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
sudden has quit [Ping timeout: 276 seconds]
rvalue has quit [Ping timeout: 256 seconds]
sudden has joined #ffmpeg-devel
rvalue has joined #ffmpeg-devel
navi has quit [Quit: WeeChat 4.1.2]
qeed has quit [Remote host closed the connection]
qeed has joined #ffmpeg-devel
lexano has quit [Ping timeout: 264 seconds]
<BBB> Lynne: I'm sorry for c99toc89, but it was a means to an end (getting msvc support when it didn't support c99). Nobody loves c89. I'm really sorry for that, is was quite horrible
<BBB> fortunately it's gone now
<Lynne> why are you apologizing? you didn't write it, did you?
<Daemon404> lol
<Daemon404> he did
<Lynne> I'm not angry, I'm impressed
thilo has quit [Ping timeout: 272 seconds]
<Lynne> libclang? I remember it being a massive perl script
thilo has joined #ffmpeg-devel
MetaNova has quit [Remote host closed the connection]
MetaNova has joined #ffmpeg-devel
derpydoo has quit [Ping timeout: 264 seconds]
<Traneptora> Copyright (c) 2012 Ronald S. Bultje <rsbultje@gmail.com>
<Traneptora> hm
mkver has quit [Ping timeout: 252 seconds]
dellas has quit [Remote host closed the connection]
<kasper93> will we need c17-to-c99 now?
<Traneptora> what do we use from C17 that isn't in C99 that can't be guarded with a preprocessor directive
<Lynne> the point isn't to add more ifdefs, it's to remove ifdefs
sudden has quit [Ping timeout: 260 seconds]
sudden has joined #ffmpeg-devel
cone-688 has joined #ffmpeg-devel
<cone-688> ffmpeg Aleksoid master:336d59643a96: avcodec/d3d12va_vc1: add support for D3D12_VIDEO_DECODE_PROFILE_VC1_D2010 guid.
<cone-688> ffmpeg Tong Wu master:82e8838165bb: avcodec/dxva2: fix different 'const' qualifiers warning
<cone-688> ffmpeg Tong Wu master:92ee7461c3f5: avcodec/d3d12va_decode: fix different 'const' qualifiers warning
<cone-688> ffmpeg Haihao Xiang master:cd31eac9997d: lavc/qsvenc: Add workaround for VP9 keyframe
elenril has quit [Ping timeout: 245 seconds]
elenril has joined #ffmpeg-devel
lemourin has quit [Read error: Connection reset by peer]
lemourin has joined #ffmpeg-devel
<jdarnley> Nice. ffmpeg doesn't see a track in an mkv. Is that a bug or a limitation or is this file hosed?
<jdarnley> that's not a question for 4AM
AbleBacon has quit [Read error: Connection reset by peer]
haihao has quit [Ping timeout: 272 seconds]
haihao has joined #ffmpeg-devel
<Lynne> shouldn't be, matroska's pretty robust in how streams are handled
<Lynne> does mkvmerge demux it correctly?
<jdarnley> mkvinfo reports all the "missing" stream
<jdarnley> I didn't try extracting it yet
<Marth64> What kind of track it? curiously
<jdarnley> vobsub/dvd_subtitle
<Marth64> LOL
<Marth64> knew it
* jdarnley sleeps
<Marth64> Did you mux this originally yourself? (i.e. wanted to ask about origin of how you muxed it, as I have a few theories)
<Marth64> but yeah maybe some other time
<Marth64> zzz important
jamrial has quit []
MisterMinister has joined #ffmpeg-devel
<Marth64> to Lynne's point, it's pretty robust. I've seen some software mux dvd subs strangely though
<Marth64> becaue dvd is evil
Martchus has joined #ffmpeg-devel
Martchus_ has quit [Ping timeout: 256 seconds]
haihao has quit [Ping timeout: 264 seconds]
haihao has joined #ffmpeg-devel
<Lynne> jkqxz: ping, could you take a look at v2 of my vulkan av1 patch?
haihao has quit [Ping timeout: 252 seconds]
haihao has joined #ffmpeg-devel
Marth64 has quit [Ping timeout: 252 seconds]
Marth64 has joined #ffmpeg-devel
haihao has quit [*.net *.split]
MisterMinister has quit [*.net *.split]
elenril has quit [*.net *.split]
vjaquez has quit [*.net *.split]
elastic_dog has quit [*.net *.split]
quietvoid has quit [*.net *.split]
gnafu has quit [*.net *.split]
paulk has quit [*.net *.split]
kekePower has quit [*.net *.split]
markh has quit [*.net *.split]
thilo has quit [*.net *.split]
uau has quit [*.net *.split]
justache has quit [*.net *.split]
BtbN has quit [*.net *.split]
staceee has quit [*.net *.split]
tmm1_ has quit [*.net *.split]
emersion has quit [*.net *.split]
Gramner has quit [*.net *.split]
aaabbb has quit [*.net *.split]
Fenrir has quit [*.net *.split]
Lynne has quit [*.net *.split]
jkqxz has quit [*.net *.split]
rix has quit [*.net *.split]
ubitux has quit [*.net *.split]
rvalue has quit [*.net *.split]
zsoltiv__ has quit [*.net *.split]
kasper93 has quit [*.net *.split]
haxar has quit [*.net *.split]
Flat has quit [*.net *.split]
c1480 has quit [*.net *.split]
deus0ww has quit [*.net *.split]
Teukka` has quit [*.net *.split]
odrling has quit [*.net *.split]
averne has quit [*.net *.split]
pross has quit [*.net *.split]
kurufu has quit [*.net *.split]
Lypheo has quit [*.net *.split]
Guest4622 has quit [*.net *.split]
kierank has quit [*.net *.split]
s55 has quit [*.net *.split]
Kei_N has quit [*.net *.split]
stevenliu has quit [*.net *.split]
c_14 has quit [*.net *.split]
jdek has quit [*.net *.split]
Traneptora has quit [*.net *.split]
JEEB has quit [*.net *.split]
haasn has quit [*.net *.split]
pengvado has quit [*.net *.split]
rom1v has quit [*.net *.split]
Chagall has quit [*.net *.split]
rcombs has quit [*.net *.split]
ChanServ has quit [*.net *.split]
marcj has quit [*.net *.split]
osvein has quit [*.net *.split]
ramiro has quit [*.net *.split]
CoreX has quit [*.net *.split]
epony has quit [*.net *.split]
deer2 has quit [*.net *.split]
skinkie has quit [*.net *.split]
llyyr has quit [*.net *.split]
signalhunter has quit [*.net *.split]
chainik1 has quit [*.net *.split]
klaxa has quit [*.net *.split]
galad has quit [*.net *.split]
any1 has quit [*.net *.split]
Arsen has quit [*.net *.split]
nitroxis has quit [*.net *.split]
uartie has quit [*.net *.split]
feiwan123 has quit [*.net *.split]
psilokos has quit [*.net *.split]
Marth64 has quit [*.net *.split]
sudden has quit [*.net *.split]
Sebastinas has quit [*.net *.split]
ocrete2 has quit [*.net *.split]
another| has quit [*.net *.split]
KyleSiefring has quit [*.net *.split]
nevcairiel has quit [*.net *.split]
xvaclav has quit [*.net *.split]
englishm has quit [*.net *.split]
pal has quit [*.net *.split]
daddesio has quit [*.net *.split]
kylophone has quit [*.net *.split]
j45 has quit [*.net *.split]
TheSashmo has quit [*.net *.split]
kepstin has quit [*.net *.split]
jkkm has quit [*.net *.split]
jessidhia has quit [*.net *.split]
jess has quit [*.net *.split]
drv has quit [*.net *.split]
auri has quit [*.net *.split]
MetaNova has quit [*.net *.split]
Coinflipper has quit [*.net *.split]
rodgort has quit [*.net *.split]
blb has quit [*.net *.split]
microchip_ has quit [*.net *.split]
beastd has quit [*.net *.split]
SuperFashi_ has quit [*.net *.split]
frankplow has quit [*.net *.split]
CAT_S has quit [*.net *.split]
Venemo has quit [*.net *.split]
ringo has quit [*.net *.split]
hbbs has quit [*.net *.split]
Son_Goku has quit [*.net *.split]
ePirat has quit [*.net *.split]
sfan5 has quit [*.net *.split]
bossjones has quit [*.net *.split]
qeed has quit [*.net *.split]
HarshK23 has quit [*.net *.split]
AMM has quit [*.net *.split]
bcheng has quit [*.net *.split]
orthoplex64 has quit [*.net *.split]
jnbek has quit [*.net *.split]
courmisch has quit [*.net *.split]
thardin has quit [*.net *.split]
cworley has quit [*.net *.split]
wellsakus has quit [*.net *.split]
BBB has quit [*.net *.split]
bencoh has quit [*.net *.split]
jkhsjdhjs has quit [*.net *.split]
ocrete has quit [*.net *.split]
stazthebox has quit [*.net *.split]
mindfreeze has quit [*.net *.split]
michaelni has quit [*.net *.split]
andrewrk has quit [*.net *.split]
tmatth has quit [*.net *.split]
cosminaught has quit [*.net *.split]
av500 has quit [*.net *.split]
natto has quit [*.net *.split]
bpmedley has quit [*.net *.split]
adema has quit [*.net *.split]
cone-688 has quit [*.net *.split]
