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 7.1 has been released! | Please read ffmpeg.org/developer.html#Code-of-conduct
sunyuechi has quit [Remote host closed the connection]
sunyuechi has joined #ffmpeg-devel
omegatron has quit [Quit: Power is a curious thing. It can be contained, hidden, locked away, and yet it always breaks free.]
sunyuechi has quit [Remote host closed the connection]
sunyuechi has joined #ffmpeg-devel
sunyuechi has quit [Remote host closed the connection]
sunyuechi has joined #ffmpeg-devel
sunyuechi has quit [Remote host closed the connection]
sunyuechi has joined #ffmpeg-devel
sunyuechi has quit [Remote host closed the connection]
sunyuechi has joined #ffmpeg-devel
Kei_N_ has joined #ffmpeg-devel
Kei_N has quit [Ping timeout: 260 seconds]
sunyuechi has quit [Remote host closed the connection]
sunyuechi has joined #ffmpeg-devel
sunyuechi has quit [Remote host closed the connection]
sunyuechi has joined #ffmpeg-devel
Marth64 has quit [Quit: Leaving]
iive has quit [Quit: They came for me...]
mkver has quit [Ping timeout: 246 seconds]
sunyuechi has quit [Remote host closed the connection]
sunyuechi has joined #ffmpeg-devel
arch1t3cht5 has joined #ffmpeg-devel
arch1t3cht has quit [Ping timeout: 244 seconds]
arch1t3cht5 is now known as arch1t3cht
sunyuechi has quit [Remote host closed the connection]
sunyuechi has joined #ffmpeg-devel
sunyuechi has quit [Remote host closed the connection]
sunyuechi has joined #ffmpeg-devel
Marth64 has joined #ffmpeg-devel
sunyuechi has quit [Remote host closed the connection]
sunyuechi has joined #ffmpeg-devel
thilo has quit [Ping timeout: 265 seconds]
thilo has joined #ffmpeg-devel
sunyuechi has quit [Remote host closed the connection]
sunyuechi has joined #ffmpeg-devel
sunyuechi has quit [Remote host closed the connection]
sunyuechi has joined #ffmpeg-devel
IndecisiveTurtle has quit [Quit: IndecisiveTurtle]
sunyuechi has quit [Remote host closed the connection]
sunyuechi has joined #ffmpeg-devel
sunyuechi has quit [Remote host closed the connection]
sunyuechi has joined #ffmpeg-devel
sunyuechi has quit [Remote host closed the connection]
Martchus_ has joined #ffmpeg-devel
Martchus has quit [Ping timeout: 246 seconds]
<fflogger>
[newticket] MasterQuestionable: Ticket #11334 ([trac] [Trac] Wiki content format regression) created https://trac.ffmpeg.org/ticket/11334
<fflogger>
[editedticket] lanczos-algorithm: Ticket #11317 ([avcodec] Can AVCodec provide more crop information when decoding via MediaCodec?) updated https://trac.ffmpeg.org/ticket/11317#comment:4
haihao has quit [Ping timeout: 260 seconds]
<Lynne>
elenril: still trying to figure out why calling frame_params early causes a failure allocating after porting everything properly
haihao has joined #ffmpeg-devel
<Lynne>
there should be no state at all being transferred over between params and init
mkver has joined #ffmpeg-devel
Martchus has joined #ffmpeg-devel
Martchus_ has quit [Ping timeout: 248 seconds]
ngaullier has joined #ffmpeg-devel
ccawley2011 has joined #ffmpeg-devel
<haasn>
ramiro: yes, after discussion on IRC
<haasn>
BBB: which patch?
