ChanServ changed the topic of #ffmpeg to: Welcome to the FFmpeg USER support channel | Development channel: #ffmpeg-devel | Bug reports: https://ffmpeg.org/bugreports.html | Wiki: https://trac.ffmpeg.org/ | This channel is publically logged | FFmpeg 7.1.1 is released
jtgd has quit [Read error: Connection reset by peer]
chair1 has joined #ffmpeg
jtgd has joined #ffmpeg
minimal has quit [Quit: Leaving]
intrac has quit [Ping timeout: 252 seconds]
UltraFuzzy has joined #ffmpeg
intrac has joined #ffmpeg
iive has quit [Quit: They came for me...]
chair8 has joined #ffmpeg
chair1 has quit [Ping timeout: 250 seconds]
chair8 is now known as chair1
chair22 has joined #ffmpeg
YUiNA has quit [Remote host closed the connection]
chair1 has quit [Ping timeout: 250 seconds]
chair22 is now known as chair1
chair13 has joined #ffmpeg
chair1 has quit [Ping timeout: 250 seconds]
chair52 has joined #ffmpeg
chair13 has quit [Ping timeout: 250 seconds]
chair52 is now known as chair1
intrac has quit [Ping timeout: 245 seconds]
intrac has joined #ffmpeg
chair1 has quit [Ping timeout: 250 seconds]
user23 has quit [Remote host closed the connection]
_JM55 has joined #ffmpeg
_JM55 has quit [Quit: Leaving]
defa has quit [Ping timeout: 245 seconds]
paulk has quit [Read error: Connection reset by peer]
paulk-bis has joined #ffmpeg
damian101__ has joined #ffmpeg
damian101_ has quit [Ping timeout: 248 seconds]
maxim_d33 has quit [Ping timeout: 245 seconds]
vincejv has quit [Quit: Bye bye! Leaving for now...]
maxim_d33 has joined #ffmpeg
burie has joined #ffmpeg
<burie>
I just want to test outputting graphics where text periodically update by user intervention with ffmpeg. I have a command like this d ffmpeg -f lavfi -i color=size=320x240:color=black -vf "drawtext=textfile=test.txt:font='Times New Roman':x=(main_w-text_w-10):y=(main_h-text_h-10):fontsize=32:fontcolor=black:box=1:boxcolor=white@0.5:boxborderw=5" -framerate 25 test.mp4 but it creates a file several minutes long in just a couple of seconds
<burie>
. How can I change the command so the output duration is the same as the amount of time ffmpeg is running so the user can change text and the output won't have a gap of several minutes in between changes
<furq>
-re before -i
<furq>
also -framerate 25 does nothing there
emmanuelux has quit [Ping timeout: 256 seconds]
<furq>
but it defaults to 25 anyway
<ossifrage>
Has anyone had problems with glitchy xHE-AAC audio [both fplay and my ipad glitch on the file, but the source claims it is fine]
lolok has joined #ffmpeg
jiffy__ has quit [Ping timeout: 248 seconds]
rodrigo-morales has quit [Server closed connection]
rodrigo-morales has joined #ffmpeg
YUiNA has joined #ffmpeg
Kimapr has quit [Ping timeout: 256 seconds]
<burie>
furq: that worked.