sepro has quit [*.net *.split]
q66 has quit [*.net *.split]
LaserEyess has quit [*.net *.split]
nto has quit [*.net *.split]
enyc has quit [*.net *.split]
Nightrose has quit [*.net *.split]
wyatt8740 has quit [*.net *.split]
philipl has quit [*.net *.split]
termos has quit [*.net *.split]
Hobbyboy has quit [*.net *.split]
j-b has quit [*.net *.split]
funman has quit [*.net *.split]
Daemon404 has quit [*.net *.split]
bilboed has quit [*.net *.split]
wbs has quit [*.net *.split]
tortoise has quit [*.net *.split]
rajivharlalka has quit [*.net *.split]
kurosu has quit [*.net *.split]
darkapex has quit [*.net *.split]
wcpan has quit [*.net *.split]
TD-Linux has quit [*.net *.split]
compn has quit [*.net *.split]
graphitemaster has quit [*.net *.split]
Manouchehri has quit [*.net *.split]
rossy has quit [*.net *.split]
dlb76 has quit [*.net *.split]
jluthra has quit [*.net *.split]
Dmitri_Ovch has quit [*.net *.split]
jdarnley has quit [*.net *.split]
RT|AO has quit [*.net *.split]
sudden has joined #ffmpeg-devel
haihao has joined #ffmpeg-devel
cone-688 has joined #ffmpeg-devel
MisterMinister has joined #ffmpeg-devel
elenril has joined #ffmpeg-devel
Marth64 has joined #ffmpeg-devel
Coinflipper has joined #ffmpeg-devel
MetaNova has joined #ffmpeg-devel
thilo has joined #ffmpeg-devel
vjaquez has joined #ffmpeg-devel
rvalue has joined #ffmpeg-devel
qeed has joined #ffmpeg-devel
wyatt8740 has joined #ffmpeg-devel
elastic_dog has joined #ffmpeg-devel
epony has joined #ffmpeg-devel
quietvoid has joined #ffmpeg-devel
HarshK23 has joined #ffmpeg-devel
j45 has joined #ffmpeg-devel
gnafu has joined #ffmpeg-devel
zsoltiv__ has joined #ffmpeg-devel
rajivharlalka has joined #ffmpeg-devel
paulk has joined #ffmpeg-devel
uau has joined #ffmpeg-devel
s55 has joined #ffmpeg-devel
TheSashmo has joined #ffmpeg-devel
markh has joined #ffmpeg-devel
kekePower has joined #ffmpeg-devel
kasper93 has joined #ffmpeg-devel
kurosu has joined #ffmpeg-devel
justache has joined #ffmpeg-devel
sepro has joined #ffmpeg-devel
rodgort has joined #ffmpeg-devel
darkapex has joined #ffmpeg-devel
marcj has joined #ffmpeg-devel
haxar has joined #ffmpeg-devel
deer2 has joined #ffmpeg-devel
bencoh has joined #ffmpeg-devel
skinkie has joined #ffmpeg-devel
AMM has joined #ffmpeg-devel
llyyr has joined #ffmpeg-devel
jkhsjdhjs has joined #ffmpeg-devel
Sebastinas has joined #ffmpeg-devel
bcheng has joined #ffmpeg-devel
ocrete has joined #ffmpeg-devel
chainik1 has joined #ffmpeg-devel
q66 has joined #ffmpeg-devel
orthoplex64 has joined #ffmpeg-devel
blb has joined #ffmpeg-devel
Flat has joined #ffmpeg-devel
Kei_N has joined #ffmpeg-devel
signalhunter has joined #ffmpeg-devel
philipl has joined #ffmpeg-devel
microchip_ has joined #ffmpeg-devel
kepstin has joined #ffmpeg-devel
jnbek has joined #ffmpeg-devel
beastd has joined #ffmpeg-devel
wcpan has joined #ffmpeg-devel
Nightrose has joined #ffmpeg-devel
Teukka` has joined #ffmpeg-devel
deus0ww has joined #ffmpeg-devel
TD-Linux has joined #ffmpeg-devel
courmisch has joined #ffmpeg-devel
natto has joined #ffmpeg-devel
BtbN has joined #ffmpeg-devel
odrling has joined #ffmpeg-devel
osvein has joined #ffmpeg-devel
stevenliu has joined #ffmpeg-devel
ocrete2 has joined #ffmpeg-devel
averne has joined #ffmpeg-devel
compn has joined #ffmpeg-devel
stazthebox has joined #ffmpeg-devel
c_14 has joined #ffmpeg-devel
jdek has joined #ffmpeg-devel
graphitemaster has joined #ffmpeg-devel
thardin has joined #ffmpeg-devel
Manouchehri has joined #ffmpeg-devel
termos has joined #ffmpeg-devel
staceee has joined #ffmpeg-devel
Hobbyboy has joined #ffmpeg-devel
Traneptora has joined #ffmpeg-devel
tmm1_ has joined #ffmpeg-devel
Daemon404 has joined #ffmpeg-devel
jkkm has joined #ffmpeg-devel
mindfreeze has joined #ffmpeg-devel
jessidhia has joined #ffmpeg-devel
jess has joined #ffmpeg-devel
another| has joined #ffmpeg-devel
emersion has joined #ffmpeg-devel
cworley has joined #ffmpeg-devel
pross has joined #ffmpeg-devel
rossy has joined #ffmpeg-devel
SuperFashi_ has joined #ffmpeg-devel
frankplow has joined #ffmpeg-devel
michaelni has joined #ffmpeg-devel
klaxa has joined #ffmpeg-devel
Gramner has joined #ffmpeg-devel
aaabbb has joined #ffmpeg-devel
kurufu has joined #ffmpeg-devel
KyleSiefring has joined #ffmpeg-devel
daddesio has joined #ffmpeg-devel
j-b has joined #ffmpeg-devel
galad has joined #ffmpeg-devel
wellsakus has joined #ffmpeg-devel
kierank has joined #ffmpeg-devel
Lypheo has joined #ffmpeg-devel
any1 has joined #ffmpeg-devel
ramiro has joined #ffmpeg-devel
auri has joined #ffmpeg-devel
drv has joined #ffmpeg-devel
bossjones has joined #ffmpeg-devel
hbbs has joined #ffmpeg-devel
ePirat has joined #ffmpeg-devel
sfan5 has joined #ffmpeg-devel
Son_Goku has joined #ffmpeg-devel
Venemo has joined #ffmpeg-devel
ringo has joined #ffmpeg-devel
CAT_S has joined #ffmpeg-devel
funman has joined #ffmpeg-devel
Arsen has joined #ffmpeg-devel
nevcairiel has joined #ffmpeg-devel
CoreX has joined #ffmpeg-devel
Guest4622 has joined #ffmpeg-devel
nitroxis has joined #ffmpeg-devel
JEEB has joined #ffmpeg-devel
andrewrk has joined #ffmpeg-devel
Fenrir has joined #ffmpeg-devel
LaserEyess has joined #ffmpeg-devel
cosminaught has joined #ffmpeg-devel
tmatth has joined #ffmpeg-devel
bilboed has joined #ffmpeg-devel
uartie has joined #ffmpeg-devel
Lynne has joined #ffmpeg-devel
feiwan123 has joined #ffmpeg-devel
jkqxz has joined #ffmpeg-devel
pengvado has joined #ffmpeg-devel
psilokos has joined #ffmpeg-devel
haasn has joined #ffmpeg-devel
av500 has joined #ffmpeg-devel
BBB has joined #ffmpeg-devel
xvaclav has joined #ffmpeg-devel
rom1v has joined #ffmpeg-devel
enyc has joined #ffmpeg-devel
ubitux has joined #ffmpeg-devel
rix has joined #ffmpeg-devel
rcombs has joined #ffmpeg-devel
Chagall has joined #ffmpeg-devel
ChanServ has joined #ffmpeg-devel
RT|AO has joined #ffmpeg-devel
tortoise has joined #ffmpeg-devel
wbs has joined #ffmpeg-devel
jluthra has joined #ffmpeg-devel
bpmedley has joined #ffmpeg-devel
nto has joined #ffmpeg-devel
Dmitri_Ovch has joined #ffmpeg-devel
jdarnley has joined #ffmpeg-devel
dlb76 has joined #ffmpeg-devel
adema has joined #ffmpeg-devel
kylophone has joined #ffmpeg-devel
pal has joined #ffmpeg-devel
c1480 has joined #ffmpeg-devel
englishm has joined #ffmpeg-devel
rajivharlalka has quit [*.net *.split]
kurosu has quit [*.net *.split]
darkapex has quit [*.net *.split]
wcpan has quit [*.net *.split]
TD-Linux has quit [*.net *.split]
graphitemaster has quit [*.net *.split]
compn has quit [*.net *.split]
Manouchehri has quit [*.net *.split]
rossy has quit [*.net *.split]
dlb76 has quit [*.net *.split]
jluthra has quit [*.net *.split]
Dmitri_Ovch has quit [*.net *.split]
jdarnley has quit [*.net *.split]
RT|AO has quit [*.net *.split]
BradleyS_ has joined #ffmpeg-devel
rajivharlalka has joined #ffmpeg-devel
kurosu has joined #ffmpeg-devel
wcpan has joined #ffmpeg-devel
darkapex has joined #ffmpeg-devel
TD-Linux has joined #ffmpeg-devel
compn has joined #ffmpeg-devel
jluthra has joined #ffmpeg-devel
graphitemaster has joined #ffmpeg-devel
rossy has joined #ffmpeg-devel
Manouchehri has joined #ffmpeg-devel
dlb76 has joined #ffmpeg-devel
RT|AO has joined #ffmpeg-devel
Dmitri_Ovch has joined #ffmpeg-devel
jdarnley has joined #ffmpeg-devel
\\Mr_C\\ has quit [Max SendQ exceeded]
sm2n has quit [Max SendQ exceeded]
BradleyS has quit [Quit: quit]
BradleyS_ is now known as BradleyS
staceee has quit [Ping timeout: 256 seconds]
wyatt8740 has quit [*.net *.split]
philipl has quit [*.net *.split]
termos has quit [*.net *.split]
Hobbyboy has quit [*.net *.split]
j-b has quit [*.net *.split]
funman has quit [*.net *.split]
Daemon404 has quit [*.net *.split]
bilboed has quit [*.net *.split]
wbs has quit [*.net *.split]
