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 7.0.1 has been released! | Please read ffmpeg.org/developer.html#Code-of-conduct
iive has quit [Quit: They came for me...]
System_Error has joined #ffmpeg-devel
System_Error has quit [Remote host closed the connection]
System_Error has joined #ffmpeg-devel
thilo_ has quit [Ping timeout: 268 seconds]
thilo_ has joined #ffmpeg-devel
* Sean_McG
peeks in
lexano has quit [Ping timeout: 264 seconds]
mateo` has quit [Ping timeout: 268 seconds]
cone-330 has joined #ffmpeg-devel
<cone-330>
ffmpeg Lynne master:208dd9fdea32: aacdec_usac: prevent get_bits(0) in get_escaped_value
<cone-330>
ffmpeg Lynne master:2554c5e39ede: aacdec_usac: error out on too many elements
<cone-330>
ffmpeg Lynne master:34e92e98c49e: aacdec_usac: error out on receiving too many channels for the current layout
<cone-330>
ffmpeg Lynne master:8ad8b5e63e28: aacdec_usac: fix off by one error check on channel numbers
<cone-330>
ffmpeg Lynne master:42e7dfe5a664: aacdec_usac: fix typo in debug message
<cone-330>
ffmpeg Lynne master:b70fdbde9cef: aacdec_usac: prevent get_bits(0) in get_escaped_value
<cone-330>
ffmpeg Lynne master:39826eacff9a: aacdec_usac: error out on too many elements
<cone-330>
ffmpeg Lynne master:0d93e04bc627: aacdec_usac: error out on receiving too many channels for the current layout
<cone-330>
ffmpeg Lynne master:f4371029f88d: aacdec_usac: fix off by one error check on channel numbers
<cone-330>
ffmpeg Lynne master:7413b81e50c6: aacdec_usac: fix typo in debug message
<cone-330>
ffmpeg James Almer master:a14440867c60: x86/float_dsp: add SSE2 and AVX versions of scalarproduct_double
Krowl has quit [Read error: Connection reset by peer]
<mkver>
Imagine one has to do the following: Write to a buffer, then read the data again and write again to the buffer (read again+write is an idct_add). When would it make sense (from a caching point of view) to perform this in an auxiliary buffer and then copy the data from this buffer into its final destination?
georgereynolds8 has quit [Ping timeout: 252 seconds]
georgereynolds8 has joined #ffmpeg-devel
ccawley2011 has joined #ffmpeg-devel
Krowl has joined #ffmpeg-devel
mateo` has joined #ffmpeg-devel
<elenril>
mkver: if you meant passlogfile, then it seems libx264 also has that option
mateo` has quit [Read error: Connection reset by peer]
<mkver>
elenril: passlogfile and stats
<elenril>
seems both were copied from libx264
mcfrdy has quit [Quit: quit]
stevenliu has joined #ffmpeg-devel
mateo` has joined #ffmpeg-devel
<elenril>
Lynne: wasn't there a news entry for aac?
mateo` has quit [Ping timeout: 260 seconds]
mateo` has joined #ffmpeg-devel
cone-902 has joined #ffmpeg-devel
<cone-902>
ffmpeg Anton Khirnov master:74159cbfc30a: lavc/hevcdec: move handling of byte alignment at the end of slice header
<cone-902>
ffmpeg Anton Khirnov master:79c0310acaf3: lavc/hevcdec: drop a useless condition
<cone-902>
ffmpeg Anton Khirnov master:67ca18dd5695: lavc/hevcdec: drop HEVCLocalContext.gb
<cone-902>
ffmpeg Anton Khirnov master:ac69e6caf6d9: lavc/hevcdec: include first row in SliceHeader.offset/size
<cone-902>
ffmpeg Anton Khirnov master:7ad9400952c0: lavc/hevcdec: drop HEVCContext.HEVClc
<cone-902>
ffmpeg Anton Khirnov master:9226514ceded: lavc/hevcdec: rename HEVCContext.ref to cur_frame
<cone-902>
ffmpeg Anton Khirnov master:db84c1c6eff9: lavc/hevcdec: rename HEVCFrame.frame to just f
<cone-902>
ffmpeg Anton Khirnov master:ba56a300a94b: lavc/hevcdec: drop HEVCContext.frame
<cone-902>
ffmpeg Anton Khirnov master:e4601cc3390e: lavc/hevc*: move to hevc/ subdir
<cone-902>
ffmpeg Anton Khirnov master:a13b89208081: lavc/hevcdec: deduplicate calling hwaccel decode_params()
<cone-902>
ffmpeg Anton Khirnov master:9576a005271e: lavc/hevcdec: drop unused HEVCContext.width/height
<nevcairiel>
lack of ucrt should directly complain about missing deps and not "crash", of course users are sometimes very imprecise
AbleBacon has joined #ffmpeg-devel
Krowl has quit [Read error: Connection reset by peer]
<BtbN>
yes, everything is "doesn't work" or "crashes"
<BtbN>
even though the "crash" is a clean exit with precise error message
<JEEB>
yup
<BtbN>
The tls_* wrappers in avformat have no maintainer?
Livio has joined #ffmpeg-devel
<JEEB>
yea, I don't think there's an explicit person. f.ex. I think for schannel one of the last people who've touched it was me >_>
<JEEB>
anyways, I am attempting to verify your schannel patch, I gave a look at sfan5's mbedtls patch set yesterday
* JEEB
screams at libbluray again
<JEEB>
le dec_init
<Lynne>
I wish libbluray would work reliably without having to mount ISOs
<JEEB>
I just have the disk file structures around in a directory
<Lynne>
between it and our mpegts demuxer, its free fuzzing for video codecs though
Krowl has joined #ffmpeg-devel
Livio has quit [Ping timeout: 256 seconds]
<nevcairiel>
libbluray pretends it can read UDF ISOs directly, i never tried though
<BtbN>
JEEB: I'm particularily unsure if it's valid for the write function. Since it'd re-encrypt on the next attempt.