ccawley2011_ has joined #ffmpeg-devel
Sean_McG has quit [Ping timeout: 260 seconds]
ccawley2011__ has joined #ffmpeg-devel
ccawley2011 has quit [Ping timeout: 246 seconds]
ccawley2011 has joined #ffmpeg-devel
ccawley2011_ has quit [Ping timeout: 252 seconds]
ccawley2011_ has joined #ffmpeg-devel
ccawley2011__ has quit [Ping timeout: 252 seconds]
ccawley2011 has quit [Ping timeout: 252 seconds]
Martchus_ has joined #ffmpeg-devel
Martchus has quit [Ping timeout: 248 seconds]
jamrial has joined #ffmpeg-devel
<haasn>
JEEB: derp, I did my previous benchmarks on a build with asm disabled
MyNetAz has quit [Read error: Connection reset by peer]
ccawley2011__ has joined #ffmpeg-devel
ccawley2011_ has quit [Ping timeout: 252 seconds]
MyNetAz has joined #ffmpeg-devel
ccawley2011 has quit [Ping timeout: 252 seconds]
Martchus_ has quit [Ping timeout: 248 seconds]
Sean_McG has joined #ffmpeg-devel
<thardin>
there we go, finally got my avio wrapping thing to clean up nicely. but I needed to add a dummy URLProtocol that opens /dev/null. guess I should add a null protocol=
<thardin>
no, better idea: add a way to specify a different close function for pb->opaque than ffurl_close()
ccawley2011 has joined #ffmpeg-devel
ccawley2011__ has quit [Ping timeout: 248 seconds]
ccawley2011_ has joined #ffmpeg-devel
ccawley2011 has quit [Ping timeout: 252 seconds]
<thardin>
there we go, sanity
Martchus has joined #ffmpeg-devel
ccawley2011 has joined #ffmpeg-devel
ccawley2011_ has quit [Ping timeout: 252 seconds]
Martchus_ has joined #ffmpeg-devel
Martchus has quit [Ping timeout: 252 seconds]
Martchus_ has quit [Ping timeout: 248 seconds]
Martchus has joined #ffmpeg-devel
ccawley2011_ has joined #ffmpeg-devel
ccawley2011 has quit [Ping timeout: 252 seconds]
ccawley2011 has joined #ffmpeg-devel
ccawley2011__ has joined #ffmpeg-devel
ccawley2011_ has quit [Ping timeout: 260 seconds]
ccawley2011 has quit [Ping timeout: 252 seconds]
Martchus has quit [Ping timeout: 252 seconds]
ccawley2011 has joined #ffmpeg-devel
ccawley2011__ has quit [Ping timeout: 276 seconds]
Martchus has joined #ffmpeg-devel
Martchus_ has joined #ffmpeg-devel
Martchus has quit [Ping timeout: 265 seconds]
<Traneptora>
haasn: do you have any plans for vf_colorspace
<Traneptora>
which is a filter that does a lot of this stuff. possibly strip it and have it call swscale?
Marth64 has quit [Quit: Leaving]
ccawley2011_ has joined #ffmpeg-devel
ccawley2011 has quit [Ping timeout: 265 seconds]
<fflogger>
[editedticket] MasterQuestionable: Ticket #11333 ([undetermined] Incomplete conversion of certain Apple H.265 MOV "hstack") updated https://trac.ffmpeg.org/ticket/11333#comment:7
ccawley2011 has joined #ffmpeg-devel
ccawley2011_ has quit [Ping timeout: 246 seconds]
Marth64 has joined #ffmpeg-devel
<haasn>
Traneptora: obsoleting it?
<haasn>
Yeah
<haasn>
Well we still need a dynamic peak detection shader
<haasn>
Filter, rather
Traneptora has quit [Remote host closed the connection]
ngaullie has joined #ffmpeg-devel
Traneptora has joined #ffmpeg-devel
<Traneptora>
haasn: probably should document that it's obsolete. can't just remove it cause spacebar heating but you can add deprecation warnings
<Traneptora>
in offtopic news my X server crashed which I've never had happen before
<haasn>
One step at a time
<haasn>
But yeah I agree on principle
<haasn>
Let’s let the new code settle a bit
<Traneptora>
true, just wanted to point out the issue
<haasn>
Probably it’s also way faster
<haasn>
Technically you can’t replicate vf_colorspace using vf_scale
<haasn>
Because it contains a very old algo
Traneptora has quit [Remote host closed the connection]
<haasn>
17:47 <haasn> Probably it’s also way faster <- to be clear, vf_scale is way faster
ngaullier has quit [Ping timeout: 248 seconds]
Traneptora has joined #ffmpeg-devel
Martchus has joined #ffmpeg-devel
<Traneptora>
apparently if you run a parallel job on a batch image process thing that eats a ton of RAM, and oomkiller kicks in, instead of killing parallel, it'll kill Xorg
<Traneptora>
for some reason
Martchus_ has quit [Ping timeout: 265 seconds]
<Traneptora>
haasn: also question (I put this in the review email) but why von kries? the page you linked cites bradford as being the "best"
<Traneptora>
that said, if you're starting in a D65 space and going to a different D65 space, does it matter if you roundtrip to D50
ccawley2011_ has joined #ffmpeg-devel
ccawley2011__ has joined #ffmpeg-devel
ccawley2011 has quit [Ping timeout: 244 seconds]
Mirarora has quit [Quit: Mirarora encountered a fatal error and needs to close]
ccawley2011_ has quit [Ping timeout: 248 seconds]
DEATH has quit [Ping timeout: 265 seconds]
DEATH has joined #ffmpeg-devel
ccawley2011 has joined #ffmpeg-devel
Martchus has quit [Ping timeout: 248 seconds]
Martchus_ has joined #ffmpeg-devel
ccawley2011_ has joined #ffmpeg-devel
ccawley2011__ has quit [Ping timeout: 244 seconds]
ccawley2011 has quit [Ping timeout: 276 seconds]
Martchus_ has quit [Ping timeout: 244 seconds]
<haasn>
Traneptora: it's a different one
<haasn>
there are multiple "von Kries" methods
<haasn>
iirc that's the general term for using a single 3x3 linear transform to get from XYZ to RGB
<haasn>
(as opposed to a nonlinear transform like e.g. CIECAM02 uses)
<haasn>
the matrix we use is taken from CAT16 and I recommend reading that document if you want to learn more
Martchus has joined #ffmpeg-devel
<Traneptora>
I've read that page about white point adaptation a few months ago
<Traneptora>
though I don't remember specifically all the details
ccawley2011 has joined #ffmpeg-devel
ccawley2011__ has joined #ffmpeg-devel
Martchus has quit [Ping timeout: 252 seconds]
ccawley2011_ has quit [Ping timeout: 244 seconds]
ngaullie has quit [Ping timeout: 244 seconds]
Martchus has joined #ffmpeg-devel
<haasn>
Traneptora: we actually use d65 as standard
<haasn>
I just picked it because it was the newest CIE standard
<Traneptora>
also as far as I understand this linear transform from gamut to gamut has to happen in linear light right
<haasn>
then it backtracks and shows a table where CAT16 is the one with the best match to the perceptual data set
<haasn>
yes
<haasn>
anyway I'm sure it's fine and the differences are subtle
<Traneptora>
how do we handle gamut mapping mode? as far as I understand the linear transform maps just treats out of gamut colors as-is (i.e. with float values out of [0.0, 1.0])
<Traneptora>
or is that handled later on in the patch (I haven't reviewed the whole set yet)
<haasn>
I decided to tie both tone and gamut mapping to a single ICC intent specification
<haasn>
not sure what you mean by gamut mapping mode
<haasn>
you mean the question of whether to gamut map in RGB or luma or whatever?
<BBB>
"I'm told the FFMPEG nightly was the reason martial law was declared this week, how do you respond to these allegations?" omg twitter what
Martchus_ has quit [Ping timeout: 246 seconds]
Martchus has joined #ffmpeg-devel
ccawley2011 has joined #ffmpeg-devel
ccawley2011__ has quit [Ping timeout: 252 seconds]
ngaullie has joined #ffmpeg-devel
ngaullie has quit [Ping timeout: 248 seconds]
ccawley2011_ has joined #ffmpeg-devel
ccawley2011 has quit [Ping timeout: 265 seconds]
<welder>
Hi, is there a guide on how to write a filter? I'd like to write an audio source reading from pipewire stream.
fflogger has quit [Ping timeout: 252 seconds]
ccawley2011 has joined #ffmpeg-devel
ccawley2011_ has quit [Ping timeout: 248 seconds]
<another|>
filter?
<welder>
There's a patch, year old or so, where the reviewer suggested that instead of -f pipewire
<jamrial>
BtbN: fwiw, after the nvEncGetSequenceParams() issue is fixed, the mov muxer needs to be updated to actually include the layer > 0 parameter set nalus in hvcC
<BtbN>
oh, that strips those too?
<BtbN>
There's people who reported successfully remuxing the apple provided hevc alpha sample with ffmpeg
<jamrial>
it's a bit annoying because the L-HEVC in ISOBMFF spec expects all nuh_layer_id > 0 NALUs to be in a separate box called lhvC
<jamrial>
for things like multivier/stereo
<jamrial>
but apple's alpha spec just puts them in hvcC
<BtbN>
so the muxer would somehow need to differentiate those, hm
<jamrial>
right now it looks for the MULTIVIEW stream disposition
<BtbN>
libx265 alpha channel support also looks like a hot mess to add
<BtbN>
mostly for lack of documentation on how to actually use it
<BtbN>
hm, at least from a quick looks, it should be fine? Nothing would set AV_DISPOSITION_MULTILAYER, so it should just write it with mov_write_hvcc_tag
<BtbN>
So it would _somehow_ need to keep trac of the layer 1 being an Alpha layer, and if so, don't stop there. Hm
<jamrial>
BtbN: if the sps/pps have no actual information about them being for an alpha layer (which we could parse in lavf), then i guess we'll need something like the multiview disposition to signal this
<BtbN>
The main sps/pps set has information that other set is for an alpha layer
<BtbN>
But I don't immediately see anything in the sps/pps itself indicating that it itself is for an alpha layer
<fflogger>
[newticket] mibby: Ticket #11335 ([avutil] av_log callback called with bad parameters (NULL AVClass)) created https://trac.ffmpeg.org/ticket/11335
<jamrial>
BtbN: ok so, we can know if a layer is alpha by looking at the VPS. it should report scalabilityid 3 (auxiliary) and auxid[layer 1] = 1 (AUX_ALPHA)
<jamrial>
or so
<BtbN>
Yeah, I'm looking at it right now. It's a bit annoying, since A TON of vps parsing needs to be added to hevc.c in avformat
<BtbN>
And I can't look at the spec, since it's unavailable