tortoise has quit [*.net *.split]
cone-688 has quit [*.net *.split]
sepro has quit [*.net *.split]
q66 has quit [*.net *.split]
LaserEyess has quit [*.net *.split]
nto has quit [*.net *.split]
enyc has quit [*.net *.split]
Nightrose has quit [*.net *.split]
cone-688 has joined #ffmpeg-devel
sepro has joined #ffmpeg-devel
q66 has joined #ffmpeg-devel
enyc has joined #ffmpeg-devel
LaserEyess has joined #ffmpeg-devel
nto has joined #ffmpeg-devel
Nightrose has joined #ffmpeg-devel
wyatt8740 has joined #ffmpeg-devel
philipl has joined #ffmpeg-devel
termos has joined #ffmpeg-devel
Hobbyboy has joined #ffmpeg-devel
Daemon404 has joined #ffmpeg-devel
j-b has joined #ffmpeg-devel
funman has joined #ffmpeg-devel
bilboed has joined #ffmpeg-devel
wbs has joined #ffmpeg-devel
tortoise has joined #ffmpeg-devel
Manouchehri has quit [Ping timeout: 246 seconds]
staceee has joined #ffmpeg-devel
sm2n has joined #ffmpeg-devel
enyc has quit [Max SendQ exceeded]
\\Mr_C\\ has joined #ffmpeg-devel
Manouchehri has joined #ffmpeg-devel
bencoh has quit [*.net *.split]
jkhsjdhjs has quit [*.net *.split]
ocrete has quit [*.net *.split]
stazthebox has quit [*.net *.split]
mindfreeze has quit [*.net *.split]
michaelni has quit [*.net *.split]
andrewrk has quit [*.net *.split]
cosminaught has quit [*.net *.split]
tmatth has quit [*.net *.split]
av500 has quit [*.net *.split]
natto has quit [*.net *.split]
bpmedley has quit [*.net *.split]
adema has quit [*.net *.split]
Marth64 has quit [*.net *.split]
sudden has quit [*.net *.split]
Sebastinas has quit [*.net *.split]
ocrete2 has quit [*.net *.split]
another| has quit [*.net *.split]
KyleSiefring has quit [*.net *.split]
nevcairiel has quit [*.net *.split]
xvaclav has quit [*.net *.split]
englishm has quit [*.net *.split]
pal has quit [*.net *.split]
kylophone has quit [*.net *.split]
epony has quit [*.net *.split]
deer2 has quit [*.net *.split]
skinkie has quit [*.net *.split]
llyyr has quit [*.net *.split]
signalhunter has quit [*.net *.split]
chainik1 has quit [*.net *.split]
galad has quit [*.net *.split]
klaxa has quit [*.net *.split]
any1 has quit [*.net *.split]
Arsen has quit [*.net *.split]
uartie has quit [*.net *.split]
nitroxis has quit [*.net *.split]
feiwan123 has quit [*.net *.split]
psilokos has quit [*.net *.split]
BradleyS has quit [*.net *.split]
marcj has quit [*.net *.split]
osvein has quit [*.net *.split]
CoreX has quit [*.net *.split]
ramiro has quit [*.net *.split]
Marth64 has joined #ffmpeg-devel
sudden has joined #ffmpeg-devel
Sebastinas has joined #ffmpeg-devel
ocrete2 has joined #ffmpeg-devel
another| has joined #ffmpeg-devel
nevcairiel has joined #ffmpeg-devel
xvaclav has joined #ffmpeg-devel
KyleSiefring has joined #ffmpeg-devel
englishm has joined #ffmpeg-devel
pal has joined #ffmpeg-devel
kylophone has joined #ffmpeg-devel
enyc has joined #ffmpeg-devel
BradleyS has joined #ffmpeg-devel
osvein has joined #ffmpeg-devel
ramiro has joined #ffmpeg-devel
marcj has joined #ffmpeg-devel
CoreX has joined #ffmpeg-devel
bencoh has joined #ffmpeg-devel
jkhsjdhjs has joined #ffmpeg-devel
ocrete has joined #ffmpeg-devel
mindfreeze has joined #ffmpeg-devel
stazthebox has joined #ffmpeg-devel
natto has joined #ffmpeg-devel
michaelni has joined #ffmpeg-devel
andrewrk has joined #ffmpeg-devel
cosminaught has joined #ffmpeg-devel
tmatth has joined #ffmpeg-devel
av500 has joined #ffmpeg-devel
bpmedley has joined #ffmpeg-devel
adema has joined #ffmpeg-devel
BradleyS_ has joined #ffmpeg-devel
BradleyS has quit [Quit: quit]
BradleyS_ is now known as BradleyS
nitroxis has joined #ffmpeg-devel
epony has joined #ffmpeg-devel
llyyr has joined #ffmpeg-devel
skinkie has joined #ffmpeg-devel
deer2 has joined #ffmpeg-devel
signalhunter has joined #ffmpeg-devel
chainik1 has joined #ffmpeg-devel
galad has joined #ffmpeg-devel
any1 has joined #ffmpeg-devel
klaxa has joined #ffmpeg-devel
Arsen has joined #ffmpeg-devel
feiwan123 has joined #ffmpeg-devel
uartie has joined #ffmpeg-devel
psilokos has joined #ffmpeg-devel
epony has quit [Max SendQ exceeded]
epony has joined #ffmpeg-devel
qeed has quit [*.net *.split]
HarshK23 has quit [*.net *.split]
AMM has quit [*.net *.split]
bcheng has quit [*.net *.split]
orthoplex64 has quit [*.net *.split]
jnbek has quit [*.net *.split]
courmisch has quit [*.net *.split]
thardin has quit [*.net *.split]
cworley has quit [*.net *.split]
wellsakus has quit [*.net *.split]
BBB has quit [*.net *.split]
qeed has joined #ffmpeg-devel
HarshK23 has joined #ffmpeg-devel
AMM has joined #ffmpeg-devel
orthoplex64 has joined #ffmpeg-devel
bcheng has joined #ffmpeg-devel
jnbek has joined #ffmpeg-devel
courmisch has joined #ffmpeg-devel
cworley has joined #ffmpeg-devel
wellsakus has joined #ffmpeg-devel
thardin has joined #ffmpeg-devel
BBB has joined #ffmpeg-devel
Marth64 has quit [Remote host closed the connection]
cone-688 has quit [Quit: transmission timeout]
qeed has quit [*.net *.split]
HarshK23 has quit [*.net *.split]
AMM has quit [*.net *.split]
bcheng has quit [*.net *.split]
orthoplex64 has quit [*.net *.split]
jnbek has quit [*.net *.split]
courmisch has quit [*.net *.split]
thardin has quit [*.net *.split]
cworley has quit [*.net *.split]
wellsakus has quit [*.net *.split]
BBB has quit [*.net *.split]
daddesio has quit [*.net *.split]
BradleyS has quit [*.net *.split]
enyc has quit [*.net *.split]
j45 has quit [*.net *.split]
TheSashmo has quit [*.net *.split]
kepstin has quit [*.net *.split]
jkkm has quit [*.net *.split]
jessidhia has quit [*.net *.split]
jess has quit [*.net *.split]
drv has quit [*.net *.split]
auri has quit [*.net *.split]
MetaNova has quit [*.net *.split]
Coinflipper has quit [*.net *.split]
rodgort has quit [*.net *.split]
blb has quit [*.net *.split]
microchip_ has quit [*.net *.split]
beastd has quit [*.net *.split]
SuperFashi_ has quit [*.net *.split]
frankplow has quit [*.net *.split]
CAT_S has quit [*.net *.split]
Venemo has quit [*.net *.split]
ringo has quit [*.net *.split]
hbbs has quit [*.net *.split]
Son_Goku has quit [*.net *.split]
ePirat has quit [*.net *.split]
sfan5 has quit [*.net *.split]
bossjones has quit [*.net *.split]
BBB has joined #ffmpeg-devel
hbbs has joined #ffmpeg-devel
j45 has joined #ffmpeg-devel
wellsakus has joined #ffmpeg-devel
thardin has joined #ffmpeg-devel
courmisch has joined #ffmpeg-devel
jnbek has joined #ffmpeg-devel
cworley has joined #ffmpeg-devel
orthoplex64 has joined #ffmpeg-devel
HarshK23 has joined #ffmpeg-devel
bcheng has joined #ffmpeg-devel
AMM has joined #ffmpeg-devel
qeed has joined #ffmpeg-devel
BradleyS has joined #ffmpeg-devel
enyc has joined #ffmpeg-devel
MetaNova has joined #ffmpeg-devel
Coinflipper has joined #ffmpeg-devel
TheSashmo has joined #ffmpeg-devel
rodgort has joined #ffmpeg-devel
blb has joined #ffmpeg-devel
microchip_ has joined #ffmpeg-devel
beastd has joined #ffmpeg-devel
kepstin has joined #ffmpeg-devel
jkkm has joined #ffmpeg-devel
jessidhia has joined #ffmpeg-devel
jess has joined #ffmpeg-devel
SuperFashi_ has joined #ffmpeg-devel
drv has joined #ffmpeg-devel
daddesio has joined #ffmpeg-devel
auri has joined #ffmpeg-devel
bossjones has joined #ffmpeg-devel
frankplow has joined #ffmpeg-devel
sfan5 has joined #ffmpeg-devel
Son_Goku has joined #ffmpeg-devel
ePirat has joined #ffmpeg-devel
ringo has joined #ffmpeg-devel
Venemo has joined #ffmpeg-devel
CAT_S has joined #ffmpeg-devel
BradleyS_ has joined #ffmpeg-devel
epony has quit [Excess Flood]
enyc has quit [Max SendQ exceeded]
BradleyS has quit [Quit: quit]
BradleyS_ is now known as BradleyS
epony has joined #ffmpeg-devel
enyc has joined #ffmpeg-devel
daddesio has quit [*.net *.split]
enyc has quit [*.net *.split]
j45 has quit [*.net *.split]
TheSashmo has quit [*.net *.split]
kepstin has quit [*.net *.split]