<burie>
Next question, if I host mp4 vp8 or h264 otherwise with html video tag mpv/firefox can stream it. how can I get ffmpeg to to allow mpv to stream/read it before ffmpeg exits and finishes writing to the file
l4yer has joined #ffmpeg
Moon_Rabbit has quit [Server closed connection]
Moon_Rabbit has joined #ffmpeg
JanC is now known as Guest262
Guest262 has quit [Killed (cadmium.libera.chat (Nickname regained by services))]
JanC has joined #ffmpeg
gordonfish- has joined #ffmpeg
acryo_ has joined #ffmpeg
FH_thecat has quit [Ping timeout: 258 seconds]
oribellow_ has joined #ffmpeg
Mister_Magister_ has joined #ffmpeg
snoriman has quit [Ping timeout: 258 seconds]
gordonfish has quit [Ping timeout: 258 seconds]
Rena has quit [Ping timeout: 258 seconds]
beastd has quit [Ping timeout: 258 seconds]
oribellow has quit [Ping timeout: 258 seconds]
acryo has quit [Ping timeout: 258 seconds]
Mister_Magister has quit [Ping timeout: 258 seconds]
Rena has joined #ffmpeg
snoriman has joined #ffmpeg
Mister_Magister_ is now known as Mister_Magister
oribellow_ is now known as oribellow
gordonfish- is now known as gordonfish
beastd has joined #ffmpeg
MooMoo has quit [Quit: Leaving]
Vonter has quit [Ping timeout: 255 seconds]
JanC has quit [Ping timeout: 260 seconds]
damian101_ has joined #ffmpeg
damian101__ has quit [Ping timeout: 256 seconds]
YUiNA has quit [Remote host closed the connection]
meklu has quit [Server closed connection]
meklu has joined #ffmpeg
damian101__ has joined #ffmpeg
damian101_ has quit [Ping timeout: 256 seconds]
lavaball has joined #ffmpeg
acidbunny has quit [Remote host closed the connection]
KombuchaKip has quit [Quit: Leaving.]
KombuchaKip has joined #ffmpeg
andm100 has joined #ffmpeg
andm100 has quit [Client Quit]
JanC has joined #ffmpeg
damian101_ has joined #ffmpeg
SohamK has quit [Server closed connection]
SohamK has joined #ffmpeg
damian101__ has quit [Ping timeout: 256 seconds]
damian101__ has joined #ffmpeg
damian101_ has quit [Ping timeout: 258 seconds]
FH_thecat has joined #ffmpeg
damian101_ has joined #ffmpeg
ChiefMooMoo has joined #ffmpeg
<ChiefMooMoo>
I'm trying to understand var_stream_map and the -map command. I'm currently looking at this example of a var_stream_map:
<ChiefMooMoo>
why does it -map 0:v -map 0:a -map 0:v -map 0:a ? Isn't that redundant?
YuGiOhJCJ has quit [Ping timeout: 272 seconds]
damian101__ has joined #ffmpeg
damian101_ has quit [Ping timeout: 256 seconds]
YuGiOhJCJ has joined #ffmpeg
lavaball has quit [Remote host closed the connection]
defa has joined #ffmpeg
lavaball has joined #ffmpeg
jmcantrell_ has quit [Ping timeout: 252 seconds]
damian101_ has joined #ffmpeg
damian101__ has quit [Ping timeout: 245 seconds]
vincejv has joined #ffmpeg
pelotron has quit [Ping timeout: 272 seconds]
pelotron has joined #ffmpeg
fgarcia has quit [Ping timeout: 248 seconds]
llinguini has quit [Server closed connection]
llinguini has joined #ffmpeg
fgarcia has joined #ffmpeg
___nick___ has joined #ffmpeg
___nick___ has quit [Client Quit]
___nick___ has joined #ffmpeg
boxrick has quit [Server closed connection]
boxrick has joined #ffmpeg
SuicideShow has quit [Ping timeout: 245 seconds]
fgarcia has quit [Quit: Remote host closed the connection]
lolok has quit [Quit: lolok]
SuicideShow has joined #ffmpeg
Hassan01 has quit [Server closed connection]
Hassan01 has joined #ffmpeg
rsx has joined #ffmpeg
Sketch has quit [Remote host closed the connection]
Sketch has joined #ffmpeg
fgarcia has joined #ffmpeg
fengdaolong has joined #ffmpeg
cmcsun has quit [Server closed connection]
sun has joined #ffmpeg
fengdaolong has quit [Remote host closed the connection]
fengdaolong has joined #ffmpeg
Blacker47 has joined #ffmpeg
lolok has joined #ffmpeg
coldfeet has joined #ffmpeg
lavaball has quit [Remote host closed the connection]
Shine_ has joined #ffmpeg
jeroen___ has joined #ffmpeg
<jeroen___>
Hello, good morning
lavaball has joined #ffmpeg
<jeroen___>
Q: I maintain bindings for the R programming language. ffmpeg-8 seems to have some breaking changes and I have a hard time understanding how to port things.
