[Ffmpeg-devel-irc] ffmpeg.log.20121018

[burek]

[00:43] <arakn0> I'm trying to do point to point streaming using rtsp but not working yet. The command im using and its output: http://pastebin.com/2TUyMRq6  Any suggestion/example that can iluminate me? Thanks!
[00:46] <burek> arakn0, why not just use udp output directly if you need p2p
[00:47] <burek> also, your ffmpeg might be old
[00:47] <burek> try 1.0
[00:49] <arakn0> burek: it's one of my requirements. In my scenario, I have an embeded device that acts a rtsp client.
[00:50] <burek> req for what?
[00:50] <burek> oh rtsp
[00:50] <burek> ok
[00:50] <burek> can you update your ffmpeg
[00:51] <arakn0> sure
[00:51] <arakn0> I built ffmpeg on the 3rd of nov
[00:51] <arakn0> but i wonder if I'm in the right track
[00:51] <burek> it doesn't matter when did you build it, but what source code did you use
[00:55] <arakn0> ffmpeg version N-45010-g5e6439a
[00:55] <llogan> arakn0: you said you built ffmpeg but your pastes look like the repo version
[00:55] <burek> that one should be ok
[01:01] <arakn0> burek: weird. I just pulled master and rebuilding.
[01:03] <arakn0> just in case, but I'm not sure that my ffmpeg version is the problem
[01:04] <arakn0> more concern about the options
[01:04] <burek> we'll see :)
[01:04] <arakn0> that I'm using
[01:05] <arakn0> burek: just fyi, I've tried rtp over udp and tcp and both work with no issues
[01:19] <addisonj> hey, so I am having some issues trying to do some processing of multicast udp stream (ffprobe gives me this: mpeg2video (Main), yuv420p, 1920x1080 [PAR 1:1 DAR 16:9], 65000 kb/s, 30.22 fps, 29.97 tbr, 90k tbn, 59.94 tbc), but I am getting lots of artifacts and corruption, when I use vlc however the stream looks great
[01:33] <addisonj> burek: thanks for the tip, here is what I am trying to do: http://pastebin.com/tn1sH7Dj
[01:33] <arakn0> burek: same problem
[01:33] <burek> addisonj, you are not using ffmpeg
[01:34] <burek> join #ubuntu and ask them to explain to you what are you using actually
[01:34] <burek> or read this
[01:36] <addisonj> burek: oh sorry, just did this on a new machine, forgot about the fork, will switch and try again
[01:36] <burek> addisonj, try using ffmpeg's git
[01:41] <arakn0> the command and its output to start an rstp server: http://pastebin.com/zKuF8Kzq
[01:43] <burek> you are still using old ffmpeg :) FFmpeg version 0.6-4:0.6-2ubuntu6.3
[01:43] <burek> latest is like 1.0
[02:07] <arakn0> burek: i don't know what happened.
[02:07] <arakn0> here it's the right version: http://pastebin.com/b6LbnGkk
[02:08] <burek> [tcp @ 0x2664c40] TCP connection to 0.0.0.0:554 failed: Connection refused
[02:09] <arakn0> I've seen that. But I don't wan to connect to that IP. I want to bind to that IP and open that port
[02:09] <burek> ffmpeg is not an rtsp server
[02:09] <arakn0> So, I can stream point to point with rtp...but not rtsp...
[02:10] <burek> you can feed the rtsp server
[02:10] <burek> but if you are going to use p2p
[02:10] <burek> what's the point of rtsp then
[02:10] <burek> why not just send udp/rtp directly
[02:11] <arakn0> the requirements
[02:11] <arakn0> I know, I know...
