BtbN 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 7.0 is released
<KillerWasp> it doesn't matter anymore. I already have the answer. Once encoded in AVPacket I can use its stored buffer to encapsulate in RTMP, and that's all it takes to send.
<KillerWasp> I don't know why it's so complicated to answer. :/
Everything has quit [Quit: leaving]
<another|> It apparently wasn't clear what your question was.
<Marth64> furq: I remmeber this one now
<Marth64> they were complaining on handbrake channel too and shared a sample
<Marth64> dont think its a bug
<KillerWasp> another|: Curious, because whoever I asked was ChatGPT, and they answered me right away. A machine understands me more, as always. lol.
jtgd has quit [Quit: WeeChat 4.4.4]
jtgd has joined #ffmpeg
xx has quit [Ping timeout: 264 seconds]
gchound has joined #ffmpeg
lusciouslover has quit [Ping timeout: 248 seconds]
lusciouslover has joined #ffmpeg
minimal has quit [Quit: Leaving]
wziko has joined #ffmpeg
wziko has quit [Ping timeout: 246 seconds]
iive has quit [Quit: They came for me...]
meinside has joined #ffmpeg
<BtbN> Multiple people told you that multiple times
<BtbN> But apparently it wasn't the correct answer then...
<BtbN> AVPackets also don't go straight into rtmp. They first get muxed into flv, and that then gets passed to rtmp
DauntlessOne4 has quit [Quit: Ping timeout (120 seconds)]
DauntlessOne4 has joined #ffmpeg
neilt has joined #ffmpeg
JanC has joined #ffmpeg
JanC is now known as Guest1559
Guest1559 has quit [Killed (copper.libera.chat (Nickname regained by services))]
neilt has quit [Quit: Leaving]
lusciouslover has quit [Read error: Connection reset by peer]
lusciouslover has joined #ffmpeg
neilt has joined #ffmpeg
neilt has quit [Remote host closed the connection]
neilt has joined #ffmpeg
emmanuelux has joined #ffmpeg
JanC has joined #ffmpeg
JanC has quit [Killed (calcium.libera.chat (Nickname regained by services))]
realies has quit [Quit: ~]
sihloo has quit [Quit: ZNC 1.9.1 - https://znc.in]
sihloo has joined #ffmpeg
lolok has joined #ffmpeg
emmanuelux_ has joined #ffmpeg
emmanuelux has quit [Ping timeout: 244 seconds]
StephenLynx has quit [Quit: Leaving]
emmanuelux_ has quit [Read error: Connection reset by peer]
fairway has joined #ffmpeg
<fairway> everybody #ffmpeg channel is not aknowledging the huge bug i found
gchound has quit [Quit: WeeChat 4.1.1]
<Marth64> i assume you're talking about the wmv issue
<CounterPillow> fairway: yes I am not, what are you gonna do about it
<fairway> marth64 yes , were you able to produce the bug on your end
MRiddickW has quit [Ping timeout: 244 seconds]
<another|> CounterPillow: you already made the mistake of answering ;)
<fairway> another| why is nobody aknowledging this bug
<CounterPillow> there is an evil conspiracy against specifically you
<Traneptora> welder: pushing q will always exit, but if you want equal length, then the easiest way to do that is to use the -shortest command with some sort of pad
<Traneptora> e.g. if you know the audio will be shorter than the video, then you can do something like: ffmpeg -i video -i audio -lavfi "aevalsrc=0,[0:a]concat=n=2:v=0:a=1[a]" -map 0:v -map "[a]" -shortest output.mkv
<Traneptora> what this command does is it takes the audio stream and adds silence at the end of it, so -shortest will cause output to end when the video stream ends (since the audio stream will never end)
<another|> fairway: maybe because it isn't such a "huge bug" and your exaggeration is annoying
<Traneptora> what "huge bug" are you even talking about
<fairway> how is this not a huge bug
<Traneptora> I have no idea what you're even referring to, I just got here
<fairway> Traneptora ffmpeg new version is reporting 30 fps file as 60 fps
<fairway> old version did not have this issue
<furq> if you found a bug then report it on the bug tracker
<furq> and tell them grüß gott from me
<CounterPillow> I will raise this bug with the UN security council, I hope they can adopt a resolution
<furq> get günther fehlinger on the case
<another|> Yes. Big bug. Very huge. Call CISA. Get a CVE. /s
<CounterPillow> furq: excuse me, we need it fixed, not nuked. Everyone knows bugs survive nukes.
<Traneptora> fairway: is the file interlaced
<fairway> Traneptora no
<Traneptora> are you sure
<fairway> if you don't believe me, check it yourself
<fairway> every other app other than ffmpeg is reporting as 30 fps video
<Traneptora> fairway: the container of the file you linked reports 60 fps
<Traneptora> container fps values are known to be frequently inaccurate
<fairway> what are you talking about
<fairway> that's not 60 fps
<fairway> it's 30 fps
<fairway> ffmpeg version 4.2.xx reports as 30 fps too
<Traneptora> well first of all ffmpeg doesn't actually report the framerate as anything if you use git master, it reports the timebase rate, which is not the same thing
<Traneptora> the ASF header in the file you linked has the pulldown flag set
<Traneptora> this means it likely went through some sort of IVTC
<fairway> Traneptora okay i meant ffprobe.exe
<Traneptora> if you just run "ffprobe input.wmv" it doesn't actually report any fps value at all, which means it's definitely not reporting an incorrect one
<CounterPillow> why is fairway refusing to acknowledge this huge problem with their report? What evil machinations could be at play?
<fairway> Traneptora okay , why is it not reporting 30 fps then. i thought tpr = fps
<Traneptora> tbr and fps are notably not the same thing
<fairway> oh ok, what is the diffence
<Traneptora> tbr is the timebase rate, it's what the container reports. it's usually a multiple of the fps, sometimes 1, sometimes not
<Traneptora> however, the actual fps is based on the timestamp of the video frames
<Traneptora> determining the actual framerate of the video typically requires you to read a few frames and see what timestamps they're placed at
<Traneptora> as most containers support non-constant framerates, this is the only way to determine it
<fairway> but that video shouldn't be 60 anything
<fairway> even 60 tbr
<fairway> Traneptora also explain why ffprobe 4.2.1 reported as 30
<Traneptora> typically the true framerate is half of the actual tickrate
<Traneptora> at least it's very common
<Traneptora> looks like old ffmpeg is attempting to heuristically guess the framerate and is getting it wrong
<Traneptora> 30.30 is notably not the framerate of the video
<Traneptora> so what you've just shown is proof that the older version had some sort of bug in it
<fairway> that's true, it should be 29.97
<fairway> but old version do show 29.97 tbr
<fairway> where new version show 59.94 tbr
<Traneptora> well the video file you linked is a different video than the one you show in that screenshot
<Traneptora> so I'm not really sure what you expected
<fairway> yes, it's little different
<Traneptora> well then I don't really know how you expected anything else
<fairway> because i cannot find ffprobe v4.2.1
<fairway> i deleted it
<Traneptora> good, five-year-old versions of software tend to have issues with them
<fairway> the point is new version is showing 30 fps as 60 tbr and 29.97 fps as 59.94 tbr
<Traneptora> indeed, that's because the timebase rate (or tickrate) is typically twice the actual fps
<Traneptora> it's not required to be, but it's usually that
<Traneptora> so the tbr number reported is most likely correct, and the older version was reporting it incorrectly
<fairway> <Traneptora> indeed, that's because the timebase rate (or tickrate) is typically twice the actual fps <---- that's bs because if i do ffprobe test.mp4, there is no issue
<fairway> only 1080p wmv
<fairway> if i do ffprobe *.mp4 it shows correct fps and correct tbr
<Traneptora> well if you read a different file and get a different result you shouldn't be surprised
<Traneptora> since it's a different file in a different format
<Traneptora> what's happening here is that the file you linked has a tickrate of 60. this is usually but not always twice the framerate, and the container reports no actual framerate. so FFmpeg chooses not to report any framerate because it is not guessing
<Traneptora> the other software that reports "30fps" is guessing based on the tickrate
<Traneptora> the guess is probably correct, but FFmpeg won't report a frame rate as a guess
<fairway> also there is no issues with 480p wmv 540p wmv only 1080p
<Traneptora> ye, that's probably because the other files you tested have the actual framerate written in the file
<fairway> only 1080p wmv is doing this weird thingg
<Traneptora> that's probably because VC-1 for a while didn't support true progressive video so what you are watching is the result of an inverse telecine
<Traneptora> (for high resolution)
<fairway> and how do i tell if "actual framerate is written in file"?
<Traneptora> if ffmpeg reports it
stolen has joined #ffmpeg
<Traneptora> if ffmpeg doesn't report it, it generally means there isn't one
<Traneptora> if you run ffprobe -select_streams v -show_streams input.wmv, and look for the avg_frame_rate field, then that's going to be your best guess
<fairway> hmm i need to get hold of old version of ffprobe.exe
<Traneptora> uh, no
JanC is now known as Guest7956
JanC has joined #ffmpeg
Guest7956 has quit [Killed (tantalum.libera.chat (Nickname regained by services))]
<Marth64> i have the sample you shared yesterday
<Traneptora> if you run, for example, ffprobe -select_streams v -show_streams 100MB_1080P_THETESTDATA.COM_WMV.wmv
<Marth64> avg_frame_rate=0/0
<Traneptora> it'll show avg_frame_rate as 0/0
<Traneptora> that's because that data isn't in the file
<Traneptora> so what you can do is check avg_frame_rate, and use that if you find it
<fairway> if ffmpeg doesn't report it, it generally means there isn't one <--- then explain why v4.2.1 reported the 29.97 30.30 fps
<Traneptora> because old ffprobe would guess
<Marth64> its literally several years old and the application has evolved
<Traneptora> if avg_frame_rate is 0/0, that means it's not there. you can check r_frame_rate as a fallback and divide by 2, but a decision was made at some point to not guess the frame rate if it's not actually known
<fairway> i get this: avg_frame_rate=30000/1001 when i did test480p.wmv
<Traneptora> that's 29.97
<fairway> i see
<Traneptora> divide 30k by 1001, that's the number you get
<Traneptora> it's closer to 29.97003, but everyone just rounds it for convenience
<fairway> so that means test480p.wmv has actual framerate written in file?
<Traneptora> most likely, yes
<fairway> if i do ffprobev421.exe -select_streams -show_streams test1080p.wmv
<fairway> i wonder what it would show as avg_frame_rate
<Marth64> do you have an option to not use wmv format and vc1 codec
<Marth64> i understand if its archived but wondering if its not and if you have other options
<fairway> no
<Traneptora> considering "old.school.wmv" was one of the filenames this is probably a file that was created a long time ago
<Traneptora> and they're just going over it now
<fairway> marth64 basically you are saying. let's avoid the issue, instead of fixing the issue"
<Traneptora> Marth64 is just pointing out that if you are currently creating new files, then VC-1 inside WMV is not an ideal choice
<Traneptora> it was unrelated to the question you had
<Marth64> no. i'm saying if this is a self inflected issue because you are using old formats and codecs, consider to use newer and more flexible options
<Marth64> but i understand why its not
<Traneptora> I mean, most likely the files were created a long time ago
<Traneptora> if I had to guess
<Marth64> ya
<Traneptora> either way fairway I need to sleep right now but that's the gist of what's going on. if you're sure you know what the framerate is you can do `ffmpeg -i input.wmv -map 0 -c copy -r 30 output.mkv` or something like that
<Traneptora> which coerces it to have 30fps. if it isn't actually 30fps this will cause errors
<Traneptora> or rather, cause issues with the video
Traneptora has quit [Quit: Quit]
qubuepe24 has joined #ffmpeg
wziko has joined #ffmpeg
wziko has quit [Max SendQ exceeded]
coldfeet has joined #ffmpeg
neilt has quit [Quit: KVIrc 5.2.6 Quasar http://www.kvirc.net/]
MRiddickW has joined #ffmpeg
coldfeet has quit [Remote host closed the connection]
fairway has quit [Ping timeout: 260 seconds]
System_Error has quit [Remote host closed the connection]
System_Error has joined #ffmpeg
xx has joined #ffmpeg
dreamon has joined #ffmpeg
TheSashmo has joined #ffmpeg
stolen has quit [Quit: Connection closed for inactivity]
Corroddity has joined #ffmpeg
coldfeet has joined #ffmpeg
kepstin has quit [Remote host closed the connection]
kepstin has joined #ffmpeg
lavaball has joined #ffmpeg
gvg has quit [Ping timeout: 272 seconds]
qubuepe24 has quit [Quit: Leaving]
gvg has joined #ffmpeg
meinside has quit [Quit: Connection closed for inactivity]
evilscreww has joined #ffmpeg
sunyibo_ is now known as sunyibo
jemius has joined #ffmpeg
YuGiOhJCJ has joined #ffmpeg
wziko has joined #ffmpeg
wziko has quit [Ping timeout: 265 seconds]
fairway has joined #ffmpeg
coldfeet has quit [Remote host closed the connection]
EmleyMoor has quit [Ping timeout: 246 seconds]
EmleyMoor has joined #ffmpeg
StephenLynx has joined #ffmpeg
kasper93_ is now known as kasper93
Ohtani has joined #ffmpeg
coldfeet has joined #ffmpeg
makidoll has quit [Remote host closed the connection]
makidoll has joined #ffmpeg
evilscreww has quit [Ping timeout: 250 seconds]
evilscreww has joined #ffmpeg
evilscreww has quit [Remote host closed the connection]
jpsollie has quit [Ping timeout: 252 seconds]
JanC has joined #ffmpeg
JanC is now known as Guest5586
Guest5586 has quit [Ping timeout: 252 seconds]
TheSashmo has quit [Read error: Connection reset by peer]
TheSashmo has joined #ffmpeg
lucasta has joined #ffmpeg
hwm4rgs has quit [Quit: Bye.]