jkkm has quit [*.net *.split]
jessidhia has quit [*.net *.split]
jess has quit [*.net *.split]
drv has quit [*.net *.split]
auri has quit [*.net *.split]
daddesio has joined #ffmpeg-devel
enyc has joined #ffmpeg-devel
TheSashmo has joined #ffmpeg-devel
j45 has joined #ffmpeg-devel
jessidhia has joined #ffmpeg-devel
jkkm has joined #ffmpeg-devel
jess has joined #ffmpeg-devel
drv has joined #ffmpeg-devel
auri has joined #ffmpeg-devel
kepstin has joined #ffmpeg-devel
enyc has quit [Max SendQ exceeded]
enyc has joined #ffmpeg-devel
\\Mr_C\\ has quit [Remote host closed the connection]
Teukka` has quit [Read error: Connection reset by peer]
Teukka has joined #ffmpeg-devel
Teukka has quit [Changing host]
Teukka has joined #ffmpeg-devel
ravenJPL has joined #ffmpeg-devel
qeed has quit [Quit: Leaving]
ravenJPL has left #ffmpeg-devel [Leaving]
Krowl has joined #ffmpeg-devel
staceee has quit [Remote host closed the connection]
sm2n has quit [Remote host closed the connection]
staceee has joined #ffmpeg-devel
sm2n has joined #ffmpeg-devel
sm2n has quit [Remote host closed the connection]
staceee has quit [Remote host closed the connection]
ngaullier has joined #ffmpeg-devel
ngaullie has joined #ffmpeg-devel
ngaullier has quit [Ping timeout: 260 seconds]
staceee has joined #ffmpeg-devel
sm2n has joined #ffmpeg-devel
Krowl has quit [Read error: Connection reset by peer]
RSATom has joined #ffmpeg-devel
<RSATom> Hello. Can anybody please give me some hint who is maintaing libavformat/cafenc.c ? I've reported issue more than a half year ago - it didn't get any attention. I even provided patch (to both ticket and ffmpeg-devel mailing list) - also no attantion... Is it maintained at all?
<RSATom> I don't try to push to anyone, just want to know if there any chance it will be reviewed?
<elenril> RSATom: there is, but sadly getting reviews is a hard problem in general
<elenril> feel free to ping the patch on the ML, as long as you don't do it too often
<RSATom> elenril: got it, thanks!
<RSATom> But I'll wait a little bit more to not bother comunity...
<RSATom> I'm just curious why it didn't get any attention if that patch fixes ability to use the same audio file recorded from WebRTC without transcoding on both Android and iOs...
<RSATom> almost the same to be precise...
epony has quit [Remote host closed the connection]
epony has joined #ffmpeg-devel
Krowl has joined #ffmpeg-devel
Krowl has quit [Client Quit]
Krowl has joined #ffmpeg-devel
dellas has joined #ffmpeg-devel
dellas has quit [Remote host closed the connection]
ccawley2011 has joined #ffmpeg-devel
<j-b> RSATom: long time no see..
<RSATom> j-b: hi, yes, glad to see you :)
odrling has quit [Remote host closed the connection]
odrling has joined #ffmpeg-devel
mkver has joined #ffmpeg-devel
dellas has joined #ffmpeg-devel
dellas has quit [Remote host closed the connection]
AntimaD has joined #ffmpeg-devel
AntimaD has quit [Ping timeout: 268 seconds]
jamrial has joined #ffmpeg-devel
<jamrial> elenril: if a demuxer calls avpriv_set_pts_info() at init but never sets any field in output packets, then no flag should be set, right?
dellas has joined #ffmpeg-devel
lexano has joined #ffmpeg-devel
omegatron has joined #ffmpeg-devel
<elenril> jamrial: correct
RSATom has quit [Remote host closed the connection]
RSATom has joined #ffmpeg-devel
<mkver> elenril: How much do we trust the container when setting these fields? E.g. Matroska has a header flag for whether a track uses lacing. If lacing is used and no default duration exists, then there might be packets with no pts. If the flag indicates that lacing is not used, do we trust this and set the flags accordingly?
zsoltiv__ has quit [Ping timeout: 276 seconds]
zsoltiv_ has joined #ffmpeg-devel
navi has joined #ffmpeg-devel
Krowl has quit [Read error: Connection reset by peer]
cone-749 has joined #ffmpeg-devel
<cone-749> ffmpeg Andreas Rheinhardt master:9ae40f282d5e: avcodec/nvdec: Constify bitstream pointee
<cone-749> ffmpeg Andreas Rheinhardt master:f6ec01147f7f: avfilter/fifo: Remove (a)fifo filters
<elenril> mkver: I'd say yes, because containers are in general unreliable
<elenril> even if the container does store timestamps, they can still be garbage
<elenril> so the best we can do is set the flag and let the caller decide what to trust
Krowl has joined #ffmpeg-devel
epony has quit [Ping timeout: 264 seconds]
epony has joined #ffmpeg-devel
<jamrial> elenril: vplayerdec sets packet duration to -1
<jamrial> how do i interpret that?
<jamrial> set duration flag, set nothing, or bug?
<elenril> stab ubitux and pipe screams to chatgpt
<elenril> this must not stand
<JEEB> lol, that reminds me of seeing the definition of packet duration
<JEEB> and hurring and durring that unknown duration and zero duration were not differentiated
<ubitux> haha negative duration are fun with subtitles; they can mean different things
<elenril> jamrial: will it actually set it to -1 on output packets?
<elenril> or is that just sent to the subtitle queue which computes actual durations
<jamrial> elenril: doesn't seem to change it in ff_subtitles_queue_read_packet()
<jamrial> oh, my bad, it does
<jamrial> funky
enyc has quit [Ping timeout: 256 seconds]
<jamrial> it also sets dts from pts
<elenril> I'd say it shouldn't, but that's unrelated
<Lynne> ubitux: like what?
<Daemon404> im also confused about how negative can be valid for subs
<elenril> <unsee previous text>
<Daemon404> it is posissible vsfilter broke the 2nd law of thermodynamics
<JEEB> Daemon404: most often in our framework negative duration is unknown duration
<ubitux> Lynne: sorry i got confused with zero vs negative duration, respectively for captions where a zero duration clears the subs, and "up until the next subtitles (if any)" (for dur<0)
<Daemon404> ah... cursed
<JEEB> aand now you get to the fun bit that avpkt IIRC defines duration as >=0
<JEEB> so negatives were not allowed?
<ubitux> there is a ton of funny shit with subtitles, like clearing only part of the subs because of time overlaps and shit
<ubitux> i think dvb has some stuff like that iirc
<Daemon404> if i really wanted to troll i would suggest we should implement the entre 608 and 708 rendering VM/iinstruction set
<Daemon404> as durations
<Lynne> ubitux: so negative duration == until next subtitles?
<ubitux> Lynne: yeah
<JEEB> AVSubtitle utilizes UINT32_MAX or so for it IIRC
<ubitux> which is -1
<ubitux> :D
* elenril stabs ubitux
<JEEB> right
<JEEB> packet.h
<JEEB> > Duration of this packet in AVStream->time_base units, 0 if unknown.
<JEEB> > 0 if unknown
<ubitux> elenril: i didn't invent subtitles, i'm merely a victim here
<JEEB> this is why I went uguuu earlier when I tried to see if I could pass through unknown subtitle durations
<ubitux> JEEB: yeah that's one funny thing when you try to harmonize the 2 concepts
<ubitux> so like one api has dur=0 for unknown/up-to-the-next
<ubitux> and the other has -1, and dur=0 means "clear"
<ubitux> all kind of nasty things like that
<elenril> durr
<JEEB> ubitux: but it also technically means that negative durations are if not not-permitted but at least not documented
<JEEB> so you get an AVSubtitle with unknown duration and then think how you should set the output AVPacket? according to doc - 0
<JEEB> which is then the same as zero duration AVPacket
<JEEB> of course that definition could be updated to include negative durations for unknown, but until now zero was unknown?! :D
* JEEB does a virtual table-flip
<ubitux> it's hard to solve all these problems individually, and it may involve a few mental breakdowns
<ubitux> you've been warned
<JEEB> personally I'd just change the doc to mention negatives as unknown
<JEEB> just like other bits, I think AVFrames had something like that too?
<elenril> could also make it a packet flag
<kepstin> hmm, that's exciting; ffmpeg's vulkan_av1.c in 6.1.1 doesn't compile if you have vulkan-headers 1.3.227 installed (which includes the newly standardized av1 stuff)
<JEEB> kepstin: yea they changed the definitions and I think we don't utilize bundled headers if the define is in the vulkan headers already?
<kepstin> compiler output is mostly complaining that a bunch of structs which had MESA in the name got renamed to have KHR, but i don't know what other changes they made.
<JEEB> there's a patch on the ML for fixing it to match current vulkan header definitions. that in theory I guess could be backported if someone checks that it not only builds but also works on hardware that has AV1 hwdec
<Lynne> nvidia still have no drivers with av1 support, only radv
<kepstin> hmm. only hardware i have with av1 decoding is intel, and the state of vulkan av1 on that seems... very preliminary. https://airlied.blogspot.com/2024/02/anv-vulkan-av1-decode-status.html
<JEEB> you at least have something :P
<JEEB> that's more than me :D
<Lynne> av1 on intel left dave with ptsd
<Lynne> to put this in perspective, he spent a month getting intel to output a single intra-only frame, but on amd hardware, it took him 3 days to have intra and most of inter working
<Lynne> both had headers published, it was a fair game
<JEEB> yea, hardware integration can be "fun"
<Lynne> and at the end of the day, intel's av1 implementation is incomplete, as it leaves film grain entirely up to shaders
* elenril wonders why does av_opt_get_video_rate() exist
<Daemon404> ask paul
<Daemon404> why do any of those accessors exist?
<Daemon404> (i wasnt aware of their existence until now)
<elenril> the other functions do make sense
<elenril> but this one is exactly the same as av_opt_get_q()
<Daemon404> *(int *)(((uint8_t *)target_obj) + o->offset) = fmt;
<Daemon404> things i saw today
<BtbN> replace it with a macro that calls it? :D
<elenril> I might, but wondering why was it added in the first place
<BtbN> Which of the two came first?
dellas has quit [Remote host closed the connection]
<elenril> rationals were much earlier
<elenril> Daemon404: the functions exist precisely to shield users from such code
<kepstin> oh, that's annoying, even just doing --disable-hwaccel=av1_vulkan doesn't fix the build, since compiling vulkan_decode.c also fails.
<Lynne> thankfully, distributions update their vulkan headers at a glacial pace
<Daemon404> elenril, fair enough
* kepstin is trying to work around a distro package build failure of ffmpeg following a vulkan headers update, so...
<JEEB> test if the patch on ML works
<elenril> is double guaranteed to be a superset of float?
<mkver> in terms of representable values, yes
nevcairiel has quit [Quit: http://quassel-irc.org - Chat comfortably. Anywhere.]
nevcairiel has joined #ffmpeg-devel
<kepstin> only thing i could do is build test the patches, probably what i'll do for the package for the time being is disable all the vulkan codec hwaccels and revisit when the drivers are available :/
<Lynne> not sure if I should backport the patch to 6.1 though
<Lynne> it's a fairly large patch, not a bugfix
Krowl has quit [Read error: Connection reset by peer]
<kepstin> yeah, the bugfix for 6.1 would be to remove the vulkan av1 support completely since nobody can use it anyways.
<JEEB> then there should probably be a small fixup which would disable the build against up-to-date headers then. so that it would still build against the old headers, but not break build-time with current ones.
<kepstin> (or at least fix it so you can build the x264/hevc hwaccels with the av1 disabled)
<JEEB> but tbh if it was experimental to begin with, not sure how bad backporting support for up-to-date headers would be?
<kepstin> h264*
<JEEB> s/experimental/unofficial/
<Lynne> kepstin: if the issue is just a filename, it could be renamed
<kepstin> ah, you think the issue might be that because of the filename, it's pulling in the system versions of the vulkan_video_codec_av1std* headers instead of the ones included with ffmpeg?
derpydoo has joined #ffmpeg-devel
<JEEB> and/or define I think?
AntimaD has joined #ffmpeg-devel
RSATom has quit [Remote host closed the connection]
RSATom has joined #ffmpeg-devel
RSATom has quit [Remote host closed the connection]
aljazmc has joined #ffmpeg-devel
Krowl has joined #ffmpeg-devel
Antima has joined #ffmpeg-devel
AntimaD has quit [Ping timeout: 268 seconds]
cone-749 has quit [Quit: transmission timeout]
<courmisch> help!!! I'm stuck in an alternate reality whence the CC is doing something!!!!
<JEEB> find truck-kun?
AbleBacon has joined #ffmpeg-devel
devinheitmueller has joined #ffmpeg-devel
<courmisch> That sounds like a terrible idea as I have not properly trained as Minoru Kageno
<JEEB> oh well. you'll need then to look out for other ways of traversing the multiverse :)
<courmisch> AFAICT, this alternate reality does not have anything extraordinary, except for the CC
AntimaD has joined #ffmpeg-devel
<courmisch> so I don't think kinetic energy-assisted suicide would work
devinheitmueller has left #ffmpeg-devel [#ffmpeg-devel]
<JEEB> :)
devinheitmueller has joined #ffmpeg-devel
Antima has quit [Ping timeout: 256 seconds]
<courmisch> severe post-fenestritis rapid deceleration sounds very unpleasant, no matter how common it seems to be on the other side of Finno-Mordorish border
<APic> ☺
AntimaD has quit [Ping timeout: 260 seconds]
<Daemon404> cheers to the community growing up a bit
<Lynne> there's nothing to cheer about, the whole thing just leaves me bitter
<Daemon404> personally, cannot agree
<Lynne> you'd rather see developers being chased off than simply not being in the situation needed to chase them off?
<Daemon404> there is no situation where paul or ng could stay acceptably with their behavior
<Daemon404> it is in no way worth it
<Daemon404> and they refuse to change it
<Daemon404> it has been literally over a decade.
<BtbN> Those two people singlehandedly chased way more other people off than it's worth, sadly.
Marth64 has joined #ffmpeg-devel
<Daemon404> corret
<Lynne> I think there always is a way, michael changed
<BtbN> Of course it sucks action like this needs to be taken, but nothing else worked
<Lynne> but that way is in the past, and we don't have a time machine
<Daemon404> note neitherr of these are premanent
<BtbN> There were literal YEARS of trying to solve things nicely
<BtbN> Didn't work
<Daemon404> you expect people to just put up with them hoping they change?
<kierank> 17:23:21 <•Lynne> I think there always is a way, michael changed
<kierank> april 1st has come early
<elenril> he did
<elenril> a lot
<Lynne> like I said, I understand why, I simply would have liked for this situation to never have happened, beyond all other considerations whether it was possible
<Daemon404> he did change
<kierank> sure, aggressive to passive aggressive
<Daemon404> but it was by force
<jamrial> kierank: stop
<Daemon404> jamrial, why
<Daemon404> he speaks nothing but the truth
<elenril> it borders on a personal attack
<Daemon404> he *is* passive aggessive
<Daemon404> insanely so
<Daemon404> it is an actual big problem.
<Daemon404> or we can pretend otherwise.
<elenril> I'd say it's still a massive improvement compared to 2010
<Daemon404> low, low bar
<Daemon404> too low.
<elenril> progress is always good
<kierank> attacks in threads marked "RFC"
<kierank> real "improvement"
<Daemon404> we shall wait to see what the pending CC complaints of michael yeild
<Daemon404> or thilo.
<kierank> 17:26:47 <•elenril> it borders on a personal attack
<kierank> being called "weak" is not a personal attack
<kierank> interesting
<elenril> regardles of whether it is, flaming back is not the answer
<elenril> it does not improve the situation
<kierank> as courmisch regularly points out in every thread where I have a differing opinion to michael I am relentless attacked on a personal level
<kierank> just because it's passive doesn't change anything
<kierank> in threads marked RFC
<Daemon404> 15:01 < Daemon404> > the problem with these threads is the two outcomes are:
<Daemon404> 15:01 < Daemon404> > * stop replying, they win
<Daemon404> 15:01 < Daemon404> > * keep replying and getting gaslit until you burn out or get in trouble, they win
<Daemon404> 15:01 < Daemon404> current status: option 2.
<Daemon404> will just leave this here and go make food
<Daemon404> the real vibe i get is that some are held to higher standards than others. the en.
<cosminaught> what do the trolls win if you ignore them?
<Daemon404> pushing crap into the codebase
<Daemon404> stf funding
<Daemon404> whichever.
<Marth64> hi good day / hope it gets better
<cosminaught> where securing stf funding is a bad thing? That seems odd
<kierank> securing stf funding being run by people who are somehow speedrunning business 101
<kierank> it's quite amazing
<kierank> oh and want government money yet in their email signatures are paranoid about the government
<kierank> the level of mental gymnastics sometimes is incredible
<Daemon404> i would take less issue if they applied for funding s themseves
<Daemon404> and nto ffmpeg itself
<Daemon404> and didnt keep it secret for months then play brinkmanship
<Daemon404> because DUE DATE
<Daemon404> its very sketch.
<cosminaught> i still don't get it, what is wrong with seeking funding for ffmpeg itself? Definitely seems like it could benefit from more funding rather than less.
<kierank> because it needs to be run by people who are responsible
<cosminaught> are you volunteering?
<kierank> sure
<Daemon404> certainly not thilo who avoids responding an hopes it goes away until you reply for him
<jamrial> please say so in the thread
<cosminaught> +1
<kierank> jamrial: why, so I can get more conspiratorial attacks
<kierank> like the above
<kierank> that I am "weak"
<kierank> "toxic[ity]"
<BtbN> Nothing is wrong with the idea in general. There's just worries that it'll be used to backdoor in some stuff people strongly disagree with.
<jamrial> no, so someone responsible can run it
<kierank> with my four days notice
<kierank> anyway at least if I ran it there wouldn't be an obvious fflabs proxy war going on
<kierank> with weird fflabs ongoings written in obvious riddles
<kierank> on the ml
<Daemon404> i cant imagine why people think it is conspiratorial when it was done in secret
<Daemon404> truly baffling
<cosminaught> it is baffling to me that people assume bad intent on all these things, trying to get more funding for ffmpeg developers seems like an inherently good thing, even if there are lessons to be learned about how to do it better next time
ngaullie has quit [Ping timeout: 260 seconds]
<Daemon404> it is almost as if the project has a 20 year history of infighting and people acting with bad intent
<Daemon404> including those specifically involved here
<kierank> what we should have had is a vote 3 months ago whether to go via fflabs or spi
<kierank> but we didn't
<kierank> or someone else
<elenril> and the conspicuous lack of replies
<Daemon404> several of my mails also lack answers.
<Daemon404> for months.
<Daemon404> including one threat where the person in question accintally sent an off-list mail to the list, admitting they were not replying on purpose
<Daemon404> why should these people have the benefit of the doubt
rvalue has quit [Ping timeout: 276 seconds]
qeed has joined #ffmpeg-devel
<pal> > i would take less issue if they applied for funding s themseves ---> that is the idea
<Daemon404> it isnt though
<Daemon404> it is being routed via SPI
rvalue has joined #ffmpeg-devel
<Daemon404> via the USA
<Daemon404> via the ffmpeg name
<pal> SPI would pay contributors individually and separately
<Daemon404> that isnt really the part people take issue with
<pal> I do not think "FFMPEG" will appear anywhere other than when referring to the codebase
<cosminaught> michael seeking funding for himself would be better than michael seeking funding for any ffmpeg developer that would like to work on maintenance? Why?
<kierank> because it's his problem if he screws up
<kierank> his reputation
<kierank> not the projects
<Daemon404> also maitenece is ill suited to discrete projects
<kierank> it's very clear the "people in question" (to use the super accurate voting terminology) are not capable of organising themselves or running this properly
<Daemon404> also a good chunk of our devs *canbot* go through SPI
<kierank> and just blame others instead
<Daemon404> due to US sanctions.
<BtbN> Isn't SPI a German thing?
<kierank> american
<Daemon404> it is american
<kierank> STF is german
<Daemon404> they want STF to pay SPI who pays devs
<Daemon404> SPI is a US charity
<Daemon404> and thus subject ot US laws
<BtbN> oh, that way around
<cosminaught> I assume it's going via SPI because there isn't any other non-profit representing ffmpeg. Might be a good time to establish a proper ffmpeg foundation?
<Daemon404> being a non-profit is not a requirement for stf to my knowlege
<kierank> "Might be a good time to establish a proper ffmpeg foundation?"
<kierank> definitely is april 1st
<kierank> there was one of those btw
<kierank> and it fought each other to death
Krowl has quit [Read error: Connection reset by peer]
<Daemon404> i also note the questions with lack of replies still have no replies, and also no explanation has been given as to why it was kept secret
<Daemon404> michael even noted other people wanted to keep it off list enrirely.
MrZeus has joined #ffmpeg-devel
<Daemon404> it is all well and good to say But Dony We Want Money, but dont ignore the rest
<cosminaught> I read that as others "wanted to first make sure this was real and actually possible" before bringing it up on the list, which seems sensible
<kierank> I have some beans to sell you if you believe that
<kierank> magic beans
<kierank> "transparency"
<kierank> 18:03:32 <Daemon404> i also note the questions with lack of replies still have no replies, and also no explanation has been given as to why it was kept secret
<kierank> 18:03:49 <Daemon404> michael even noted other people wanted to keep it off list enrirely.
<kierank> The ffmpeg way (tm)
<Daemon404> it is a community that is prone to disagreements an paranoia, nobody can be shocked that dumping this on the list 2 weeks before and lacking a lot of explanation would go well.
<Daemon404> and yes it is the ffmpeg way to ignore hard things and hope they go away.
dellas has joined #ffmpeg-devel
<another|> I'm beginning to think applying for funding or donations for ffmpeg is ill-advised.
<Daemon404> we dont even use the 100k we have aleady in SPI
<another|> There's too much fighting and drama to pour money into it at this stage.
<Daemon404> when i pointed that out, i was told it wasnt enough so didnt get used
<jamrial> Daemon404: travel cost reinbursment, mainly
<Daemon404> when i asked what it wasnt enough for
<cosminaught> there was an attempt to use the existing SPI money to fund development and that also seemed to devolve into using SPI money to fund development is bad
<Daemon404> i got told to suggest something
<cosminaught> michael for example proposed using some of the existing SPI money to fund swscale improvements
<Daemon404> i note that isnt being proposed for STF because STF still wouldnt be enough
<cosminaught> what isn't being proposed to STF? Last I saw swscale improvements is one of the proposed SOWs on the STF wiki
<Daemon404> if so i missed that, aplogies
<kierank> it's not about what is or what isn't being proposed
<kierank> it's about the way it's being run
<Daemon404> ^ that
<cosminaught> that part I don't disagree with, it can definitely be run better and it would benefit if someone would step up to help run it better
<haasn> re: STF, as I understood there will be a second opportunity in a couple of months, maybe that would be a better fit for the time being
<haasn> it seems most of the problems here are due to it being announced too late, and now it feels rushed/forced
<Daemon404> if there is, i do not object
<JEEB> pretty much, so if there is another deadline as well, that would probably alleviate certain concerns
<Daemon404> and agree with haasn on the 2nd bit.
<JEEB> yup
<haasn> I cannot find the quote atm
<haasn> in the gigantic backlog of emails related to this topic
<Daemon404> interesting
<jamrial> honestly yes, i guess we should wait for the next submission opening
<Daemon404> that is a very odd way to set up a funding program
<Daemon404> but... governments, i guess
<kierank> Daemon404: it was due to some german government budge issue
<kierank> someone explained at fosdem
<haasn> tl;dr they tried to re-use covid-19 emergency money for non-covid-19 emergency reasons
<haasn> and it turns out, according to the constitutional court, the government is not just allowed to print as money as it wants
kierank has left #ffmpeg-devel [#ffmpeg-devel]
epony has quit [Remote host closed the connection]
<pal> can someone summarize the objection(s) to individuals applying to receive a grant for doing work on ffmpeg
<Traneptora> as far as I understand the biggest gripe with the whole process is that apparently in private discussions have been going on
<BtbN> nobody objected to that aspect of it
<Traneptora> and then suddenly it's like "quick we have two weeks or the money is GONE"
<Traneptora> and everyone is like "hold up can we talk about this?" and the answer is NO WE HAVE TWO WEEKS
<haasn> pal: afaict the problem is that it's not individuals applying, because the minimum grant is 150k, we have to have some central person to apply for the entire project
<BBB> cosminaught: "what do the trolls win if you ignore them?" - from what I've seen, the problem is that certain parties won't engage with our project (submit patches, ..) because of "the trolls"
<haasn> and there are disagreements about who that central person/entity is and how the process is being handled
<Traneptora> and, there's no clearly established methods of whether developers are getting individually funded to work on an open-source project, what happens if their code is rejected, etc.
<Traneptora> and these sorts of questions need some discussion to establish. it's a decent amount of money so we should know what happens in these scenarios
<BBB> cosminaught: I agree that generally speaking, funding is always good
<pal> hassn: the best I can see "us" doing is packaging several SOWs, each from individual devs, and let SPI manage each of these SOWs individually
<Traneptora> but the general frustration is how rushed the whole process is, iirc. like, obviously, money is better than no money. but when you have money you have questions about conditions for payment and that sort of thing
<Traneptora> questions we don't yet have good answers to
<pal> Traneptora: here's how I understand the money flow: STF -> SPI -> individual devs with an SOW with SPI
<pal> ... "FFMPEG" is not involved since there is no "FFMPEG" org
<cosminaught> BBB: I agree that it's also good to get rid of the trolls, no doubt, but until such time it's always good policy to ignore their trolling attempts rather than feed them
<Traneptora> sure, but what strings are attached? in order to be paid for the work, is it required for that work to be merged?
<Traneptora> what happens if that work isn't merged on technical grounds?
<pal> >in order to be paid for the work, is it required for that work to be merged?< ---> I think we should leave that to each dev to propose in their SOW
<BBB> cosminaught: yes
<Traneptora> does the developer still get paid? if they do get paid anyway, doesn't it look bad for FFmpeg to reject a paid work?
<Traneptora> these are all questions without easy answers
<pal> Traneptora: the answer might be different for each dev/project
<Traneptora> you see, we don't know if that's the case
<Traneptora> should the individual developer be responsible for putting "and the code gets merged into master" in their SOW? why would they do that if they had that option?
<pal> FWIW if I had to write an SOW I would include one milestone for submitting the patch and another once the patch is merged
<pal> (pretty standard approach AFAIK)
<cosminaught> that's my proposal as well, it shouldn't be 100% on merge or 100% even without merge
<Traneptora> and what happens if that milestone isn't delivered?
<haasn> iirc somebody raised the issue that having money go from germany -> usa -> europe is stupid for tax reasons?
<Traneptora> how much less funding do they get?
<pal> Traneptora: which milestone?
<Traneptora> if they submit a patch that is never merged. how does that affect the developer's pay?
<pal> Traneptora: yes, it would affect pay, but the amount is for the dev to propose based on the type of project, etc.
<pal> My big question is whether SPI is willing/interested/capable of managing these SOWs... totally unclear to me
<Traneptora> as a developer, if putting "merged into master" risks me working for free or significantly less on a paid project proposal, when that merge is outside of my control, I'm heavily incentivized to not list that as a milestone.
<Traneptora> because if I can't deliver it (which is not something I can ultimately control), then I get paid a lot less for that work
<pal> Traneptora: yes, I would not assign 90% of the payment on the merge milestone!
<pal> (personal preference of course)
<Traneptora> would you assign a nonzero quantity? what %? these are questions we don't have answers to
<cosminaught> not merged into master means the value of the work hasn't really been realized though. It's different for a commercial thing where the requesting company can still privately use the patch even if it doesn't get merged
<Traneptora> I'm not raising these questions because i want you to answer them, I'm raising them because I think these answers should be discussed, and that we weren't given adequate time to work out those details
<pal> Traneptora: my point is that I do not think there is one size fits all and that a group can choose the same threshold for everyone for every project
<Traneptora> sure, but you need to discuss that before saying 'quick submit now or lose funding'
<pal> ... it should be left to individual projects/devs
<pal> To be sure, I am totally happy to do nothing and wait to help at the next opportunity
<Traneptora> I'm mostly frustrated by the timeline of things. why should there be such a time crunch if the work to get this funding has been going on for months
<pal> ... just trying to help get folks compensated for their work
<pal> Traneptora: I agree and dislike time crunches/death marches
<haasn> the question of whether to pay for submission or for merging only really matters if your idea of a project is to work for 6 months and then submit a 300-commit series
<haasn> rather than incrementally merging during development, which is what we _should_ be doing
<pal> haasn: +1
<pal> also... depends on whether they are bug fixes (low risk) vs demuxer for exotic file format for which not test case exists and that requires a modification to the core API (super high risk)
<pal> in any case, I am happy to help (with the understanding that the chances of success are low per all the comments made) or do nothing.
<pal> ... and thanks for all the insights above :)
<haasn> you can start a new codec by implementing only the basic barebones of possible features, formats, syntax elements etc. and make it more complete with subsequent commits
<haasn> that still allows incremental work to happen without an all-or-nothing merge
<michaelni> What does FFmpeg gain from "merge to git master" over "submitted for review to the mailing list. As well as taking care of all reasonable review comments." ? In what cases would that not be equal ?
<cosminaught> in the case of a bugfix merge means the bug is actually fixed, not merged means the bug hasn't actually been fixed
<michaelni> what is blocking that patch in this example ?
<michaelni> i mean it would be posted and all reasonable comments taken care of
<michaelni> so it could be comited unless theres something unreasonable
<cosminaught> if nothing is blocking the patch then it would be merged, a distinction only arises when something is blocking it from being merged
<cworley> anyone able to merge these series? they've been on the ML for a while without objection https://patchwork.ffmpeg.org/project/ffmpeg/list/?series=10643 https://patchwork.ffmpeg.org/project/ffmpeg/list/?series=10615
<cosminaught> the problem with "something unreasonable" is that this is very subjective, we've seen lots of debates about what is reasonable or unreasonable in terms of objections
<michaelni> yes it can be
<cosminaught> having some % of the payment tied to resolving all objections and actually merging the code seems sensible, if that means escalating to the TC because of unreasonable objections then that's part of the work
<cosminaught> and the TC can sort out whether it's unreasonable or not
<michaelni> if the contractor has to do unexpected extra work due to unreasonable objections and TC that extra time should be payed
<another|> michaelni: what do you think about multiple milestones?
<michaelni> another|, the biggest problem is that this is rushed and iam not exactly sure what the community exactly wants, i understand that some people feel really strongls on "merged to git master" but i think it isnt really helpfull to ffmpeg as it could lead to more hostilities in the review process
<BBB> cworley: is probably OK, but I have no idea what video codec that is. can you merge yourself or do you need someone to merge for you?
<another|> so what about a milestone for submitting and one for merging?
<cworley> BBB: i don't have write access
<BBB> what video codec is this?
<BBB> I'm not familiar with dxv
<BBB> (trying to see if someone else would be a better reivewer)
<michaelni> another|, if theres consensus that people want that i can resubmit my SoW with that. But if theres just 2-3 people who want that and the majority not caring i would prefer avoiding subjecting me to extra stress in the reviews
<cworley> proprietary codec used by VJ software called Resolume -- i believe Paul and Vittorio are familiar with it
<cworley> https://patchwork.ffmpeg.org/project/ffmpeg/list/?series=10721 as well -- v3 just rewrites the commit message to make FATE happy -- https://ffmpeg.org/pipermail/ffmpeg-devel/2024-February/320688.html has demo files for the alpha change
aljazmc has quit [Quit: Leaving]
<cworley> the codec is extremely similar to hap, if you know anyone familiar with that
aljazmc has joined #ffmpeg-devel
<Lynne> cworley: so those 3 patchsets?
<cworley> yes
<BBB> pal: btw, congrats
<Lynne> cworley: it changes FATE results right now, right?
<cworley> only the patchset that adds FATE tests. the other patchsets don't change FATE because the existing samples don't adequately cover alpha-containing texture formats
<BBB> re: submit vs. merge, I think I've said enough about that... but fwiw, merge has never caused issues for me
<BBB> Lynne: tnx... I was about to do the same thing
<cworley> thx
<Lynne> cworley: could you link to the dxt encoder files needed for fate tests?
microchip_ has quit [Quit: There is no spoon!]
cone-765 has joined #ffmpeg-devel
<cone-765> ffmpeg Connor Worley master:c4e9556cf546: lavc/texturedsp: fix premult2straight inversion
<cone-765> ffmpeg Connor Worley master:3b6a515c5f8f: lavc/dxv: treat DXT5-tagged files as DXT4
<cone-765> ffmpeg Connor Worley master:1eeee68d8eb6: lavc/dxv: fix incorrect back-reference index calculation in DXT5 decoding
microchip_ has joined #ffmpeg-devel
<Lynne> oh, okay, I'll push the fate tests for the encoder too
<cworley> wonderful, thank you
<cone-765> ffmpeg Connor Worley master:5fbb6fc636cc: fate/video: add DXV3 DXT1 encoding test
<cone-765> ffmpeg Connor Worley master:395469c1b091: fate/video: add DXV3 HQ tests
<Lynne> cworley: cheers, pushed
<cosminaught> > if the contractor has to do unexpected extra work due to unreasonable objections and TC that extra time should be payed
<cosminaught> michaelni: precisely why part of the payment milestone should be tied to merging, to compensate for the effort of getting things merged
<BBB> like a slight plus in price because a merge was part of the requirements? that seems ok
<cosminaught> otherwise either the patch doesn't get merged (and the work has no value to end users) or the extra time spent to get it merged is done for free
<BBB> I was mostly worried about the loss of value if it is not merged, yes
<BBB> but both are true
<cosminaught> yeah, between the two that's my concern as well, given this is already contentious the optics of paying for maintenance work only to have it sit unmerged are going to be terrible
<courmisch> If you want to be paid on submission rather than merge, don't go via the community, as BBB already pointed out on the ML.
<BBB> right... it's in the project/community's best interest to focus as much as possible on merge, since that's where the benefit is realized
<courmisch> As added bonus, no drama about deadlines and no need to argue to death about what the projects should be.
<courmisch> All this mess just because Michael absolutely wants to have the moral FFmpeg rubberstamp of approval
<courmisch> Also what kierank wrote; it does not look like it's been planned well and professionally, so it's better to keep FFmpeg's name out anyhow
<courmisch> In the end, it looks to me that this is really just a FFlabs board room proxy battle, and we could dispense with that.
<BBB> let's not jump to conclusions
<BBB> and let's not bring fflabs into this
<courmisch> I'm not jumping to conclusions and I did not bring FFlabs into this.
<BBB> I think fundamentally michael is allowed to ask for a vote where he asks for GA approval of his STF submission, your so-called rubber stamp. is it hasty? yes, probably. but I don't think it's malicious
<michaelni> BBB, ill resubmit the SoW with additional "merged to git master" milestones
<BBB> hm... how does that affect the GA vote?
<Daemon404> the ga vote did not have sow frozen
<Daemon404> it is a living wiki
<BBB> o_O
<Daemon404> because sow werent all done / ready anyway
<courmisch> BBB: I don't think Michael is being malicious here. I do reckon though that he all but admitted that it was a consequence of a fight between Thilo and one or several other FFlabs parties
<BBB> hm...
<Daemon404> i must admit it it is pretty crazy to see people referring to their own employer like that in public
<haasn> elenril: you know what, I'm completely silly
<haasn> elenril: we already have a distinction between hw->hw and sw->hw, it's called hwmap for the former
<BBB> Daemon404: it makes me a bit uncomfortable, yes
<haasn> the ability to pass hwframes to hwupload is due to passthrough functionality
<BBB> I, for one, think fflabs is great in that it pays some ffdevs to work on ffmpeg
<haasn> i.e. hwupload is a no-op if you feed it a frame in the same hwformat as you got out
<courmisch> Daemon404: oh yes. But it's not as bad as Thilo throwing accusations against other FFlabs parties during the GA-related votes
<BBB> but what do I know. I'm guessing fairytale castles look different from the inside
<courmisch> so at this point, :shruh:
<courmisch> shrug*
<BBB> hey courmisch wanna share any gossip from videolan asso's unicorn castle?
