lavaball has quit [Remote host closed the connection]
rvalue has quit [Ping timeout: 264 seconds]
Jhonny2x4 has quit [Quit: WeeChat 4.2.1]
Jhonny2x4 has joined #ffmpeg
fossdd has joined #ffmpeg
<BtbN>
you can consider support for those dropped, but it might just as well just work
epony has joined #ffmpeg
<BtbN>
nobody tests it anymore
rvalue has joined #ffmpeg
navi has quit [Quit: WeeChat 4.1.2]
fossdd has quit [Ping timeout: 255 seconds]
fossdd has joined #ffmpeg
<Juest>
last time i tried it appears that ffmpeg 5.1 doesn't run on windows 7
<Juest>
or 5.0.2
<Juest>
i dont remember
fossdd has quit [Ping timeout: 276 seconds]
Jhonny2x4 has quit [Quit: WeeChat 4.2.1]
fossdd has joined #ffmpeg
fossdd has quit [Ping timeout: 255 seconds]
fossdd has joined #ffmpeg
Tano has joined #ffmpeg
fossdd has quit [Ping timeout: 272 seconds]
fossdd has joined #ffmpeg
omegatron has quit [Quit: Power is a curious thing. It can be contained, hidden, locked away, and yet it always breaks free.]
bitbinge has quit [Ping timeout: 255 seconds]
fossdd has quit [Ping timeout: 276 seconds]
fossdd has joined #ffmpeg
<kepstin>
i'd honestly be surprised if you couldn't build and run modern ffmpeg on windows 7, but prebuilt ones might not work and you might have to disable some features if they require newer windows.
minimal has quit [Quit: Leaving]
jarthur has quit [Quit: jarthur]
fossdd has quit [Ping timeout: 255 seconds]
lexano has quit [Ping timeout: 272 seconds]
fossdd has joined #ffmpeg
cxc has quit []
JanC_ has joined #ffmpeg
JanC is now known as Guest1583
JanC_ is now known as JanC
Guest1583 has quit [Ping timeout: 256 seconds]
fossdd has quit [Ping timeout: 256 seconds]
fossdd has joined #ffmpeg
fossdd has quit [Ping timeout: 268 seconds]
fossdd has joined #ffmpeg
LionEagle has quit [Ping timeout: 272 seconds]
LionEagle has joined #ffmpeg
five618480 has quit [Remote host closed the connection]
five618480 has joined #ffmpeg
fossdd has quit [Ping timeout: 276 seconds]
fossdd has joined #ffmpeg
lemourin has quit [Read error: Connection reset by peer]
lemourin has joined #ffmpeg
fossdd has quit [Ping timeout: 264 seconds]
<FlorianBad>
If I was making -thread_queue_size proportional to something so that it stops complaining, but still doesn't run out of memory, what should it in theory be proportional to? I was thinking resolution surface (width x height)
fossdd has joined #ffmpeg
<FlorianBad>
My script encodes for example 12 resolutions of input each to 5 bitrates of output (in 12 commands, 1 per resolution, no rescale). And of course with the low rest it's fine, but when we get to 8K footage it becomes difficult to nail the right thread_queue_size
<FlorianBad>
And I can't just try and see because it takes absolutely forever. So that leads me back to my question: in theory to what should it be proportional?