<BtbN>
But I also wouldn't know how _else_ to handle it, since the function can't know if it's a re-call with the same data, or fresh data
<JEEB>
yea
microchip__ has joined #ffmpeg-devel
<BtbN>
looking at the EncryptMessage function, it has a seqNo parameter. Which might work? But it says not to use it with SSP, which I think is the TLS stuff
microchip_ has quit [Ping timeout: 268 seconds]
microchip__ is now known as microchip_
Livio has joined #ffmpeg-devel
<BtbN>
My only other idea would be to first try a non-blocking zero-byte-write to the tcp socket, and if that succeeds, go on but turn of non-block
HarshK23 has joined #ffmpeg-devel
Livio has quit [Ping timeout: 268 seconds]
Livio has joined #ffmpeg-devel
AbleBacon has quit [Read error: Connection reset by peer]
psykose has joined #ffmpeg-devel
microchip_ has quit [Quit: There is no spoon!]
Krowl has quit [Read error: Connection reset by peer]
ngaullie has quit [Ping timeout: 268 seconds]
ccawley2011 has quit [Ping timeout: 260 seconds]
microchip_ has joined #ffmpeg-devel
ccawley2011 has joined #ffmpeg-devel
<Lynne>
that suckless format, no yuv, no subsampling, no colorspaces, no animated pictures, big-endian only, rgba-only
psykose_ has joined #ffmpeg-devel
<Lynne>
20 GOTO STONE AGE 10
psykose has quit [Ping timeout: 256 seconds]
psykose_ is now known as psykose
<BtbN>
Why not just use bmp?
microchip_ has quit [Ping timeout: 240 seconds]
<Lynne>
palette
microchip_ has joined #ffmpeg-devel
* Sean_McG
snickers
cone-902 has quit [Quit: transmission timeout]
ravenJPL has joined #ffmpeg-devel
ccawley2011 has quit [Ping timeout: 252 seconds]
Traneptora has quit [Ping timeout: 246 seconds]
Traneptora has joined #ffmpeg-devel
ccawley2011 has joined #ffmpeg-devel
Hobbyboy has quit [Quit: The BNC has broken!]
Hobbyboy has joined #ffmpeg-devel
microchip__ has joined #ffmpeg-devel
microchip_ has quit [Ping timeout: 256 seconds]
HarshK23 has quit [Quit: Connection closed for inactivity]
microchip__ is now known as microchip_
<courmisch>
how much is jamrial getting paid by Intel and AMD to make Arm look bad like this?
<jamrial>
lol
<jamrial>
i'm literally picking low hanging fruit
<jamrial>
this last one is actually useful, for a change
<courmisch>
uh, rgb24 to YUV? aren't screen captures padded to rgb32 ?
<Lynne>
courmisch: is there a reason you're not giving X60 timings for the vc1 dct?
<BBB>
rgb24_to_uv_1080_ssse3: 876.8
<BBB>
rgb24_to_uv_1080_avx: 887.8
<BBB>
?
<BBB>
oh avx
<BBB>
nevermind
<BBB>
I'm blind
<BBB>
cool results
<BBB>
why do we test w=8 :)
<courmisch>
Lynne: RDCYCLE does not work on X60
<courmisch>
and since this is all fixed size stuff that fits in 128-bit vectors, X60 is not going to look any better than C908
<Lynne>
ah, right
<courmisch>
there's probably a way to get it work involving some u-boot and/or SBI patching but err
microchip_ has quit [Quit: There is no spoon!]
mkver has joined #ffmpeg-devel
microchip_ has joined #ffmpeg-devel
Krowl has joined #ffmpeg-devel
<courmisch>
not supposed to multiply 8 bits by 32 bits
philipl has quit [Quit: leaving]
<courmisch>
hmm, yeah okay I should probably use FMA
philipl has joined #ffmpeg-devel
<courmisch>
UV index is now 0 has the sun is down. RGB to UltraViolet will have to wait tomorrow.
Krowl has quit [Read error: Connection reset by peer]
iive has joined #ffmpeg-devel
AbleBacon has joined #ffmpeg-devel
Livio has quit [Ping timeout: 246 seconds]
System_Error has quit [Remote host closed the connection]
<kurosu>
BBB: re mv wraparound, and with that one trick, you can now call yourself versatile. The whole .5% bdrate gain of it.
System_Error has joined #ffmpeg-devel
<BBB>
I bet the 0.5% gain is patented
* BBB
runs
<nevcairiel>
isnt that a sure thing?
<nevcairiel>
i wouldnt have thought 360 video is an important enough use-case to invent a whole mode for, however
<nevcairiel>
even if its kinda interesting
ccawley2011 has quit [Ping timeout: 252 seconds]
System_Error has quit [Quit: Leaving]
System_Error has joined #ffmpeg-devel
<ravenJPL>
is xHE-AAC going to get a designed profile for aac?
<ravenJPL>
right now probing does not show much
<ravenJPL>
designated*
<Lynne>
it should, thanks for reminding me
<Lynne>
JEEB: do you know anything about AOT == 39?
<Lynne>
Alex Converse added it back 15 years ago as 45, but then that got changed through the years to 39
ccawley2011 has joined #ffmpeg-devel
<Lynne>
I can't find any info on it being documented
ccawley2011 has quit [Read error: Connection reset by peer]