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
<BtbN>
hm, are there any averse effects to changing the timestamps a demuxer returns from 1/1000 to 1/1000000000? Cause that's the precision flv now in theory supports.
<BtbN>
changing the timebase I mean
<Lynne>
vlc ran into issues when we considered changing the default matroska timebase years ago
<BtbN>
the problem is, there is no way to know in advance if a stream will have ns precision timestamps, or not
HarshK23 has quit [Quit: Connection closed for inactivity]
<BtbN>
Either there is a nanosecond offset, or there isn't
<BtbN>
So in theory they could start appearing an hour in...
<jamrial>
that sounds like a good argument for changing it, then, assuming it doesn't break parsing old files
<BtbN>
They'd just have a lot of 0s at the end of their timestamps
<BtbN>
At the moment, I'd be inclined to just hold off on switching the timebase
<BtbN>
Cause it's VERY unlikely such files even exist yet
<BtbN>
So the support I added just is aware of it, converts it to ms (so likely to 0), and adds it
IndecisiveTurtle has joined #ffmpeg-devel
thilo has quit [Ping timeout: 264 seconds]
thilo has joined #ffmpeg-devel
thilo has quit [Changing host]
thilo has joined #ffmpeg-devel
<Lynne>
BtbN: could you reply to the "avutil/vulkan_glslang: windows build improvements" thread?
<BtbN>
I switched away from glslang to shaderc
<BtbN>
Cause glslang was such a major pain to keep working
<BtbN>
Then again, it's difficult... offering a way to query it via pkg-config at least gives users an option to build a working environment with a custom pkg-config file, whereas with hardcoded libs in configure will always require patching configure
<BtbN>
mind you, shaderc is not all that much better, it's still a royal pain to deal with. But not quite as bad as glslang
<BtbN>
I'll look into it in more detail tomorrow
System_Error has quit [Remote host closed the connection]
<BtbN>
I also don't use msvc at all. THAT sounds like just pure pain to get working
<BtbN>
.pc files should work fine for C++ libraries though, as long as the libraries properly add -lstdc++ to Libs.private
<BtbN>
Now, how well that translates to MSVC... no clue
mkver has quit [Ping timeout: 252 seconds]
System_Error has joined #ffmpeg-devel
_av500_ has joined #ffmpeg-devel
av500 has quit [Ping timeout: 252 seconds]
System_Error has quit [Remote host closed the connection]
Martchus has joined #ffmpeg-devel
Martchus_ has quit [Ping timeout: 252 seconds]
Martchus_ has joined #ffmpeg-devel
Martchus has quit [Ping timeout: 265 seconds]
System_Error has joined #ffmpeg-devel
Xaldafax has quit [Quit: Bye...]
jamrial has quit []
Traneptora has joined #ffmpeg-devel
steven-netint has quit [Ping timeout: 265 seconds]
System_Error has quit [Remote host closed the connection]
Martchus has joined #ffmpeg-devel
Martchus_ has quit [Ping timeout: 272 seconds]
chinmay is now known as cinmay
rvalue- has joined #ffmpeg-devel
rvalue has quit [Ping timeout: 260 seconds]
rvalue- is now known as rvalue
System_Error has joined #ffmpeg-devel
jluthra has quit [*.net *.split]
Nightrose has quit [*.net *.split]
jkkm has quit [*.net *.split]
Dmitri_Ovch has quit [*.net *.split]
tortoise has quit [*.net *.split]
bossjones has quit [*.net *.split]
tmatth has quit [*.net *.split]
jluthra has joined #ffmpeg-devel
jkkm has joined #ffmpeg-devel
Dmitri_Ovch has joined #ffmpeg-devel
Nightrose has joined #ffmpeg-devel
tortoise has joined #ffmpeg-devel
bossjones has joined #ffmpeg-devel
tmatth has joined #ffmpeg-devel
Compn has quit [Remote host closed the connection]
Compn has joined #ffmpeg-devel
Kei_N has joined #ffmpeg-devel
Kei_N_ has quit [Read error: Connection reset by peer]
HarshK23 has joined #ffmpeg-devel
DodoGTA has left #ffmpeg-devel [#ffmpeg-devel]
abdu has joined #ffmpeg-devel
abdu has quit [Client Quit]
abdu has joined #ffmpeg-devel
abdu has quit [Quit: Client closed]
abdu has joined #ffmpeg-devel
DodoGTA has joined #ffmpeg-devel
aaabbb has quit [Ping timeout: 252 seconds]
aaabbb has joined #ffmpeg-devel
pross has quit [Ping timeout: 252 seconds]
pross has joined #ffmpeg-devel
<haasn>
the alpha channel in yuva is always full range?
DodoGTA has quit [Quit: DodoGTA]
<Lynne>
yeah, I think so, I remember this being discussed long ago
DodoGTA has joined #ffmpeg-devel
mkver has joined #ffmpeg-devel
kasper93 has quit [Read error: Connection reset by peer]
kasper93 has joined #ffmpeg-devel
witchymary has quit [Ping timeout: 252 seconds]
witchymary has joined #ffmpeg-devel
jamrial has joined #ffmpeg-devel
rvalue has quit [Ping timeout: 252 seconds]
rvalue- has joined #ffmpeg-devel
rvalue- is now known as rvalue
ccawley2011 has joined #ffmpeg-devel
jamrial has quit []
jamrial has joined #ffmpeg-devel
microchip_ has joined #ffmpeg-devel
putacho has joined #ffmpeg-devel
microchip_ has quit [Client Quit]
elChippo has quit [Ping timeout: 244 seconds]
putacho is now known as elChippo
<haasn>
nice
<haasn>
my C "slow path" outpeforms swscale's asm by 4.4x
<haasn>
simply because it doesn't roundtrip through planar yuv
<haasn>
for rgba -> rgba scaling
<haasn>
didn't even write any fast paths yet for combined reading-and-scaling
<haasn>
so this is: read packed rgba input, write to temporary buffer
<haasn>
sample from temporary buffer and write to packed rgba output
<haasn>
as four separate steps
<haasn>
but with a combined read-and-sample path we can skip the temporary buffer and get another 2x
<haasn>
pretty excited about the way my shitty C "PoC" is coming along
<Lynne>
it does what?