waleee has quit [Ping timeout: 272 seconds]
fossdd has quit [Ping timeout: 255 seconds]
fossdd has joined #ffmpeg
waleee has joined #ffmpeg
fossdd has quit [Ping timeout: 256 seconds]
fossdd has joined #ffmpeg
fossdd has quit [Ping timeout: 256 seconds]
Tinos has joined #ffmpeg
fossdd has joined #ffmpeg
epony has quit [Read error: Connection reset by peer]
zsoltiv__ has joined #ffmpeg
zsoltiv_ has quit [Ping timeout: 255 seconds]
qaph has joined #ffmpeg
kron has quit [Ping timeout: 256 seconds]
AbleBacon has quit [Read error: Connection reset by peer]
qaph is now known as kron
jagannatharjun has joined #ffmpeg
Suchiman has quit [Quit: Connection closed for inactivity]
Vonter has quit [Ping timeout: 260 seconds]
Vonter has joined #ffmpeg
waleee has quit [Ping timeout: 240 seconds]
flotwig_ has quit [Excess Flood]
flotwig has joined #ffmpeg
dead1 has joined #ffmpeg
coala has joined #ffmpeg
Ox7C5_ has joined #ffmpeg
JanC has quit [Ping timeout: 256 seconds]
JanC has joined #ffmpeg
<FlorianBad>
I know I already asked this but I'd like to give it one more try ;) > Is there a way to define the (approximate) duration of EACH INDIVIDUAL segments, so just like seg_duration but as a list (similar to what -force_key_frames does for keyframes)
<FlorianBad>
(that's for audio, which is why I can't just use the keyframes for that)
jtgd has quit [Quit: WeeChat 4.2.1]
lusciouslover has quit [Ping timeout: 268 seconds]
lusciouslover has joined #ffmpeg
jtgd has joined #ffmpeg
<FlorianBad>
What I mean is something like: seg_duration_seconds_list=2.5,2.7,2,3.1,2.4 as opposed to an average: seg_duration=2.5
FH_thecat has quit [Quit: Leaving]
jagannatharjun has quit [Quit: Connection closed for inactivity]
fling has quit [Remote host closed the connection]
fling has joined #ffmpeg
ivanich has joined #ffmpeg
ivanich has quit [Remote host closed the connection]
ivanich has joined #ffmpeg
ivanich has quit [Remote host closed the connection]
<dv_>
can SPS / PPS NALUs be inserted in front of any frame, not just IDR frames?
<dv_>
in h264 I mean
<dv_>
with intra refresh, you don't have IDR frames (other than the very first frame), so this sounds logical
<JEEB>
you could verify the H.264 spec since it is free, but indeed I am not sure there is a requirement for them to be only at random access points (IDR or random access point SEI + I)
epony has joined #ffmpeg
ahc has quit [Quit: Leaving]
Wenbin_Chen_ has quit [Quit: Wenbin_Chen_]
Ox7C5_ has quit [Ping timeout: 260 seconds]
Ox7C5_ has joined #ffmpeg
FH_thecat has joined #ffmpeg
hightower3 has quit [Ping timeout: 260 seconds]
Suchiman has joined #ffmpeg
Ox7C5_ has quit [Ping timeout: 240 seconds]
Ox7C5_ has joined #ffmpeg
epony has quit [Remote host closed the connection]
HarshK23 has joined #ffmpeg
bschindler has joined #ffmpeg
epony has joined #ffmpeg
beaver has joined #ffmpeg
bitbinge has joined #ffmpeg
bitbinge has quit [Ping timeout: 255 seconds]
bitbinge has joined #ffmpeg
coala has quit [Quit: WeeChat 4.1.3]
bschindler has quit [Ping timeout: 250 seconds]
Ox7C5_ has quit [Quit: Lost terminal]
\\Mr_C\\ has joined #ffmpeg
Gaboradon has joined #ffmpeg
JanC_ has joined #ffmpeg
JanC has quit [Killed (zinc.libera.chat (Nickname regained by services))]
JanC_ is now known as JanC
YuGiOhJCJ has joined #ffmpeg
bschindler has joined #ffmpeg
beaver has quit [Remote host closed the connection]
beaver has joined #ffmpeg
bschindler has quit [Ping timeout: 250 seconds]
lec_thege80 has quit [Read error: Connection reset by peer]
lec_thege804 has joined #ffmpeg
Blacker47 has joined #ffmpeg
Dotz0cat has quit [Ping timeout: 256 seconds]
Dotz0cat has joined #ffmpeg
bitoff_ has quit [Ping timeout: 256 seconds]
lec_thege804 has quit [Ping timeout: 272 seconds]
lec_thege80 has joined #ffmpeg
lec_thege80 has quit [Read error: Connection reset by peer]
lec_thege804 has joined #ffmpeg
epony has quit [Remote host closed the connection]
fossdd has quit [Ping timeout: 272 seconds]
Tano has quit [Quit: WeeChat 4.1.2]
fossdd has joined #ffmpeg
navi has joined #ffmpeg
epony has joined #ffmpeg
lavaball has joined #ffmpeg
lucasta has joined #ffmpeg
MootPoot has quit [Quit: Connection closed for inactivity]
fossdd has quit [Ping timeout: 256 seconds]
Tinos has quit [Ping timeout: 250 seconds]
fossdd has joined #ffmpeg
bitbinge has quit [Ping timeout: 255 seconds]
bitoff has joined #ffmpeg
foul_owl has quit [Ping timeout: 246 seconds]
omegatron has joined #ffmpeg
foul_owl has joined #ffmpeg
Jhonny2x4 has joined #ffmpeg
vlm has joined #ffmpeg
TuxJobs has joined #ffmpeg
<TuxJobs>
I now have almost exclusively lossless FLAC files of the only music I want to listen to. But now I'm feeling nostalgic for how lossy MP3s sounded in the late 1990s and early 2000s. Since I haven't kept those ancient MP3s and they cannot be found online anymore, I'm trying to make copies of some of my FLACs as lossy MP3s to listen to sometimes. The command I use is `ffmpeg -i lossless.flac -b:a 32k lossy.mp3`, and it works, but I can't hear any of those
<TuxJobs>
characteristic "pops" that always were present at seemingly random locations in all those old MP3s. Does ffmpeg have a feature to add those in?