<JEEB>
docs/APIchanges IIRC should have a list of changes, otherwise feel free to query around here while linking to relevant spots from code.ffmpeg.org
* jeroen___
My code is giving a lot of these errors:
<jeroen___>
Option 'sample_fmts' is not a runtime option and so cannot be set after the object has been initialized
<jeroen___>
Option 'pix_fmts' is not a runtime option and so cannot be set after the object has been initialized
<JEEB>
that message effectively seems to be an avfilter one (since while it is global, I only see it utilized there)
<JEEB>
jeroen___: I think I possibly utilized something else in my avfilter code (will have to double-check when I get to my private laptop), but basically that would limit things to the arguments of graph_create_filter
<JEEB>
most likely since it gets initialized at that point and you don't want to let the API user touch it after that logic runs (possibly new logic added due to the newly added options)
<JEEB>
yea this was actually part of a patch set to just improve the initialization check logic
<JEEB>
so previously there indeed was a initialized int, but I bet it was not generally observed while it should have been
<JEEB>
now it actually became properly strict since the checking logic is centralized
<JEEB>
yea AV_OPT_FLAG_RUNTIME_PARAM is an old thing
<jeroen___>
Is there an easy way I can adapt my bindings for ffmpeg8?
<JEEB>
yea, pretty sure there was a backwards compatible manner of doing this which previously and currently works
<JEEB>
if you want to keep using that avfilter_graph_create_filter function, I'm pretty sure one of its arguments can just be utilized for setting that stuff that you're setting atm
<JEEB>
as I see it being utilized in fftools/ffmpeg_filter.c
<JEEB>
then there's doc/examples which contains various stuff which are more or less limited to something very specific
<JEEB>
alternatively you can first just allocate/create the sink/src
<JEEB>
then set the options like you currently are
<JEEB>
and then init
<JEEB>
yea, that helper function is avfilter_graph_alloc_filter and avfilter_init_str
<jeroen___>
Hmm yes replacing avfilter_graph_create_filter with avfilter_graph_alloc_filter and avfilter_init_str seems to get rid of the printed warnings, but my tests fail because the trasnformation no longer seem to output the correct lenght :/
<JEEB>
they are an improvement and expect that the old ones will at some point start failing. but that's currently separate from the runtime option change
durstloescher has joined #ffmpeg
xx has quit [Remote host closed the connection]
xx has joined #ffmpeg
JanC is now known as Guest229
Guest229 has quit [Killed (silver.libera.chat (Nickname regained by services))]
JanC has joined #ffmpeg
<jeroen___>
OK that means I have yet another unrelated problem in ffmpeg-8
fengdaolong has quit [Ping timeout: 260 seconds]
<jeroen___>
If I transform an audio file to adjust the sample rate, it just seems to stretch the duration of the audio.
<jeroen___>
One test (that works with ffmpeg 4.x-7.x) reduces the sample rate from 44100 to 16000. But with ffmpeg-8 the output file goes from 30s to 82.8s. So I guess it didn't resample but just slowed down the audio.
<JEEB>
sounds like instead of getting aresample auto-inserted, you are now just setting the input information to that sample rate
<JEEB>
this should work with older versions as well, but you can call on the graph `avfilter_graph_set_auto_convert(fgt->graph, AVFILTER_AUTO_CONVERT_NONE)`, which will instead of auto-insertion error out
<JEEB>
and if it does not fail now, you are essentially configuring the input to be interpreted like that
<JEEB>
(I did not check if you add an aresample filter in there, if you did actually do that then sorry for missing it)
durstloescher has quit [Ping timeout: 248 seconds]
YuGiOhJCJ has quit [Quit: YuGiOhJCJ]
Shine_ has quit [Ping timeout: 244 seconds]
signalhunter has quit [Server closed connection]
signalhunter has joined #ffmpeg
Kei_N has joined #ffmpeg
Kei_N_ has quit [Ping timeout: 248 seconds]
<jeroen___>
Hmm we do seem to set `AVCodecContext->sample_rate` for the encoder and later `av_opt_set_bin(buffersink_ctx, "sample_rates", ....)` on the filter. That used to work. Is aresample a property I would need to set with `av_opt_set_bin` ?
<JEEB>
if you have the auto-conversion on, the filter graph configuration step would check the values from input and output, and auto-insert the aresample filter if the values for output differ from what the input is (just like the scale filter for pixel format conversions etc)
<JEEB>
if you don't have auto-conversion on, it will instead error out and rqeuire you to manually add an instance of the aresample filter into the graph
Blacker47 has quit [Quit: Life is short. Get a V.90 modem fast!]
<jeroen___>
Setting `avfilter_graph_set_auto_convert` to `AVFILTER_AUTO_CONVERT_NONE` or `AVFILTER_AUTO_CONVERT_ALL` does not seem to change any behavior, afaict. Is the default to have it on?
<JEEB>
the default is on
<JEEB>
and that matches with you setting the values to the same between input and output, since that would of course mean that things passthrough
<JEEB>
while on the older version if you were doing any proper resampling it would then insert it, and thus setting it to NONE would actually fail
<JEEB>
(also setting the logging verbosity level to verbose or higher - verbose being the last non-spammy one - would log that)
<JEEB>
avfilter_graph_dump also dumps the resulting filter graph if you ever need that for debugging
Fenrir has quit [Server closed connection]
Fenrir has joined #ffmpeg
Fenrir has quit [Changing host]
Fenrir has joined #ffmpeg
YUiNA has joined #ffmpeg
moviuro has quit [Remote host closed the connection]
<jeroen___>
FWIW, setting `av_opt_set_bin(buffersink_ctx, "samplerates" ...)` does not do anything anymore in ffmpeg-8. It only works with the new `av_opt_set_array(buffersink_ctx, "samplerates"...)`.
<jeroen___>
Ugh sorry I meant setting the old `"sample_rates"` does not work anymore.
Blacker47 has joined #ffmpeg
<jeroen___>
Sadly this means more special casing to make the bindings work with ffmpeg 6,7,8
chowbok has quit [Read error: Connection reset by peer]
chowbok has joined #ffmpeg
moviuro has quit [Quit: Reboot? Or did my jail(8) just die?]
l4yer has quit [Remote host closed the connection]
l4yer has joined #ffmpeg
lavaball has quit [Remote host closed the connection]
lavaball has joined #ffmpeg
JanC has quit [Ping timeout: 245 seconds]
termos__ has quit []
turlando_ has joined #ffmpeg
turlando has quit [Ping timeout: 260 seconds]
JanC has joined #ffmpeg
Kimapr has quit [Remote host closed the connection]
Kimapr has joined #ffmpeg
JanC has quit [Ping timeout: 260 seconds]
JanC has joined #ffmpeg
lusciouslover has quit [Remote host closed the connection]
lusciouslover has joined #ffmpeg
HarshK23 has joined #ffmpeg
JanC has quit [Killed (calcium.libera.chat (Nickname regained by services))]
JanC has joined #ffmpeg
lavaball has quit [Ping timeout: 256 seconds]
acidbunny has joined #ffmpeg
jeroen___ has quit [Quit: Connection closed for inactivity]
lavaball has joined #ffmpeg
Guest40 has joined #ffmpeg
Guest20 has joined #ffmpeg
iive has joined #ffmpeg
jmcantrell_ has joined #ffmpeg
Guest40 has quit [Ping timeout: 250 seconds]
jmcantrell_ is now known as jmcantrell
<Guest20>
Hello, I have a question.
<Guest20>
I have a .webm file that contains both audio and video. When extracting the audio to another .webm file using ffmpeg -i original.webm -vn -c:a copy audio.webm and then checking the hash of the stream with ffmpeg -i audio.webm -f md5 -, it matches up with the hash of the audio stream of the original file (as is expected). Since the original file's
<Guest20>
audio stream uses opus encoding (exiftool shows A_OPUS), I can also extract the audio stream to an .ogg file with ffmpeg -i original.webm -vn -c:a copy audio.ogg` However, when checking the audio hash of the .ogg file, that hash does not match up with that of the original stream, even though I used c:a copy, which, as far as my understanding goes,
<Guest20>
should mean that the stream is not modified. Sorry if I am missing something - did I just come across a bug?