[02:11] <burek> well ok
[02:11] <burek> then don't use ffmpeg
[02:11] <burek> use something else, vlc for example
[02:12] <burek> use something that can act as an rtsp server
[02:12] <burek> i think ffserver also can
[02:12] <burek> let me check
[02:12] <arakn0> yes, I was going to mention that
[02:12] <arakn0> ffserver
[02:12] <burek> http://ffmpeg.org/sample.html
[02:12] <burek> ctrl+f rtsp
[02:14] <arakn0> yes, I see that. I was going in that direction too. It wasn't 100% if I could stream using just ffmpeg
[02:14] <burek> http://ffmpeg.org/trac/ffmpeg/wiki/Streaming%20media%20with%20ffserver
[02:14] <burek> that might help too
[02:15] <burek> the only problem is I'm not sure how to specify -vcodec copy inside ffserver's config file
[02:16] <burek> but since you want to stream a file
[02:16] <burek> there is no need for that and you can safely ignore me :)
[02:16] <burek> just see the sample and rtsp examples there :)
[02:23] <arakn0> burek: thanks so much. I'm going to study it and I'll let you know!
[02:25] <burek> :beer: :)
[02:36] <addisonj> burek: our issues disappear with ffmpeg, avconv--
[02:36] <addisonj> thanks for the push in the right direction :)
[02:36] <burek> +1 :)
[03:53] <dericed> Hi all, for an unknown audio source is there a way to transcode it to an uncompressed output without possible loss in bit depth? For instance with video I can use -c:v rawvideo which will produce uncompressed video in native chroma subsampling and bit depth. There doesn't seem to be a rawaudio equivalent.
[05:03] <jivetalkingturke> hello i have a question, i need to input 3 videos crop and position into one ouput.. can this be done?
[05:03] <jivetalkingturke> like a tile effect
[05:04] <llogan> jivetalkingturke: sup cracka.
[05:04] <jivetalkingturke> i am capturing 3 webcams at 720p and would like to join them into one video output
[05:05] <llogan> it is possible. let me find an example.
[05:05] <jivetalkingturke> ok thanks
[05:06] <dericed> the side-by-side example in the overlay filter documentation is close https://ffmpeg.org/ffmpeg.html#overlay-1
[05:06] <jivetalkingturke> ok.. thanks i'll have a look
[05:06] <jivetalkingturke> overlay is the word i needed....
[05:08] <jivetalkingturke> thank you.. i'll have a look at overlaying and see what i can get...
[05:09] <llogan> that should be a good starting point. my example is ancient anyway
[05:10] <llogan> this binder has too many examples
[05:10] <jivetalkingturke> no worries thanks lllogan...
[05:10] <jivetalkingturke> and thank you dericed
[05:11] <dericed> ffmpeg -f lavfi -i testsrc -f lavfi -i testsrc -f lavfi -i testsrc -filter_complex "[0:0]pad=iw*2:ih*2[a];[1:0]negate[b];[2:0]vflip[c];[a][b]overlay=w[x];[x][c]overlay=w:h" -y  -t 1 out.avi
[05:11] <dericed> aw, he's leave the room once I finish an example
[05:12] Action: llogan gives dericed +10 FFmpeg Bucks anyway
[05:12] <dericed> sweet FFmpeg Bucks!! can i apply to a trac bounty?
[05:16] <llogan> dericed: consider adding your example here: https://ffmpeg.org/trac/ffmpeg/wiki/FilteringGuide
[05:17] <llogan> i see similar requests every once in a while
[05:19] <dericed> llogan: working on it
[05:21] <dericed> llogan: where do i upload thumbnails for wiki?
[05:22] <llogan> dericed: click on "Attach file" button on bottom of page, IIRC.
[05:23] <dericed> thx
[05:23] <llogan> ah...duh. i forgot there already is an example at FancyFilteringGuide
[05:23] <llogan> s/Guide/Examples
[05:24] <llogan> but a three or 4 input version would still be interesting
[05:24] <llogan> dericed: sorry about that
[05:25] <llogan> and that example still only have one input.
[05:31] <dericed> llogan: my image doesn't work but it is added to here: https://ffmpeg.org/trac/ffmpeg/wiki/FilteringGuide
[05:32] <llogan> dericed: thanks
[05:39] <llogan> dericed: i fixed the image. apparently you have to attach it to the specific page you are editing. i didn't know that until now, and if i did i forgot because my memory is => lesser ape.