<TuxJobs>
I assumed that they were the result of the low bitrate chosen, but I also tried 16k and it sounds worse than I remember the old MP3s, yet zero "pops".
<kepstin>
"pops" were likely caused by either download errors introducting artifacts or poor quality mp3 encoders. ffmpeg will normally use lame by default, which is a lot better encoder now than most people used in the 90s)
bitbinge has joined #ffmpeg
<rodeo>
you can't even build current git master with a very old lame, requires 3.98.3
lexano has joined #ffmpeg
rsx has joined #ffmpeg
<rodeo>
n0.5.10 has no official minimum lame requirement although that doesn't necessarily means it will work with any version
<kepstin>
indeed; write some stuff to cd, throw it in a box of sand and shake it a bit, then rip it on a windows xp box with windows media player and the fraunhofer mp3 encoder, should sound suitably terrible :)
LionEagle has quit [Quit: Leaving]
<Marth64>
i dream of those days
<Marth64>
bad quality but good times
<furq>
you forgot that it should be a cd-r that was created from 128k mp3s in the first place
waleee has quit [Quit: WeeChat 4.1.2]
elastic_dog has quit [Quit: elastic_dog]
elastic_dog has joined #ffmpeg
minimal has joined #ffmpeg
ttys000 has joined #ffmpeg
bencc1 has quit [Ping timeout: 272 seconds]
alexherbo2 has joined #ffmpeg
YuGiOhJCJ has quit [Quit: YuGiOhJCJ]
MootPoot has joined #ffmpeg
fossdd has quit [Ping timeout: 255 seconds]
fossdd has joined #ffmpeg
Ingvix has quit [Ping timeout: 264 seconds]
Ingvix has joined #ffmpeg
Ogobaga has joined #ffmpeg
fossdd has quit [Ping timeout: 255 seconds]
Jhonny2x4 has quit [Quit: WeeChat 4.2.1]
Jhonny2x4 has joined #ffmpeg
fossdd has joined #ffmpeg
Jhonny2x4 has quit [Client Quit]
Jhonny2x4 has joined #ffmpeg
LionEagle has joined #ffmpeg
Jhonny2x4 has quit [Quit: WeeChat 4.2.1]
fossdd has quit [Ping timeout: 272 seconds]
fossdd has joined #ffmpeg
fossdd has quit [Ping timeout: 268 seconds]
theobjectivedad has quit [Ping timeout: 240 seconds]
fossdd has joined #ffmpeg
Ogobaga has quit [Ping timeout: 272 seconds]
Ogobaga has joined #ffmpeg
<TuxJobs>
Just to be clear, the "pops" I mentioned earlier seemed frequent and sounded more like some kind of sound effect from a NES game than any sound that a vinyl record would make while playing back a record.
<TuxJobs>
I wish I had an example.
<TuxJobs>
"Bad early MP3s" is apparently not nostalgic for too many more people besides myself.
theobjectivedad has joined #ffmpeg
<kepstin>
hmm, yeah, a recurring noise of that sort does seem like the sort of thing you'd get from a cd ripper using realtime or fast extraction and relying on the cd drive's builtin error hiding on a cheap drive, rather than anything specific to the mp3 encoding.