<another|>
hmm.. can reproduce. weird
System_Error has quit [Remote host closed the connection]
JanC is now known as Guest7234
Guest7234 has quit [Killed (tantalum.libera.chat (Nickname regained by services))]
JanC has joined #ffmpeg
Guest40 has joined #ffmpeg
Guest40 has quit [Write error: Broken pipe]
Guest40 has joined #ffmpeg
Guest20 has quit [Ping timeout: 250 seconds]
manwithluck has quit [Ping timeout: 256 seconds]
SketchyInternetC has joined #ffmpeg
SketchyInternetC is now known as Guest20
Guest20 is now known as SketchyInternet
Guest40 has quit [Ping timeout: 250 seconds]
lolok has quit [Quit: lolok]
Kimapr has quit [Remote host closed the connection]
Kimapr has joined #ffmpeg
Blacker47 has quit [Quit: Life is short. Get a V.90 modem fast!]
jmcantrell has quit [Ping timeout: 245 seconds]
Narrat has joined #ffmpeg
jmcantrell_ has joined #ffmpeg
___nick___ has quit [Ping timeout: 256 seconds]
System_Error has joined #ffmpeg
vincejv has quit [Ping timeout: 256 seconds]
chair1 has joined #ffmpeg
SketchyInternetC has joined #ffmpeg
vincejv has joined #ffmpeg
SketchyInternet has quit [Ping timeout: 250 seconds]
Traneptora has quit [Quit: Quit]
realies9346 has quit [Quit: ~]
HarshK23 has quit [Quit: Connection closed for inactivity]
Traneptora has joined #ffmpeg
System_Error has quit [Remote host closed the connection]
System_Error has joined #ffmpeg
chair95 has joined #ffmpeg
chair1 has quit [Ping timeout: 250 seconds]
jmcantrell_ is now known as jmcantrell
chair95 is now known as chair1
psykose has quit [Remote host closed the connection]
psykose has joined #ffmpeg
chair1 has quit [Ping timeout: 250 seconds]
another| has quit [Server closed connection]
another| has joined #ffmpeg
SketchyInternetC has left #ffmpeg [#ffmpeg]
lavaball has quit [Remote host closed the connection]
Kimapr_ has joined #ffmpeg
Kimapr has quit [Remote host closed the connection]
fristed has quit [Server closed connection]
fristed has joined #ffmpeg
Haven0320-2 has joined #ffmpeg
JanC has quit [Killed (calcium.libera.chat (Nickname regained by services))]
JanC has joined #ffmpeg
Forza has quit [Ping timeout: 258 seconds]
fristed has quit [Ping timeout: 258 seconds]
fgarcia has quit [Ping timeout: 258 seconds]
fristed has joined #ffmpeg
Haven0320 has quit [Ping timeout: 258 seconds]
chowbok has quit [Ping timeout: 258 seconds]
chowbok has joined #ffmpeg
fgarcia has joined #ffmpeg
emersion has joined #ffmpeg
emersion has left #ffmpeg [#ffmpeg]
YUiNA has joined #ffmpeg
Narrat has quit [Quit: They say a little knowledge is a dangerous thing, but it's not one half so bad as a lot of ignorance.]
Shuriko has quit [Ping timeout: 256 seconds]
Kimapr_ has quit [Remote host closed the connection]
Forza has joined #ffmpeg
Kimapr_ has joined #ffmpeg
Miyu has joined #ffmpeg
hackkitten has quit [Read error: Connection reset by peer]
chair1 has joined #ffmpeg
Kimapr_ has quit [Remote host closed the connection]
Kimapr_ has joined #ffmpeg
Miyu has quit [Ping timeout: 260 seconds]
hackkitten has joined #ffmpeg
minimal has quit [Ping timeout: 245 seconds]
YUiNA has quit [Remote host closed the connection]