[12:00] <zamba> hi there.. i want to create a timelapse of some web cam images
[12:01] <zamba> the images are sorted like: cam/<year>/<month>/<day>/img-<year><month><day>-<hour><minute>.jpg
[12:01] <zamba> and i want to pick one image each day
[12:02] <zamba> any smart way ffmpeg can do that?
[12:02] <Tjoppen> write a shell script to rename them to 1.jpg 2.jpg etc.
[12:02] <Tjoppen> that's about it
[12:03] <zamba> *shrug*
[12:03] <zamba> ffmpeg can't take its input from a text file?
[12:03] <zamba> so i can just put the absolute paths in there?
[12:04] <Tjoppen> ffmpeg can't take explicit names in image sequences, AFAIK
[12:04] <zamba> maybe look to mplayer, then
[12:05] <Tjoppen> no, just write a shell script to rename them
[12:05] <zamba> there's LOTS of images
[12:05] <Tjoppen> or, copy
[12:05] <Tjoppen> so? just throw disk and time at the problem. that or use an NLE like kdenlive
[12:05] <zamba> and i don't want to rename any images, that's irreversible, and i don't want to copy them either, because that's a huge overhead
[12:05] <zamba> i use mplayer instead
[12:05] <Tjoppen> ln -s then
[12:06] <zamba> ok, that's more interesting
[12:06] <zamba> ln is a more interesting approach, yeah
[12:06] <zamba> but the find job takes forever..... :p
[12:07] <zamba> 1440 images a day for 6 months :)
[12:08] <Tjoppen> somethin glike I=0; for f in `find .|grep jpg` ; do ln -s $f $I.jpg && I=`expr $I + 1` ; done
[12:08] <Tjoppen> then ffmpeg -i %d.jpg yadda yadda
[12:08] <zamba> yes :)
[12:09] <zamba> but they need to be sorted
[12:09] <Tjoppen> I hacked support for pipe separated image sequence names at work, but it's somewhat hackish
[12:09] <Tjoppen> just pipe the find output to sort then
[12:10] <Tjoppen> lunch
[12:18] <tapas> hi, i'm fiddling around with avconv and ffserver..
[12:19] <tapas> running the server works fine.. just avconv doesn't like to produce the feed:  http://pastesite.com/82142
[12:20] <tapas> i also tried just plain avconv -f video4linux2 -i /dev/video0 http://localhost:8090/feed1.ffm
[12:20] <tapas> same error
[12:20] <tapas> i wonder what avconv tries to tell me.. what codecs did it try? what paremeters did it try?
[12:20] <tapas> which ones failed? :D
[12:21] <tapas> works fine in vlc btw.. [just to note that the hardware does work]
[12:22] <tapas> and in mplayer..
[12:26] <tapas> hmm, i guess it's a broken package in ubuntu..
[12:44] <zamba> i tried building my time lapse now, but i'm only getting 9 seconds worth of video
[12:45] <zamba> my files are named 0001.jpg to 0584.jpg
[12:45] <zamba> and i'm running: ffmpeg -i %04d.jpg -r 30 -s 640x480 -vcodec libx264 -b 2000k cam-01.mp4
[12:46] <zamba> Input #0, image2, from '%04d.jpg':
[12:46] <zamba>   Duration: 00:00:23.36, start: 0.000000, bitrate: N/A
[12:46] <zamba>     Stream #0.0: Video: mjpeg, yuvj420p, 2560x1920 [PAR 1:1 DAR 4:3], 25 fps, 25 tbr, 25 tbn, 25 tbc
[12:46] <zamba> why does it say duration 00:00:23.36? how can it know that?
[12:48] <durandal_1707> from fps and number of images
[12:48] <zamba> but why is only a 9 second long video created?
[12:49] <zamba> ah, error in input
[12:49] <zamba> three broken images
[12:49] <zamba> let's try that again
[12:50] <zamba> argh.. the images have to be in order
[12:51] <zamba> and there's three jpegs that are broken
[12:51] <zamba> ffmpeg has no option to "skip" broken input?
[12:56] <zamba> just removed those images.. size 0 anyway
[12:56] <zamba> that was the problem
[13:51] <jivetalkingturke> hello... was wondering if someone could help me with my terminal command for ffmpeg..
