traneptora>
yea but in that case power consumption matters for battery life,
<Traneptora>
not for, well, global warming reasons
<courmisch>
totally for global warming reasons, totally not greenwashing
* courmisch
totally does not work for a mobile phone compan... err
<Traneptora>
crypto farms actually have a measurable impact on electricity use. just need to power them with a windmill
<BBB>
nicolas suffers from the classic pathological condition of needing to have the last word
<BBB>
he can't just end a discussion without having been the last to talk and make a point
<Traneptora>
I just straight up told him I wouldn't respond to his inevitable ramble
<BBB>
he believes that not having the last word means he lost the argument
<Traneptora>
and I'm gonna stick to that
<BBB>
poor nicolas. I hope he grows out of that at some point
<courmisch>
acute ultimoverbitis
<elenril>
highly unlikely
<Traneptora>
I used to be like that online but I was like, 15
<BBB>
I express a wish, I did not associate a probability with it
<elenril>
he's way too old to grow out of it
<Traneptora>
some arguments just straight up aren't worth it
<elenril>
anyway, I wish people would just stop replying to that thread
<elenril>
it serves no useful purpose
<BBB>
ok
<courmisch>
actually chronic ultimoverbitis, sorry I don't have a medical degree
<elenril>
michaelni: are you going to do more testing for CLI threading set, or was that broken case the only one?
deus0ww has quit [Ping timeout: 256 seconds]
<j-b>
courmisch: " To be fair, he also abundantly abused JB there, although I do not know how much of it was sarcasm vs actual attack." I consider all of that as a joke, tbh.
deus0ww has joined #ffmpeg-devel
MrZeus_ has joined #ffmpeg-devel
* elenril
abuses j-b with a large trout
dellas has joined #ffmpeg-devel
<j-b>
I prefer honey
___nick___ has joined #ffmpeg-devel
<courmisch>
elenril: hentai
___nick___ has quit [Client Quit]
___nick___ has joined #ffmpeg-devel
microchip_ has quit [Quit: There is no spoon!]
microchip_ has joined #ffmpeg-devel
kurosu has quit [Quit: Connection closed for inactivity]
ccawley2011 has quit [Remote host closed the connection]
ccawley2011 has joined #ffmpeg-devel
ccawley2011 has quit [Remote host closed the connection]
ccawley2011 has joined #ffmpeg-devel
mkver has quit [Ping timeout: 268 seconds]
AbleBacon has joined #ffmpeg-devel
ccawley2011 has quit [Read error: Connection reset by peer]