<BBB> B)
<haasn> elenril: so actually all of my agonizing about the design was for naught
<BBB> I think unicorn is appropriate given that we all went to tokyo
<courmisch> BBB: I'm probably the least aware, as I'm neither a board member, nor associated with VideoLabs.
<BBB> but you're a member right?
<courmisch> as are dozens of people
<JEEB> it was funny when I helped some person who happened to be related to a company, and got thanked as "Jan from FFlabs" xD
<another|> please don't keep unicorns in castles, mkay? That's non species-appropriate husbandry
<courmisch> and I'm still a bit salty about getting banned for a whole month 1.5 years ago
<cosminaught> as far as I can tell the vote is mostly about submit to stf now vs don't submit now (presumably submit later). So anyone who thinks submitting in February is rushed should probably vote no
<devinheitmueller> courmisch: I think perhaps the confusion is that BBB asked about the *videolan* association meeting, not *videolabs*.
<courmisch> devinheitmueller: no
<courmisch> yes but there is confusion
<courmisch> +no
<courmisch> there's ton of stuff being discussed inside videolabs
<courmisch> that I don't know of, or learn later
<devinheitmueller> courmisch: Oh sure. I wouldn’t expect you to have any knowledge of what’s going on in videolabs, but BBB was asking about videolan. That said, I am curious myself and haven’t seen any minutes from the meeting. I screwed up and thought it was on Sunday, only to find out on Saturday after I started my day that it was already over.
<courmisch> ask the secretary
<j-b> lol
<j-b> If I had managed VideoLAN and/or videolabs like the discussions right now here, those entities would have died years ago...
<j-b> Too many of you have too little understanding of business, contracts and management of people.
qeed has quit [Remote host closed the connection]
<j-b> Anyway, STF, if managed by SPI and Wiki and SoW will only get bounties/features, and they can only be accepted on releases (not even merge)
<j-b> Which is the contrary of what STF claims they want.
<Daemon404> that is what i still dont understand, and have said multiple times
<Daemon404> discrete projects are antithetical to stable maitenence
<j-b> Maintenance is beyond hard to put into SoW
<JEEB> yea, since it's an ongoing thing
<Daemon404> i just assumed STF doesnt understand OSS or software
<Daemon404> because government
<j-b> SoW accept only features, notably since it's limited in scope.
<j-b> and so if a feature takes 2x times, you need to bite the bullet. Like we do on commercial contracts
<BBB> j-b: to be fair, those of us that do understand don't want to talk about it, because we understand by having learned it the hard way <3
<j-b> BBB: yup, those looong nights because you gave the wrong deadline because the project is taking too much :´)
<BBB> I don't want to talk about it :')
<Daemon404> i heard you once had to make dav1d 2x faster
<devinheitmueller> If we’re talking about non-maintenance consulting work, as a general practice I put a specific clause into all my contracts making clear that payment is *not* contingent on changes being merged upstream. I’ve had too many cases with open source projects where the work done isn’t “generally applicable” enough, where significant changes need to be made to frameworks that will never get past review,
<devinheitmueller> where a feature isn’t considered important enough to the developers to give appropriate attention or merge, or sometimes developers on the upstream project are just a**holes.
<BBB> it depends on the work, a general rule is difficult to make
<BBB> for maintenance stuff (security fixes, coverity stuff, bugfixes, ..), I'd say it's easier to ask for then in completely new features, especially fringe features, yes
<BBB> and I don't know what that dav1d thing is you're talking about, my iphone has a hardware video decoder
<Daemon404> lol
<devinheitmueller> :-p
<BBB> (I bought a new one just so I can say that :-p)
<BBB> #avwon
<Lynne> does C have some weird limits on 32-bit systems I don't know about?
<Lynne> I get an "array is too large (512 elements)" error on a struct that's never allocated or used on struct
<Lynne> *stack
<Lynne> the type of the array is definitely not large enough for 512 entries to be over 4GB
<Daemon404> would be news to me
<Lynne> ah... I forgot I had a really huge struct in there
<JEEB> :D
<cosminaught> j-b: why would payment only be done on release rather than merge?
<jamrial> elenril: https://github.com/jamrial/FFmpeg/tree/ts_flags what i got so far
<cosminaught> > SoW accept only features, notably since it's limited in scope.
<cosminaught> This also doesn't sound right, I don't see why a SoW cannot be defined for doing maintenance rather than feature work
mkver has quit [Ping timeout: 264 seconds]
<haasn> does anybody know (or have the ability to test) vsrc_ddagrab?
<BtbN> I wrote the thing, soo, kinda
<JEEB> yea it is pretty usable on windows
<haasn> BtbN: can we move probe_output_format (+ dependencies) to query_formats() ?
<JEEB> not sure if it works on wine
<haasn> I need to know what the underlying device format is at format configuration time, for swfmt negotiation
<BtbN> hm, I can't say that on top of my head
<BtbN> JEEB: I'd be very surprised, since it interacts with the graphics driver at a relatively low level for wine
<haasn> oh, we don't even *have* a query_formats yet :D
<haasn> in that filter
<BtbN> I _think_ as long as probe_output_format has run successfully before the filter starts operating, it doesn't particularily care
<haasn> alternatively we could make allow_fallback=0 required behavior, but that would be more of a regression in functionality
<haasn> creating the hwdevice inside query_formats is established convention (vf_hwupload, vf_libplacebo)
<BtbN> Could get a bit complicated to move though, since a whole bunch of stuff in config_props depends on the results
<BtbN> looking at the doxy for config_props and query_func, ddagrab is not doing things correctly right now
<BtbN> Though technically it is, since the format is always D3D11
<haasn> BtbN: can you test this perhaps? https://0x0.st/Hdzb.txt
<haasn> oh, I suppose that won't compile for you without my series
<haasn> Wait, I'll throw up a branch
<BtbN> Is init_hwframes_ctx really AFTER all the other ones?
darkapex has quit [Remote host closed the connection]
<haasn> it would appear so
darkapex has joined #ffmpeg-devel
<BtbN> The patch is quite hard to read, but at least on first glance, init_dxgi_dda not being called unconditionally looks wrong
<haasn> the patch is a bit hard to read because I moved code to a different level of indentation
<haasn> (thanks git)
<haasn> github smart?
<BtbN> well, I told it to ignore whitespaces
<haasn> oh, git diff -w
<BtbN> One potential issue I see is that there might be a wonky error state
<BtbN> Or does returning EINVAL/any error from query_formats stop any retries?
<BtbN> Cause right now, if any of the error states in the "if (!dda->device_ctx)" happen, and then the function is re-tried, iit will potentially think it's already initialized, while it's actually not
iive has joined #ffmpeg-devel
<j-b> cosminaught: because this is often what happens for SoW in open source
BtbN has quit [Remote host closed the connection]
BtbN has joined #ffmpeg-devel
<BtbN> haasn: where did you get ff_set_common_sw_formats from? That function does not seem to exist? Is it just supposed to be ff_set_common_formats?
kierank has joined #ffmpeg-devel
<BtbN> Or is that from your branch?
<kierank> what did I miss
<kierank> apart from angry emails from paul
<BtbN> Oh, yeah. There's a lot more in that branch than that one patch
<BtbN> It's definitely missing an include for "format.h" though
<jamrial> kierank: libera.chat has logs
<kierank> I joke
<jamrial> didn't know if you knew, so might as well tell you about it :p
ccawley2011 has quit [Read error: Connection reset by peer]
cone-765 has quit [Quit: transmission timeout]
s55 has quit [Quit: ZNC 1.8.2 - https://znc.in]
s55 has joined #ffmpeg-devel
aljazmc has quit [Quit: Leaving]
<haasn> BtbN: fixed
epony has joined #ffmpeg-devel
HarshK23 has quit [Quit: Connection closed for inactivity]
<BtbN> haasn: it still works fine at least
<BtbN> hm, it REALLY misbehaves on my new 10bit/HDR screen
<BtbN> The moment I move my mouse cursor, it at runtime changes the format from 16 bit float to standard bgr
<BtbN> Does the same on master though
derpydoo has quit [Ping timeout: 264 seconds]
<BtbN> The only way I can get it to work reliably is if I force it to the 16 bit float format by default. It'll jump back to it on its own first "chance" it gets. Still not exactly sure what triggers it.
<BtbN> sadly, there are no consumers in ffmpeg which support AV_PIX_FMT_RGBAF16
<BtbN> I really don't understand why it decides to suddenly revert to BGRA the moment I move the mouse
<BtbN> no, the other way around actually. It starts in BGRA mode, and the moment I move the mouse, it suddenly realizes it's 16 bit float
<BtbN> hm, actually. Both cases can happen it seems. the hell
Daemon404 has left #ffmpeg-devel [#ffmpeg-devel]
<BtbN> Yeah, anything but 8bit capture is COMPLETELY busted
<BtbN> And to me it looks like the bustedness is on Microsofts side
<BtbN> Even if I tell it that I only accept 8 bit bgra, it will randomly switch to rgbf16 whenever it feels like it
<BtbN> mostly triggered by moving the mouse. Even on another screen