[13:52] <jivetalkingturke> i am trying to run multiple jobs at the same time
[13:53] <jivetalkingturke> i don't want to use a bash script.. would rather do it all from terminal
[13:53] <jivetalkingturke> i am using 3 webcams 1 audio interface and 1 microphone
[13:54] <jivetalkingturke> would like to capture them at the same time
[14:23] <jivetalkingturke> anyone know about multiple simultaneous input?
[15:07] <t4nk124> hey guys, I'm updating my project to use the latest version of ffmpeg - it's working for single channel streams but anything stereo is coming out the other side very distorted - the playback rate is fine but the sound is like it's dropped 2 octaves - example: http://hosting.ispyconnect.com/example.mp3
[15:09] <durandal_1707> t4nk124: you have not support for planar sample formats
[15:09] <durandal_1707> so only mono will work
[15:09] <durandal_1707> almost every decoder is switching to that
[15:10] <durandal_1707> some, like flac and alac supports both
[15:11] <t4nk124> when i try and use AV_SAMPLE_FMT_S16 I get an error: Specified sample_fmt is not supported.
[15:11] <durandal_1707> as i said only some decoders support selecting sample fmt
[15:12] <t4nk124> ... but it works if it's single channel
[15:12] <t4nk124> and this used to work with an old version of ffmpeg
[15:12] <durandal_1707> t4nk124: as i said decoders switched to planar sample format
[15:13] <t4nk124> ah ok so basically what are my options?
[15:13] <durandal_1707> and interleaving is now handled in libswresample
[15:13] <durandal_1707> interleave output manually or use libswresample
[15:14] <durandal_1707> you will need to take care of channel layout - that it match with output device
[15:14] <durandal_1707> and libswresample AFAIK do not support that
[15:14] <durandal_1707> ^custom remapping
[15:15] <durandal_1707> so you are force to do it manually until this get cleaned up either in libswr or in lavc itself
[15:15] <durandal_1707> *forced
[15:16] <t4nk124> do you know of any examples? I'm way out of my depth :(
[15:16] <durandal_1707> but if you are interested in stereo this should not matter at all
[15:16] <t4nk124> ideally it would spit out a stereo mp3
[15:17] <durandal_1707> you meen encoding?
[15:18] <durandal_1707> *mean
[15:18] <t4nk124> yes
[15:19] <durandal_1707> you just need to deinterleave samples
[15:20] <t4nk124> ok sounds promising - how do i do that?
[15:20] <durandal_1707> but this is overkill considering that libmp3lame support interleaved samples, so fork just removed that code....
[15:21] <durandal_1707> actually not, it was never done that way, deinterleaving was done in encoder istself :roll
[15:22] <t4nk124> yeah this was all working fine in the previous version - now i can't encode anything stereo into mp3 using AV_SAMPLE_FMT_S16
[15:23] <durandal_1707> you need to give encoder each channel in separate plane
[15:24] <t4nk124> so i should be using AV_SAMPLE_FMT_S16P ?
[15:24] <durandal_1707> previously: Chan.1_sample.1|Chan.2_sample.1|Chan.1_sample.2|...
[15:26] <durandal_1707> now: Chan.1_sample.1|Chan.1_sample.2|Chan.1_sample.3 | ... | Chan.1_sample.N | Chan.2_sample.1 | ... | Chan.2_sample.N
[15:27] <durandal_1707> t4nk124: yes, but make sure you give deinterleaved samples
[15:30] <t4nk124> hmmm... all iv'e got coming into the method is a pointer to a byte array
[19:40] <Amnesia> hi question, I'm trying to decode my webcam's output to something compatible in multiple oses using a pipe
[19:40] <Amnesia> what type of codecs/contains would be advisable?
[19:40] <Amnesia> (it has to become a stream)
[19:47] <Endorgh> hi everyone! After 3 hours searching on internet, I haven't found out how I can preserve metadata after a conversion. I'm using version 0.7.13 under FreeBSD 9.0. Any suggestions?
[20:13] <creep> hi
[20:14] <creep> do somebody know how to set the headphone's left to audio track 1 right to track 2 converted to mono ?
[20:14] <creep> or maybe usgin gmplayer?
[20:45] <divVerent> creep: ffmpeg sure can do it using -filter_complex
[20:45] <divVerent> but I don't know exactly how
[20:45] <divVerent> look for "amerge" in the manpage, it does something like that
[20:55] <alezakos> Good morning/evening. Can I make ffmpeg produce TIFF images in the RGB colourspace?
[20:57] <durandal_1707> alezakos: several rgb colourpspaces are supported (except planar))
[21:01] <arakn0> I'm trying to stream an mp4 file over RTSP using just ffserver, but for some reason ffserver crashes. http://pastebin.com/pjGDzRH8  Any help/hint ???
[21:02] <alezakos> Good. As my experience on this field is zero, could you please provide me with help on how to split a video to tiffs readable by imagemagick?
[21:05] <durandal_1707> alezakos: ffmpeg -i input out%06d.tiff
[21:09] <alezakos> Unfortunately, that command produces tiffs in the YCbCr colourspace, and imagemagick doesn't support it (the colours are weird)
[21:10] <klaxa> arakn0: not sure if this is still up to date, but the documentation says streaming from files is broken: http://ffmpeg.org/ffserver.html#What-can-this-do_003f
[21:11] <klaxa> alternatively you could try using ffmpeg as a feed
[21:11] <klaxa> i think? that kinda worked for me
[21:12] <durandal_1707> alezakos: your input is than not in rgb colorspace which means you need to specify it via -pix_fmt rgba
[21:14] <Endorgh> Hi folks! I've ffmpeg compiled with 'libaacplus' on FreeBSD, and it doesn't keep into output file the metadata of input file. I've tried all -map options but nothing works. Any ideas?
[21:14] <alezakos> durandal_1707: That works perfectly! Thank you very much!
[21:14] <arakn0> klaxa: damn.... but it works for mpg :(
[21:15] <durandal_1707> Endorgh: what container?
[21:15] <klaxa> arakn0: like i said, no idea if that's still up to date, i have given up on setting up ffserver myself, because it lacks some functionality i want (streaming matroska with subtitles)
[21:17] <Endorgh> durandal_1707: flac to aac
[21:20] <Endorgh> durandal_1707: exactly this command:     ffmpeg -i file_in.flac -ab 60k file_out.aac
[21:21] <Endorgh> durandal_1707: if I type     ffmpeg -i file_in.flac -ab 60k file_out.mp4     the file_out.mp4 has the medatada, but its length is only 2 minutes, not the same length of input file... Is very strange
[21:21] <JEEB> uhh, raw AAC doesn't have anything really to put metadata into
[21:21] <JEEB> a container like "MP4" is needed to accomodate that
[21:24] <Endorgh> JEEB: hence, is needed to specify ".mp4" extension
[21:24] <Endorgh> ?
[21:24] <durandal_1707> Endorgh: flac to mp4 with libaacplus works fine here
[21:25] <JEEB> Endorgh, or anything else that has possible spots for the tags you need, yes
[21:25] <Endorgh> durandal_1707: do you recommend libaacplus in place of FAAC?
[21:25] <Endorgh> ok I understand...
[21:25] <JEEB> no, you should use fdk
[21:25] <JEEB> instead of both faac and libaacplus
[21:25] <JEEB> handles both HE-AAC and LC-AAC
[21:25] <JEEB> https://github.com/mstorsjo/fdk-aac
[21:26] <Endorgh> ohh, great info!
[21:26] <JEEB> you can't give out binaries with it, just like faac and libaacplus, but it's the least bad aac encoder atm :P
[21:27] <durandal_1707> JEEB: you did blind test?
[21:28] <JEEB> not me, but in general that has been the concensus. It's the fraunhofer encoder and it definitely is better than faac and vo-aacenc with LC
[21:28] <Endorgh> JEEB: hehe, the least bad aac encoder. Then, what is the best one?
[21:28] <divVerent> durandal_1707: really need no blidn test to compare with faac
[21:29] <divVerent> given I DO hear difference between 128kbit/s faac and original
[21:29] <divVerent> faac really IS that bad
[21:29] <JEEB> and it doesn't sound bad with HE-AAC so it's at least on par with libaacplus
[21:29] <JEEB> yeah, faac is pretty bad
[21:29] <JEEB> and vo-aacenc is at times even worse than the ffaac
[21:29] <divVerent> blind test may be needed for 192kbit/s faac :P
[21:29] <JEEB> (internal aac encoder)
[21:29] <cbsrobot> JEEB: sorry for the late replay, but about the libaac_fdk delay issue
[21:29] <divVerent> and the ffmpeg builtin "aac" encoder is buggy
[21:29] <divVerent> it does clipping ;)
[21:29] <divVerent> or a noise similar to that
[21:29] <JEEB> well, every encoder does that
[21:29] <cbsrobot> I was not able to get it to work
[21:29] <JEEB> oh
[21:30] <cbsrobot> I thought it is a samplerate issue
[21:30] <divVerent> I mean, apparently the "aac" encoder in ffmpeg has some overflow issue somewhere
[21:30] <divVerent> it really is only good if "there is nothing else available"
[21:30] <cbsrobot> but it does not seem so
[21:30] <divVerent> and if you really NEED aac
[21:30] <JEEB> divVerent, more like the ffaac encoder was left to be after some development :/
[21:30] <divVerent> the aac codec needs -strict -2 for a reason :P
[21:30] <JEEB> at least it's better than the WMA encoder
[21:30] <JEEB> lol
[21:31] <divVerent> jeeb:didn't test THAT one
[21:31] <JEEB> Endorgh, not sure -- Fraunhofer or Apple's (Dolby's?)
[21:31] <divVerent> but I actually did use the aac encoder for a while
[21:31] <divVerent> and so did I use faac
[21:31] <divVerent> for libfdk_aac I still have to look for the artifacts ;)
[21:31] <JEEB> well, the internal aac encoder is the best that you can still distro
[21:31] <divVerent> even though I changed bitrate from 128 to 96 when changing from faac to libfdk_aac
[21:31] <JEEB> vo_aacenc according to a blind test is not really better
[21:32] <divVerent> jeeb: that I don't believe
[21:32] <divVerent> unless vo_aacenc has similar issues :P
[21:32] <divVerent> but well
[21:32] <divVerent> the average case of the "aac" codec is fine
[21:32] <JEEB> it's just the reference code >_>
[21:32] <divVerent> it just has a bad worst case
[21:32] <divVerent> if you want, I can switch to using it again
[21:32] <JEEB> (and yes, I know that fraunhofer and friends base on the reference code, too)
[21:32] <divVerent> and wait till I get a good sample to demonstrate the "clipping" issue
[21:32] <JEEB> but at least fraunhofer generally has been optimized
[21:32] <divVerent> when I did encounter one in the past, I verified the weird noise goes away when reducing volume before encoding
[21:33] <cbsrobot> JEEB: Do you know how to prevent the .5 sec delay in libfdk_aac ?
[21:33] <JEEB> cbsrobot, if it's the priming samples you need to signal them in the container and make sure the thing that's playing it back actually understands the way of setting priming samples
[21:34] <JEEB> I think ffmpeg should support that (maybe even by default?), and L-SMASH's "MP4" muxer does it too
[21:34] <cbsrobot> aac in mp4, played back with quicktime
[21:34] <JEEB> basically you should check the first PTS of the first audio frame/packet, you should have the amount of priming needed there
[21:34] <JEEB> (there's a negative pts IIRC)
[21:35] <cbsrobot> how do I check that ?
[21:35] <cbsrobot> hexedit ?
[21:35] <JEEB> no idea, people usually use the API to get that figure :P
[21:35] <cbsrobot> or ffprobe ?
[21:36] <JEEB> also check if the audio length is the same in ffmpeg before and after
[21:36] <JEEB> when you do f.ex. ffmpeg -i derp.mp4 and have it show stuff
[21:36] <JEEB> (ffprobe probably shows too)
[21:37] <cbsrobot> Ok thanks
[21:38] <cbsrobot> I'll try and ping you later (maybe)
[21:38] <JEEB> basically if there's a length difference then priming samples aren't for whatever reason pushed there :P
[21:38] <JEEB> (output should be somewhat longer)
[21:39] <JEEB> if that is the case, you'd have to get the amount of priming needed and re-mux it with L-SMASH's muxer or something from raw AAC and setting the amount of priming samples
[21:41] <cbsrobot> basicly it a mov with prores and pcm sound convertet to h264 and aac in mp4
[21:48] <Endorgh> ffmpeg -i input_file.flac -ab 60k output_file.mp4, generates an output file with less time length than the original input file. Why?? O_O
[21:49] <Endorgh> it results incomprehensible for me...
[21:53] <durandal_1707> Endorgh: i said i can not reproduce it, you are using buggy version
[21:55] <creep> nah i set software mixer on gmplayer and opened up same video 2 times, synced them, and set one language with balance left on one, and the other language on the other gmplayer on the right, i can watch CSI in both language now
[21:56] <Endorgh> durandal_1707: ok, understood!
[21:58] <durandal_1707> Endorgh: what version you are using?
[22:00] <Endorgh> ffmpeg-0.7.13_6,1
[22:01] <JEEB> uhh, does that even have libaacplus >_>
[22:01] <JEEB> also, that is /quite/ old
[22:02] <Endorgh> its the default version on FreeBSD port collection
[22:02] <JEEB> you should build it yourself then, current git if possible
[22:03] <JEEB> you are not going to get much support or help with as old version as that.
[22:03] <JEEB> as sad as it is
[22:03] <durandal_1707> Endorgh: there is ffmpeg-011 and ffmpeg-devel
[22:03] <durandal_1707> and ffmpeg1 (1.0)
[22:06] <Endorgh> durandal_1707: mmm d'oh!
[22:37] <dominikd> is there a "just work" setting for ffmpeg?
[22:38] <dominikd> as in eat as many stream encodings as possible
[22:38] <dominikd> without throwing errors
[23:08] <Amnesia> question, what are the required arguments for webm?
[23:15] <wlan2> required? use extension .webm (?)
[23:15] <wlan2> And then it automatically chooses vp8 and vorbis
[23:15] <Amnesia> hm don't know what parameters to give with webm
[23:16] <Amnesia> Encoder (codec none) not found for output stream #0:0
[23:16] <Amnesia> and
[23:16] <Amnesia> Could not write header for output file #0 (incorrect codec parameters ?)
[23:18] <Amnesia> ffmpeg -f video4linux2  -i /dev/video0 -vcodec mpeg4 -f webm -y /dev/stdout | nc -lp 5555
[23:21] <dirty_d> Stream #0:2(eng): Subtitle: subrip
[23:21] <dirty_d> is this the same as SRT fromat?
[23:25] <iive> yes, srt is subrip
[23:25] <Amnesia> iive: how can I determine what arguments are required for a specific container/codec ?
[23:26] <iive> no idea. sorry.
[23:26] <iive> btw, if you try to stream it through a pipe, nut may be the container.
[23:27] <creep> ok mplayer hung again, shit "MPlayer interrupted by signal 2 in module: unknown" would someone tell mplayer devs that module unknown has  bugs?
[23:27] <Amnesia> hm ok thx
[23:27] <Amnesia> I seriously suck at ffmpeg-.-
[23:27] <iive> flv may be a little slow, there is v3 in development, no idea how to select it.
[23:27] <iive> creep: LoL!
[23:28] <iive> what was signal2? bus error or ctrl+c?
[23:29] <Amnesia> iive: looks like asf is my best bet so far:)
[23:29] <Amnesia> thanks
[23:30] <creep> iive<< controlc did the job now
[23:31] <creep> there are really lame errors like double pressing play, or pressing controls while playlist is open resulting infinite loops, or setting window to always on top and hitting some buttons on controls
[23:56] <iive> creep: interesting, i assume these are gmplayer related. could you try to fill some of them in the bugzilla, in case they are not already there.
[00:00] --- Fri Oct 19 2012