<kepstin>
there's a reason projects like cdparanoia and exact audio copy exist; to prevent that sort of thing :)
<bencoh>
so decode to pcm, replace random bits with 0s, and re-encode, basically
<kepstin>
if you want to get tricky about it, you'd look into the cd audio format and make sure to align your random chunks of 0s with the frame size.
<kepstin>
(and space them based on the typical length of music encoded in one revolution of a cd)
<kepstin>
(might also try a "hold"; take a sample and repeat the same sample value for a period of time, instead of replacing with 0s - that would change the character of the glitch a bit, and might be more representative of how some drives worked)
rvalue has quit [Ping timeout: 272 seconds]
<rodeo>
for all we know some of these came from the playback hardware and not the mp3 encode, I remember a long long time ago I bought a cheap CD player w/MP3 support and burned some CDs with multiple albums in MP3 format, and it sounded bad, but eventually I realised regular CDs sounded bad as well, with artifacts including cracks and pops; I returned it and exchanged for a high-end CD player also w/MP3 and the CDs I burned for the bad player
<rodeo>
sounded about just fine on the new one
rvalue has joined #ffmpeg
five618480 has quit [Remote host closed the connection]
five618480 has joined #ffmpeg
Ogobaga has quit [Ping timeout: 276 seconds]
theobjectivedad has quit [Remote host closed the connection]
theobjectivedad has joined #ffmpeg
JanC_ has joined #ffmpeg
JanC has quit [Killed (lead.libera.chat (Nickname regained by services))]
JanC_ is now known as JanC
rv1sr has joined #ffmpeg
mrelcee has quit [Quit: I want Waffles!]
mrelcee has joined #ffmpeg
theobjectivedad has quit [Remote host closed the connection]
theobjectivedad has joined #ffmpeg
bitbinge has joined #ffmpeg
fossdd has quit [Ping timeout: 276 seconds]
theobjectivedad has quit [Remote host closed the connection]
fossdd has joined #ffmpeg
theobjectivedad has joined #ffmpeg
<TuxJobs>
I'm pretty sure it was not the playback, though.
<TuxJobs>
I actually thought or assumed that this was a famous, known thing.
rvalue has quit [Ping timeout: 252 seconds]
<TuxJobs>
"Oh, yeah, those weird pops/blops in early MP3s! I remember those!"
<tykling>
the ntfs effect
<tykling>
(and fat32 before that)
rvalue has joined #ffmpeg
fossdd has quit [Ping timeout: 256 seconds]
<TuxJobs>
tykling: Hmm?
<tykling>
:) the blips and errors in my mp3s were from various crashes poorly handled by bad filesystems over the years, the problem kind of just stopped once the world moved to zfs and other checksummed and journalled filesystems
<tykling>
same reason we used to get warez in a bunch of rar files with checksums and now we dont
fossdd has joined #ffmpeg
FlorianBad has joined #ffmpeg
reynaldo has quit [Quit: Lost terminal]
fossdd has quit [Ping timeout: 256 seconds]
fossdd has joined #ffmpeg
Jhonny2x4 has joined #ffmpeg
fossdd has quit [Ping timeout: 268 seconds]
fossdd has joined #ffmpeg
alexherbo2 has quit [Remote host closed the connection]
alexherbo2 has joined #ffmpeg
Tano has joined #ffmpeg
fossdd has quit [Ping timeout: 272 seconds]
fossdd has joined #ffmpeg
wyatt8740 has joined #ffmpeg
wyatt8750 has quit [Ping timeout: 255 seconds]
theobjectivedad has quit [Remote host closed the connection]
theobjectivedad has joined #ffmpeg
lns has quit [Quit: leaving]
Haxxa has quit [Quit: Haxxa flies away.]