hwm4rgs has joined #ffmpeg
hwm4rgs has quit [Client Quit]
hwm4rgs has joined #ffmpeg
dreamon has quit [Ping timeout: 265 seconds]
YuGiOhJCJ has quit [Quit: YuGiOhJCJ]
rsx has joined #ffmpeg
Vonter has quit [Quit: WeeChat 4.4.3]
coldfeet has quit [Remote host closed the connection]
wziko has joined #ffmpeg
TheSashmo has quit [Read error: Connection reset by peer]
TheSashmo has joined #ffmpeg
alexherbo2 has joined #ffmpeg
wziko has quit [Ping timeout: 246 seconds]
rsx has quit [Quit: rsx]
gchound has joined #ffmpeg
alexherbo2 has quit [Remote host closed the connection]
alexherbo2 has joined #ffmpeg
wziko has joined #ffmpeg
wziko has quit [Max SendQ exceeded]
alexherbo2 has quit [Remote host closed the connection]
haritz has quit [Read error: Connection reset by peer]
haritz has joined #ffmpeg
rex has quit [Ping timeout: 252 seconds]
Magissia has quit [Ping timeout: 244 seconds]
rex has joined #ffmpeg
Dagger has quit [Ping timeout: 265 seconds]
billchenchina- has joined #ffmpeg
Dagger has joined #ffmpeg
billchenchina has quit [Ping timeout: 260 seconds]
billchenchina has joined #ffmpeg
billchenchina- has quit [Ping timeout: 260 seconds]
jemius has quit [Quit: Leaving]
<noobaroo> hey, i have some hour long videos of static camera footage in a mostly barren room with tiny little activity throughout, not really in particular groups of time
<noobaroo> 30fps is pointless so I'm thinking of re-encoding to 12fps but honestly it would be even better if i could manually trim it to 5 seconds burst cuts in a loop so it has normal 30fps but skips . Like play 5, skip 5, repeat until end
<noobaroo> I dont know how to google this question though, can someone offer me a start in the right direction?
Magissia has joined #ffmpeg
minimal has joined #ffmpeg
emmanuelux has joined #ffmpeg
<vlt> noobaroo: I would propably use a script to generate a trim and concat filter_complex string :D
zsoltiv_ has joined #ffmpeg
KillerWasp has left #ffmpeg [YOU KIDDING ME?? YOU KIDDING MEEE???? (PC drinking beer) *kernel panic!*]
lavaball has quit [Quit: lavaball]
coldfeet has joined #ffmpeg
soreau has joined #ffmpeg
rvalue- has joined #ffmpeg
rvalue has quit [Ping timeout: 248 seconds]
<soreau> Is it possible to fade out the last few seconds of audio when creating a time lapse video from still images in one step?
<soreau> I was kind of expecting something like -af "afade=t=out:st=-3:d=3" to work, but apparently negative values don't work
rvalue- is now known as rvalue
MRiddickW has quit [Ping timeout: 248 seconds]
coldfeet has quit [Remote host closed the connection]
dreamon has joined #ffmpeg
wziko has joined #ffmpeg
lucasta has quit [Quit: Leaving]
jemius has joined #ffmpeg
\\Mr_C\\ has joined #ffmpeg
billchenchina has quit [Remote host closed the connection]
jemius has quit [Quit: Leaving]
YuGiOhJCJ has joined #ffmpeg
realies has joined #ffmpeg
realies has quit [Client Quit]
dreamon has quit [Ping timeout: 276 seconds]
hsv has joined #ffmpeg
hsv has quit [Client Quit]
Marth64 has quit [Quit: Leaving]
wziko has quit [Ping timeout: 265 seconds]
Marth64 has joined #ffmpeg
wziko has joined #ffmpeg
gchound has quit [Quit: WeeChat 4.1.1]
Peetz0r has quit [Remote host closed the connection]
Dagger has quit [Ping timeout: 265 seconds]
Dagger has joined #ffmpeg
SuicideShow has quit [Ping timeout: 246 seconds]
chandash has joined #ffmpeg
SuicideShow has joined #ffmpeg
meinside has joined #ffmpeg
Dagger has quit [Ping timeout: 265 seconds]
Dagger has joined #ffmpeg
chandash has quit [Quit: WeeChat 4.4.2]
chandash has joined #ffmpeg
wziko has quit [Ping timeout: 248 seconds]