<Lynne>
how messed up was swscale to need a yuv step for simple rgba rescaling?
<nevcairiel>
swscale just has one very generic path, and if there isnt a fast path for your request it'll use it
<Lynne>
yeah, generally the generic path would be through xyz or rgb, not something with limited gamut like yuv
<kepstin>
doesn't yuv (well, YCbCr)have the exact same gamut as whatever rgb space it's derived from tho? certainly a possible loss of precision issue to convert to yuv then back to rgb.
<kepstin>
huh, apparently the way yuv transforms are defined, the yuv color spaces end up being able to represent some colors that don't fit int he original rgb space.
ccawley2011 has quit [Ping timeout: 244 seconds]
Gramner has quit [Remote host closed the connection]
_av500_ is now known as av500
Gramner has joined #ffmpeg-devel
abdu has quit [Ping timeout: 240 seconds]
abdu has joined #ffmpeg-devel
<Compn>
ffcolorspace
<Compn>
should patent some new colors.
Marth64 has joined #ffmpeg-devel
abdu96 has joined #ffmpeg-devel
abdu has quit [Ping timeout: 240 seconds]
ccawley2011 has joined #ffmpeg-devel
<Traneptora>
kepstin: they have the same gamut, but colors outside of [0.0, 1.0] are out-of-gamut colors
<Traneptora>
and the YUV transform matrix doesn't always map the unit cube to the unit cube
<sdc>
somewhat offtopic but I've been reading the jpeg spec and implementing a decoder based off of it, with the goal of getting more familiar with video codecs eventually. in the process of finishing up decoding progressive jpegs.
<sdc>
I was thinking about trying to make the jump to mpeg1 next but was wondering if anyone had any other suggestions for next steps or general approach?
ccawley2011 has joined #ffmpeg-devel
ccawley2011__ has joined #ffmpeg-devel
ccawley2011_ has quit [Ping timeout: 245 seconds]
ccawley2011 has quit [Ping timeout: 276 seconds]
abdu96 has quit [Quit: Client closed]
<sdc>
more on topic, I finished up a v1 for simd vvc chroma deblocking that I'm waiting to get reviewed by nuo mi before submitting to the ml if anyone wants to take a look https://github.com/ffvvc/FFmpeg/pull/275/files
<sdc>
any feedback appreciated!
abdu96 has joined #ffmpeg-devel
abdu43 has joined #ffmpeg-devel
abdu96 has quit [Ping timeout: 240 seconds]
System_Error has quit [Remote host closed the connection]
System_Error has joined #ffmpeg-devel
Marth64 has joined #ffmpeg-devel
abdu43 has quit [Quit: Client closed]
abdu43 has joined #ffmpeg-devel
<Traneptora>
sdc: may be a good idea to send to ML anyway
<Traneptora>
so people can review it there
HarshK23 has quit [Quit: Connection closed for inactivity]
DauntlessOne4 has quit [Read error: Connection reset by peer]
DauntlessOne4 has joined #ffmpeg-devel
DauntlessOne4 has quit [Remote host closed the connection]
DauntlessOne4 has quit [Read error: Connection reset by peer]
ccawley2011__ has quit [Ping timeout: 252 seconds]
<BBB>
should we maybe get a replacement in the CC until jamrial's back? or is the CC OK as-is?
<BBB>
I guess I'm an outsider so I'll let current CC decide
<jamrial>
no, it's not ok as is
DauntlessOne4 has joined #ffmpeg-devel
<Marth64>
I am hoping not and don't think I've done anything wrong but if I am in any way the issue please let me know.
<jamrial>
you're not, why do you think so?
<jamrial>
when i say it's not ok as is it's because i resigned, and anton is unlikely to participate right now
<jamrial>
so there's three people
<Marth64>
Thank you for clarifying. I just wanted to ensure. I hope the best for you guys.
<Marth64>
being the overall newcomer I wasn't sure
<Traneptora>
Marth64: i'm sure you're aware of the drama going on wrt running the project rn. but you definitely aren't the source of the drama, or fuel on the fire
<Traneptora>
I'm trying very hard to stay out of it as much as possible. I have strong opinions but I don't think they'll be useful
<Marth64>
fair enough and understandable. I'm trying to do my job faithfully by (1) helping with localized / privately reported issues and (2) trying to schedule a meeting with whats left of the CC to go organize and prioritize the community problems at large. I am hoping to make a difference once we can meet. I inferred that people are probably travelling and tied up making it hard to connect right now
<Marth64>
tied up with travels*
<Traneptora>
as far as I am aware though most in-person meetings are in Europe
<Traneptora>
so those of us (like me) in the United States have a bigger overhead to travel
Mirarora has quit [Quit: Mirarora encountered a fatal error and needs to close]
<Marth64>
Travel is hard for me too, I can/will settle for virtual or even text based on IRC
<Marth64>
Anyhow, hope everyone is doing well. I have had some IRL stuff pop up but will pick up tech contributions again soon