waleee has joined #ffmpeg
Haxxa has joined #ffmpeg
minimal has quit [Quit: Leaving]
Vonter has quit [Ping timeout: 272 seconds]
Vonter has joined #ffmpeg
navi has quit [Quit: WeeChat 4.1.2]
navi has joined #ffmpeg
navi has quit [Client Quit]
navi has joined #ffmpeg
catsoften has quit [Ping timeout: 264 seconds]
catsoften has joined #ffmpeg
theobjectivedad has quit [Read error: Connection reset by peer]
fossdd has quit [Ping timeout: 256 seconds]
theobjectivedad has joined #ffmpeg
fossdd has joined #ffmpeg
theobjectivedad has quit [Remote host closed the connection]
theobjectivedad has joined #ffmpeg
theobjectivedad has quit [Remote host closed the connection]
theobjectivedad has joined #ffmpeg
fossdd has quit [Ping timeout: 256 seconds]
fossdd has joined #ffmpeg
waleee has quit [Quit: updating system]
catsoften has quit [Read error: Connection reset by peer]
bitoff_ has joined #ffmpeg
waleee has joined #ffmpeg
catsoften has joined #ffmpeg
waleee has quit [Client Quit]
bitoff has quit [Ping timeout: 260 seconds]
waleee has joined #ffmpeg
vlm has quit [Quit: Leaving]
fossdd has quit [Ping timeout: 256 seconds]
fossdd has joined #ffmpeg
theobjectivedad has quit [Remote host closed the connection]
theobjectivedad has joined #ffmpeg
alexherbo2 has quit [Remote host closed the connection]
rv1sr has quit []
alexherbo2 has joined #ffmpeg
fossdd has quit [Ping timeout: 272 seconds]
five618480 has quit [Remote host closed the connection]
theobjectivedad has quit [Remote host closed the connection]
five618480 has joined #ffmpeg
theobjectivedad has joined #ffmpeg
theobjectivedad has quit [Remote host closed the connection]
theobjectivedad has joined #ffmpeg
manwithluck has joined #ffmpeg
Blacker47 has quit [Quit: Life is short. Get a V.90 modem fast!]
fossdd has joined #ffmpeg
theobjectivedad has quit [Remote host closed the connection]
theobjectivedad has joined #ffmpeg
theobjectivedad has quit [Remote host closed the connection]
bencc1 has joined #ffmpeg
<bencc1>
I'm trying to pause/resume screen capture with SIGSTOP/SIGCONT with continuous timestamps
<bencc1>
but the video has gaps every time the ffmpeg process is paused
<bencc1>
I've tried to add setpts and asetpts but it doesn't help
<TuxJobs>
tykling: "from various crashes" <-- On the computers making the MP3s?
theobjectivedad has joined #ffmpeg
<bencc1>
this is my command with setpts and asetpts:
alexherbo2 has quit [Remote host closed the connection]
alexherbo2 has joined #ffmpeg
<furq>
bencc1: did you check ffprobe -show_frames
fossdd has quit [Ping timeout: 272 seconds]
<furq>
TuxJobs: i guess he means bit flips etc on any computer between the ripper and the downloader
<bencc1>
furq: I'll check now
<furq>
including through bad piracy software or an overloaded switch at the lan party that was set up by someone's mate who reckons he knows what he's doing
<furq>
and that ends up ruining the quake tournament
manwithluck has quit [Quit: THE END]
<furq>
but what you described sounds like ancient mp3 encoder artifacts to me
manwithluck has joined #ffmpeg
<furq>
they were not very good in the 90s especially since they insisted on the lie that 128k cbr was CD QUALITY
fossdd has joined #ffmpeg
<BtbN>
I doubt anything in ffmpeg is prepared to sensibly handle SIGSTOP for a live capture
lolok has quit [Remote host closed the connection]
theobjectivedad has quit [Read error: Connection reset by peer]
theobjectivedad has joined #ffmpeg
<furq>
if it works at all i guess you'd need -fps_mode vfr
<bencc1>
BtbN: so how can I pause and resume x11grab and pulseaudio capture?
theobjectivedad has quit [Remote host closed the connection]
<BtbN>
stop the whole process, and start it again later
<bencc1>
furq: I'm trying to get a constant framerate
<BtbN>
and then stitch together the two videos
<TuxJobs>
furq: That just doesn't add up. You're talking about the 1990s... not some sort of ancient, alternate steampunk timeline with crude metal pipes and stuff...
<bencc1>
will I get constant audio and video framerate after concatenating the videos?
<furq>
you can't have cfr if there are seconds-long gaps though
<TuxJobs>
Things swoooshed through the network cables reliably on LANs.
<furq>
without dup frames
<bencc1>
furq: I thought I can rewrite the pts
<furq>
TuxJobs: yeah like i said i think it's probably artifacts
<TuxJobs>
furq: But you're implying that it's from unreliable network transfer and whatnot?
<furq>
it could be
<furq>
or just people having jank files and then sharing them
<bencc1>
and here the suggest to use setpts=N/FR/TB
<bencc1>
you are saying that it is not supposed to work?
theobjectivedad has joined #ffmpeg
<TuxJobs>
ChatGPT just answered: 'The "pops" or "blip" sounds that were often heard in MP3 songs in the late 1990s were generally a result of the encoding process used at that time. MP3 is a lossy audio compression format, which means it discards some audio data in order to reduce file size. This compression technique employs a "psychoacoustic model" to determine which sounds are less perceptible to human hearing and can be safely removed. During the encoding
<TuxJobs>
process, data related to certain sounds might be discarded or altered. This can sometimes create artifacts or glitches in the audio, leading to the presence of pops or blips in the resulting MP3 files. These artifacts are typically more noticeable when higher levels of compression are used or when the encoding algorithm is less sophisticated. Furthermore, in the late 1990s, when MP3 technology was still relatively new, encoding software and algorithms
<BtbN>
it's not anything anyone ever thought about when writing the code
<TuxJobs>
were not as advanced as they are today. This also contributed to the prevalence of those glitch sounds in MP3 files during that era.' But who knows if that's just made up nonsense like it often is?
<BtbN>
Can you not?
<furq>
that sounds like a much more verbose version of the guess that i made
<TuxJobs>
Damn FFMPEG for using such a sophisticated MP3 encoder!
<furq>
which is better than those things normally manage
<furq>
i guess you didn't ask it to do any maths
<TuxJobs>
furq: You were talking about network transfers corrupting the files, though? Not the encoding process?
<furq>
i said it was probably artifacts and earlier i mentioned using l3enc or mp3enc if you want that sound
<furq>
other people said it was bad cd rippers or old network transfers with no checksumming onto filesystems with no checksumming
<furq>
idk what you mean by pops so it could be all of the above
<TuxJobs>
I wish I could just physically travel back to that time and hear for myself, at my old computer... ;S
<furq>
there are still l3enc and mp3enc windows binaries online
<TuxJobs>
If I could find an example MP3 song with the glitch in it, I would've loved to link to it.
<furq>
as well as a million other awful mp3 encoders
<furq>
i think mp3enc is the correct one for the era though
<furq>
there was a windows acm component for it that was very commonly pirated back in the day
<FlorianBad>
If anyone is interested in very high quality test/encoder-challenging footage I gathered a few videos downloaded from Youtube here: https://florian.club/pub/video-test-footage/
mccobsta has joined #ffmpeg
fossdd has joined #ffmpeg
<another|>
"very high quality" "downloaded from Youtube"
<another|>
lol
theobjec- has quit [Remote host closed the connection]
theobjectivedad has joined #ffmpeg
hightower3 has joined #ffmpeg
theobjectivedad has quit [Remote host closed the connection]
Jhonny2x4 has quit [Read error: Connection reset by peer]
theobjectivedad has joined #ffmpeg
hightower2 has quit [Ping timeout: 272 seconds]
minimal has joined #ffmpeg
LionEagle has quit [Read error: Connection reset by peer]
<FlorianBad>
another|, well, yeah they're pretty damn good if your goal is to test codecs because it's well beyond the kind of loss you are trying to notice when encoding, especially since you'll obviously downscale that a bit
<FlorianBad>
Also you would't want a 2TB file for 2 minutes of test video
<another|>
not even 8k 420p would be 2TB
LionEagle has joined #ffmpeg
hightower4 has joined #ffmpeg
hightower3 has quit [Ping timeout: 246 seconds]
<BtbN>
Let me introduce you to the Phantom T4040
<BtbN>
"The industry’s fastest 4Mpx camera at over 9300 fps at 2560 x 1664 and up to 444K fps."
<another|>
okay, that's a super specific use case
<furq>
what an ugly sentence
<BtbN>
Yeah, it will not be 2560x1664 anymore at 444k fps
<bencc1>
I'm trying to use select between to pause/resume screen capture: