Ffmpeg-devel-irc
Threads by month
- ----- 2026 -----
- June
- May
- April
- March
- February
- January
- ----- 2025 -----
- December
- November
- October
- September
- August
- July
- June
- May
- April
- March
- February
- January
- ----- 2024 -----
- December
- November
- October
- September
- August
- July
- June
- May
- April
- March
- February
- January
- ----- 2023 -----
- December
- November
- October
- September
- August
- July
- June
- May
- April
- March
- February
- January
- ----- 2022 -----
- December
- November
- October
- September
- August
- July
- June
- May
- April
- March
- February
- January
- ----- 2021 -----
- December
- November
- October
- September
- August
- July
- June
- May
- April
- March
- February
- January
- ----- 2020 -----
- December
- November
- October
- September
- August
- July
- June
- May
- April
- March
- February
- January
- ----- 2019 -----
- December
- November
- October
- September
- August
- July
- June
- May
- April
- March
- February
- January
- ----- 2018 -----
- December
- November
- October
- September
- August
- July
- June
- May
- April
- March
- February
- January
- ----- 2017 -----
- December
- November
- October
- September
- August
- July
- June
- May
- April
- March
- February
- January
- ----- 2016 -----
- December
- November
- October
- September
- August
- July
- June
- May
- April
- March
- February
- January
- ----- 2015 -----
- December
- November
- October
- September
- August
- July
- June
- May
- April
- March
- February
- January
- ----- 2014 -----
- December
- November
- October
- September
- August
- July
- June
- May
- April
- March
- February
- January
- ----- 2013 -----
- December
- November
- October
- September
- August
- July
- June
- May
- April
- March
- February
- January
- ----- 2012 -----
- December
- November
- October
- September
- August
- July
- June
- May
- April
- March
- February
- January
- ----- 2011 -----
- December
- November
- October
- September
- August
- July
- June
- May
- April
- March
- February
- January
- ----- 2010 -----
- December
- November
- October
- September
- August
- July
- June
- May
- April
- March
- February
- January
June 2014
- 1 participants
- 64 discussions
[00:21] <J_Darnley> Oh my god! gcc doesn't warn you if you stupidly do this:
[00:21] <J_Darnley> for (int i; ...; ...;)
[00:30] <nevcairiel> What happens if you do this?
[00:34] <BtbN> random stuff. But it should warn about the uninitialized var
[00:46] <J_Darnley> nevcairiel: BtbN is right. Without initialising i can be any value
[00:47] <J_Darnley> I happened to get a large negative number
[00:47] <J_Darnley> which wasn't very useful for use as an index in an array.
[00:47] <J_Darnley> It caused a segfault but it was fairly easy to trace back in this case.
[00:48] <J_Darnley> Perhaps this is an argument to not use this C99 convenience
[00:49] <BtbN> if it realy does not generate a warning, you should propably report a bug to gcc
[00:50] <J_Darnley> ugh. I guess I should
[00:50] <J_Darnley> First I better confirm it with a test case.
[00:51] <nevcairiel> Did you enable the warnings for uninitialized stuff? Not sure which level includes those
[00:52] <J_Darnley> Wall and Wextra
[00:54] <J_Darnley> I will check that before I attempt to report
[00:54] <nevcairiel> For kicks try again with Wuninitialized, no idea if its included
[00:55] <J_Darnley> hmm doesn't look like it.
[01:01] <cone-53> ffmpeg.git 03Michael Niedermayer 07master:525a165d9985: avfilter/vf_pullup: workaround bug in gcc 4.4.3 on ARM
[01:08] <J_Darnley> Oh it already exists: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=45467
[01:09] <J_Darnley> With a dubious closed status
[01:13] <J_Darnley> I'll post it again because it didn't warn me in a very real case.
[01:30] <BBB> jamrial: if you care, new patch on github to convert int16 functions also (needs testing on win64, and some benching)
[01:30] <BBB> jamrial: Im probably gonna leave it as-is (maybe convert double) and focus on more useful things again, so if you want to do fma4/xop, thatd be cool
[01:30] <BBB> (over-optimizing a function like this is kinda silly)
[01:32] <J_Darnley> Actually I'm not going to post. It looks like this might be the 7th most reported bug.
[01:37] <jamrial> yeah, i'll do fma/xop
[01:38] <jamrial> btw, benched float on win64 this morning and got 65->63 common, 110->92 linear
[01:39] <jamrial> maybe J_Darnley can bench as well to double check, because i found it unusual. quite the boost for linear and a really small change for common
[01:47] <BBB> gcc did a really good job on common
[01:48] <BBB> like, a really good job
[01:48] <BBB> so Im not totally surprised it only goes down by a bit
[01:48] <Daemon404> obligatory "which gcc"
[01:48] <BBB> I was surprised how good code it generated
[01:48] <BBB> that too
[01:50] <jamrial> btw, you should add your name to resample.asm. probably replacing mine for that matter
[01:51] <jamrial> you pretty much rewrote the entire thing after all
[01:51] <BBB> jamrial: .. maybe - wanna convert the double sse2 function for me?
[01:51] <BBB> Im feeling lazy
[01:51] <BBB> kids arent helping productivity much
[01:53] <BBB> and updated patch on github to add me as co-author of that file
[01:53] <jamrial> i'll give it a try
[01:54] <BBB> shouldnt be terribly hard, but who knows
[01:55] <BBB> hopefully the float can serve as most-of-the-template (and multi-format support already exists with int16)
[02:23] <J_Darnley> BBB jamrial: can I bench it tomorrow or would you like me to do it now?
[02:24] <jamrial> do it whenever you can, no hurry
[02:24] <jamrial> it's just to cross check results. the code is already commited and confirmed faster
[02:34] <BBB> J_Darnley: tomorrow is fine, no hurry
[02:57] <jamrial> BBB: win64 and 32bits int16 (sse2 and mmxext) all work
[03:00] <cone-358> ffmpeg.git 03Michael Niedermayer 07master:f02a4408149c: ffmpeg_opt: route -ab to -b:a
[03:25] <jamrial> BBB: got double working. on top of the int16 patch
[03:26] <cone-358> ffmpeg.git 03Michael Niedermayer 07release/0.10:6a968073daa7: avutil/lzo: Fix integer overflow
[03:26] <cone-358> ffmpeg.git 03Luca Barbato 07release/0.10:e7f5dacd55de: lzo: Handle integer overflow
[03:26] <cone-358> ffmpeg.git 03Reinhard Tartler 07release/0.10:359383c98308: Prepare for 0.8.13 Release
[03:26] <cone-358> ffmpeg.git 03Reinhard Tartler 07release/0.10:e122fb594a5f: Update Changelog for 0.8.13
[03:26] <cone-358> ffmpeg.git 03Michael Niedermayer 07release/0.10:d77ad6ec2d14: Merge commit 'e7f5dacd55deeee8a866020b8463f829b2c5971f' into release/0.10
[03:26] <cone-358> ffmpeg.git 03Michael Niedermayer 07release/0.10:a907fc0b21fa: Merge commit 'e122fb594a5feb6729cce86a70aafd93d10202d8' into release/0.10
[03:31] <cone-358> ffmpeg.git 03Michael Niedermayer 07release/0.10:9153b33a742c: update for FFmpeg 0.10.14
[03:33] <Timothy_Gu> michaelni: tickets 3748-3751 are spams. Can you add the IP of the offender to BadIP?
[03:41] <michaelni> Timothy_Gu, added
[03:43] <Timothy_Gu> BTW, i WHOIS'd the IP http://myip.ms/view/ip_addresses/3423563264/204.15.110.0_204.15.110.255
[03:43] <Timothy_Gu> And the ISP seems to be called "Hoyos Consulting"
[03:52] <Timothy_Gu> michaelni: another one
[04:09] <cone-358> ffmpeg.git 03Michael Niedermayer 07fatal: ambiguous argument 'refs/tags/n0.10.14': unknown revision or path not in the working tree.
[04:09] <cone-358> Use '--' to separate paths from revisions
[04:09] <cone-358> refs/tags/n0.10.14:HEAD: ffmpeg_opt: route -ab to -b:a
[10:35] <iamtakin1iteasy> hello, i am trying to generify FreeType2/fontconfig handling in ffmpeg, since after introduction of aalib video filter, this part starts to be common with drawtext video filter, but i am not sure what to do with public API and ABI. i think better to explain this starting from the public header of freetypeutils component of libavfilter: http://paste.eientei.org/show/238/ -- here i am trying to
[10:35] <iamtakin1iteasy> offer a generic interface to both freetype and fontconfig at once, but each of them may be unavailable at compilation time, depending on ./configure options, so if in my structure FFFreetypeContext i can't simply use FT_Library and FcConfig structures, but rather have to either guard them with #ifdef <lib> (which will be changing structure size) or using void* pointers (which is not type-safe),
[10:36] <iamtakin1iteasy> so my question would be: what is preferable - type-safety or constant struct size? or maybe sacrifice clarity of code and use #ifdef <lib> FT_Library *ft_library #else void * ft_library #endif construction?
[10:39] <ubitux> while the fontconfig can be considered optional, is it really the case for FT?
[10:40] <ubitux> like, let's take drawtext for instance: it has a hard dep on FT
[10:40] <ubitux> so you're wrapper won't be necessary if FT is not available
[10:41] <ubitux> it looks like to me the question only stands for FC
[10:41] <iamtakin1iteasy> vf_drawtext is hard dependent on FT, but as far as i understood vf_drawtext itself is optional depending on FT selection
[10:42] <iamtakin1iteasy> but this is video filter
[10:42] <ubitux> that doesn't change the fact that this wrapper -in the case of drawtext at least- won't be build unless FT is available
[10:43] <ubitux> what is this aalib filter you're talking about?
[10:44] <wm4> meh
[10:44] <iamtakin1iteasy> hm, got it, it is more a matter of buildsystem rather than code.
[10:44] <wm4> why not just use libass
[10:45] <iamtakin1iteasy> aalib filter didn't passed review yet
[10:45] <ubitux> ah it was submitted?
[10:45] <wm4> also, for aalib the hardcoded tty font would be better
[10:45] <ubitux> missed that
[10:45] <wm4> probably
[10:45] <iamtakin1iteasy> basically, it uses aalib to render frames as ascii-art and rastarizes them back with FT
[10:45] <ubitux> http://ffmpeg.org/pipermail/ffmpeg-devel/2014-June/159169.html ah, mmh ok.
[10:46] <iamtakin1iteasy> current version on review is outdated a bit, so don't look hard on it
[10:46] <iamtakin1iteasy> i have santized it's code a lot during this time
[10:46] <wm4> iamtakin1iteasy: see libavutil/xga_font_data.h and libavcodec/ansi.c
[10:47] <ubitux> yeah that sounds like you could use some vga fixed width char
[10:47] <ubitux> and you wouldn't need FT/FC
[10:48] <iamtakin1iteasy> but it is a point of aalib filter i want :/ i wish to select font sizes and faces to fine-tune the ascii-art effect
[10:48] <iamtakin1iteasy> for instance, Terminus is not anything like Liberation Mono
[10:49] <ubitux> ok
[10:50] <iamtakin1iteasy> but still, FC is being optional and pretty nice feature to have
[10:50] <iamtakin1iteasy> should i have it's handle as void* or guareded with #if LIBFONTCONFIG ?
[10:50] <wm4> never use void* for opaque things
[10:50] <ubitux> wm4: why?
[10:51] <wm4> ubitux: because it sucks
[10:51] <iamtakin1iteasy> well, C sucks at type-safety when it comes to any type of pointer anyways :/
[10:51] <wm4> opaque structs are better because they give you a minimum of type-safety
[10:51] <wm4> C doesn't suck, people who think they know C but don't suck
[10:52] <wm4> and that's why ever shitty API does typedef void *something_t;
[10:52] <iamtakin1iteasy> well, FcInitLoadConfigAndFonts() returns pointer anyways
[10:53] <iamtakin1iteasy> so even in FC API it would be FcConfig*
[10:53] <ubitux> #if CONFIG_LIBFONTCONFIG #include ... #define FCTYPE FcConfig #else #define FCTYPE void #endif
[10:53] <ubitux> :-°
[10:54] <ubitux> but well, you could just add some idefery inside the struct
[10:55] <wm4> iamtakin1iteasy: you could provide something to generate and load bitmap fonts in a custom format
[10:55] <wm4> still sounds better than letting fontconfig/freetype use spread
[10:56] <iamtakin1iteasy> well, maybe it is for a developer, but isn't a bit too much inconvinient for end-user?
[10:56] <wm4> what enduser wants an aa filter
[10:56] <iamtakin1iteasy> using FT/FC you can easily tweak sizes and faces
[10:56] <iamtakin1iteasy> me!
[10:56] <iamtakin1iteasy> :)
[10:56] <wm4> a single person?
[10:57] <iamtakin1iteasy> well, i hope not only me
[10:57] <iamtakin1iteasy> but i wish to stream films/animation in ascii-art
[10:57] <wm4> also there are already things like libcaca which do this
[10:57] <iamtakin1iteasy> no!
[10:57] <wm4> I think aalib also has a X backend
[10:57] <iamtakin1iteasy> libcaca is different thing
[10:57] <iamtakin1iteasy> it doesn't produces neat aa-like output
[10:57] <wm4> and oh joy, libavdevice indeed has a caca output
[10:58] <iamtakin1iteasy> but it's output, which simply displays the result on a screen
[10:58] <iamtakin1iteasy> not rastarizing it back to frames
[10:58] <wm4> so?
[10:58] <ubitux> wm4: the point is to have a customed ascii-like rendered video
[10:59] <wm4> so useful
[10:59] <ubitux> so what
[10:59] <iamtakin1iteasy> it deosn't fit my use-case. i want to take a regular movie, process it with aafilter and stream it over RTMP
[10:59] <iamtakin1iteasy> preferably in real-time
[10:59] <wm4> are you running some sort of joke video streaming site?
[10:59] <iamtakin1iteasy> yes
[11:00] <iamtakin1iteasy> and there are titles that *must* be watched in aa
[11:00] <iamtakin1iteasy> SE:Lain for instance
[11:02] <wm4> anyway, fontconfig is a big problem and is foreign to all major platforms (except Linux, if you want to call that major)
[11:05] <ubitux> wm4: do you think !linux user will care about aa output? ;)
[11:05] <wm4> I think nobody will care about it
[11:05] <ubitux> (do you have an alternative?)
[11:06] <wm4> alternative to what?
[11:06] <ubitux> FC
[11:06] <wm4> if there was, libass would already use it
[11:06] <ubitux> then there is no reason not to use it
[11:07] <wm4> libass will probably be free of FC on windows and osx in the near future
[11:07] <wm4> with great effort
[11:07] <iamtakin1iteasy> isn't OSX has it's version of FC?
[11:08] <wm4> all platforms have their own font lookup APIs
[11:08] <ubitux> wm4: replaced with what?
[11:08] <iamtakin1iteasy> looks like, if not using FC, FFmpeg will have to invent it's own :/
[11:08] <wm4> ubitux: coretext on osc
[11:09] <wm4> *osx
[11:09] <ubitux> why isn't that part of FT btw?
[11:09] <wm4> outside of the scope of FT
[11:10] <ubitux> and FC is not able to use coretext on OSX?
[11:10] <wm4> apparently not
[11:11] <wm4> they don't care about other platforms anyway
[12:00] <wm4> DVD is so fucked up
[12:01] <wm4> I wonder whether it would make sense to provide a libavformat dvd demuxer
[12:01] <wm4> but since dvd requires dealing with menus and user interaction, probably not
[12:05] <ubitux> same as bluray
[12:05] <ubitux> and it would definitely help
[12:08] <wm4> both dvd and bluray allow title switching
[12:08] <wm4> which doesn't fit into the lavf api at all
[12:08] <wm4> since audio/subtitle streams and stream languages will change
[12:09] <wm4> changing the title is basically like loading a different file
[12:09] <wm4> and what would you do about user interaction (menus)?
[12:09] <ubitux> i'd say it can be ignored
[12:10] <wm4> ignored?
[12:10] <ubitux> we have support for bluray, and it's more than useful
[12:10] <ubitux> yes, the dvdnav
[12:10] <wm4> haha
[12:10] <wm4> no
[12:10] <wm4> I don't want a shitty half-assed solution (I already have that in my own code)
[12:11] <ubitux> well, in the ffmpeg scope it has its usage
[12:11] <wm4> and excluding menus (btw. bluray also has menus) belongs to that
[12:11] <ubitux> (i don't need menu to transcode a bd)
[12:11] <wm4> yeah right I forgot lav* is for transcoding only
[12:11] <ubitux> that's not what i'm saying
[12:12] <ubitux> i'm saying even with no menu it covers a large use-case
[12:13] <ubitux> now if you want interaction, that probably requires a different lavd
[12:13] <wm4> "different lavd" ??
[12:14] <ubitux> improved* lavd
[12:14] <ubitux> different api
[12:15] <wm4> that doesn't change anything about the title switching problems
[12:15] <ubitux> what's this title switching thing?
[12:15] <wm4> I explaine dit above
[12:15] <ubitux> well
[12:15] <ubitux> lavf could raise packets of every stream/file
[12:16] <ubitux> in parallel
[12:16] <wm4> not sure I understand
[12:16] <ubitux> not very optimal but it would fit in the current design
[12:16] <ubitux> wm4: every title would be opened
[12:17] <wm4> the API doesn't allow that
[12:17] <wm4> libdvdnav API I mean
[12:18] <ubitux> should we use an AVOption hack just like the ogg thing to control the demuxing?
[12:18] <wm4> what ogg thing?
[12:19] <ubitux> the tags
[12:19] <ubitux> already forgot what it was doing actually
[12:19] <wm4> updating tags via side-data
[12:20] <ubitux> ok
[13:30] <cone-161> ffmpeg.git 03Clément BSsch 07master:b18d1b094e90: avfilter/buffersrc: reindent
[13:30] <cone-161> ffmpeg.git 03Clément BSsch 07master:d5f817793e29: avfilter/overlay: reindent
[13:35] <BBB> jamrial: cool, thanks, I guess I can send the int16 one out
[13:35] <BBB> michaelni: btw, about swr int16/32, Im pretty 100% sure that the int32 codepath routinely overflows? if input is 32 bits and filter is 30 and src_incr is another up to 20 or so, that never fits in 64 bit
[13:36] <BBB> michaelni: I mean this line: val += (v2 - val) * (FELEML) frac / c->src_incr;
[13:37] <BBB> jamrial: or you can send the int16 one out as part of your patchset if you like
[13:48] <J_Darnley> Oh yeah! (As the koolaid guy might say) I've got the superscope working.
[14:24] <michaelni> BBB, fixed, ill post a patch soon
[14:29] <Daemon404> [11:01] <+wm4> I wonder whether it would make sense to provide a libavformat dvd demuxer <-- pls no
[14:30] <wm4> Daemon404: what would you have against it?
[14:30] <Daemon404> "demuxer"
[14:30] <Daemon404> menus and stuff
[14:30] <Daemon404> etc
[14:30] <wm4> I know
[14:31] <Daemon404> im sure j-b has strong feelings here ;)
[14:31] <wm4> IMO, actually libdvdnav should contain a mpeg demuxer
[14:32] <Daemon404> as in ps?
[14:32] <wm4> yes
[14:32] <Daemon404> wouldnt be a terrible idea
[14:33] <Daemon404> but nobody is going to do that
[14:33] Action: Daemon404 looks at libdvdnav;s code
[14:33] <wm4> yeah, parts of it look like copy&pasted mplayer code
[14:33] <BBB> michaelni: itd be cool if the division in that statement became a multiplication, esp. for bitexact int math matching (for float, I already changed it to a multiply)
[14:35] <BBB> michaelni: since div requires me to go outside simd which is kinda silly
[14:35] <BBB> (for int)
[14:37] <Daemon404> wasnt there some big text file of equivalent mul/shr for divs
[14:40] <Daemon404> heh... facebook ads are showing me HEVC books
[14:40] <Daemon404> ~stalking~
[14:56] <michaelni> BBB, patches posted, just saw your comment now, feel free to change the int C side to what is matches and is convenient for SIMD
[14:57] <michaelni> what matches
[14:59] <BBB> ok
[14:59] <BBB> can I also change src_incr to be max 16 bits for int16?
[14:59] <BBB> that makes a lot of things simpler
[14:59] <BBB> 15bits is even nicer (pmulhrsw)
[15:01] <michaelni> for the linear interpol, 15bits should be fine, but the state variables that are used in resampling need more than 15bit i think. Even "48khz" wont fit in 15bit
[15:01] <Daemon404> grmbl, i cant find said list, which was posted on reddit of all places
[15:02] Action: Daemon404 rememberes massimo's thing
[16:31] <cone-414> ffmpeg.git 03Michael Niedermayer 07master:cddbf95c5f0d: doc/examples/resampling_audio: use av_freep() for saftey
[16:31] <cone-414> ffmpeg.git 03Michael Niedermayer 07master:3f42434600f2: avcodec/huffyuvencdsp: try to fix misaligned access
[17:28] <cone-414> ffmpeg.git 03Michael Niedermayer 07master:7f5d75f95b65: cmdutils: ignore sws_flags if swscale is not enabled
[20:31] <jamrial> BBB: https://github.com/jamrial/FFmpeg/commits/swr
[20:36] <BBB> will check in 1 hr or so, ok?
[20:37] <jamrial> ok
[21:12] <BBB> Ill probably make some small changes to the int16 patch over time, but perhaps after committing the original one
[21:12] <BBB> so all inline is gone now right?
[21:13] <BBB> also, sugestion for float/double
[21:13] <BBB> mostly, the difference is just a s/d diff in instruction
[21:13] <BBB> so
[21:14] <BBB> can we make s/d a optional 4th macro argument, and merge the code for them? so itd be mulp%4, addp%4, etc.
[21:16] Action: Daemon404 wonders what posessed BBB to hack on swr
[21:16] <BBB> ...
[21:16] <BBB> :)
[21:24] <nevcairiel> The inline asm bugged him
[21:30] <Daemon404> perhaps you mean offended
[21:33] <jamrial> BBB: double has a couple instructions less in the horizontal add
[21:34] <jamrial> but aside from that yeah, it's mostly s/d diff
[21:46] <BBB> ok, I guess partial merge with a few %ifidn %1, float & do the extra step in the hadd & %endif
[21:46] <BBB> but that would still look nice
[21:51] <BBB> cpuflag)avx) still works right? (re: fma patch)
[21:51] <BBB> since fma means cpuflag(avx) returns true, right?
[21:51] <jamrial> yeah
[21:53] <jamrial> i changed the cpuflag(avx) check with mmsize == 32 since i added both an xmm fma4 and ymm fma3 versions, and both imply avx
[21:54] <BBB> hm& I see, right
[21:54] <BBB> ok thats fine then
[21:54] <BBB> I dont understand the last bit of the xop patch
[21:54] <BBB> line 507-514
[21:54] <BBB> https://github.com/jamrial/FFmpeg/commit/c1ea59f7273db604e6f5270f999da2a5e1…
[21:55] <jamrial> that's basically the HADDD macro
[21:55] <BBB> it may just be me misparsing it, but it looks like the pshufd was removed?
[21:55] <jamrial> yes, i replaced with PSHUFLW
[21:55] <BBB> so pshuflw == pshufd?
[21:56] <BBB> weird macro
[21:56] <BBB> anyway if its the same code I dont mind
[21:56] <BBB> odd macro name perhaps
[21:56] <jamrial> no, it's pshuflw for mmsize == 16, and pshufw for mmsize == 8
[21:56] <BBB> hm& oh I see
[21:56] <BBB> right
[21:56] <BBB> doh
[21:56] <BBB> ok thats ok then
[21:56] <BBB> that took me forever to parse
[21:56] <BBB> ok rest is good
[21:57] <BBB> so I guess at your option merge a little of the pd/ps code if you think it makes sense (since later you remove that again for the fma3/4 code), and otherwise its ok
[21:57] <BBB> do you have benchies?
[21:58] <jamrial> i could instead just use the HADDD macro like in common which will be cleaner, but i decided to interleave the instructions instead
[21:58] <jamrial> should do the former? out of order execution should in theory take care of it
[21:59] <jamrial> i benched fma3 on ubitux's rig and it was faster than avx. especially for linear
[21:59] <BBB> cool
[21:59] <jamrial> didn't bench any of the int16
[21:59] <BBB> (neither did I)
[21:59] <BBB> :-p
[22:00] <jamrial> hehe
[22:05] <jamrial> i suppose resampling of double precision is rarely used, right?
[22:09] <BBB> I guess so
[22:10] <BBB> I mean, Im by no means an expert
[22:10] <BBB> but the audio I worked on was typically float, or int16 or the lower-end stuff
[22:11] <jamrial> yeah. probably not worth adding an avx/fma version for it then
[22:29] <cone-376> ffmpeg.git 03Michael Niedermayer 07master:c5a405c4f070: swresample/resample_template: flip order of operations in swri_resample_linear() for 32bit
[22:33] <jamrial> ok, pushed new version with the ps/pd changes
[22:49] <cone-634> ffmpeg.git 03Matthew Oliver 07master:09adb1dd364f: configure: enable lto for icl compiler.
[22:49] <cone-634> ffmpeg.git 03Michael Niedermayer 07master:86576129bd22: Merge remote-tracking branch 'cehoyos/master'
[23:05] <BBB> jamrial: ok I had no other comment, so feel free to merge it all (or submit it all)
[23:11] <jamrial> ok, i'll send it later today
[23:12] <jamrial> BBB: btw, you're still checking for HAVE_SSE2_INLINE for the yasm port in your int16 patch
[23:15] <kierank> "If i understood well google with vp9 while working on 10/12 bit wants to switch everything to 16 bit internally and other decoders already do this."
[23:15] <kierank> wtf
[23:15] <kierank> are they nuts
[23:15] <kierank> unattributed quote however
[23:45] <nevcairiel> Those people usually don't understand anything correctly
[23:58] <BBB> jamrial: just change it :)
[23:58] <jamrial> alright
[00:00] --- Mon Jun 30 2014
1
0
[00:04] <c_14> dahat: Check under windows? https://trac.ffmpeg.org/wiki/CompilationGuide
[00:44] <dahat> c_14: I did, that ended up taking me to http://ffmpeg.org/platform.html#Microsoft-Visual-C_002b_002b-or-Intel-C_002… ... and when trying to do a /.configure --toolchain=msvc (or other suggested configure arguments)... I end up being told that (short version) "C compiler test failed"... and while there is a suggestion of trying --enable-cross-compile... that becomes whole other rabbit hole I've not been able to
[01:10] <ajeesh> http://pastebin.com/QUMK0wRi
[01:11] <ajeesh> cannot convert some wmv files
[01:45] <ajeesh> cannot convert to flv for some wmv files
[03:57] <hellhound> hey guys, I was wondering if there was a way to embed subtitles with ffplay
[05:40] <Gregor> I have a ridiculous question: Is it possible to make ffmpeg discard all I-frames for a given input (I'm trying to intentionally create an invalid video here, to demonstrate the oft-seen effect of P/B-frames applied to invalid picture data)
[05:41] <Gregor> Or perhaps even better, to replace the data of an I-frame.
[06:12] <Gregor> It's surprisingly difficult to find information on how to intentionally corrupt video ^^Ž
[08:17] <bewees> hi, I'd like to merge two mp4 video files without reencoding. I tried: ./ffmpeg -i 1.mp4 -i 2.mp4 merged.mp4 without success. Not sure what Im missing
[08:17] <bewees> My ffmpeg version is: ffmpeg version 2.2.3
[08:20] <bewees> This the output of my attempt: http://pastebin.com/EP3hWqVJ
[08:21] <relaxed_> bewees: (MP4Box is part of gpac) MP4Box -cat 1.mp4 -cat 2.mp4 -new combined.mp4
[08:26] <bewees> relaxed_, Thanks, that worked fine. Is ffmpeg using gpac for mp4 conversion too, but it failed because of a build error?
[08:33] <bewees> I guess not, because else gpac would be a dependencies of ffmpeg
[08:34] <relaxed_> bewees: https://trac.ffmpeg.org/wiki/How%20to%20concatenate%20%28join,%20merge%29%2…
[08:35] <relaxed_> I just find mp4box easier to use
[08:40] <bewees> Thanks
[11:15] <smartcat> hi, friends, a question about mpeg2 encoding. I try to encoding a HD video to 60Mbps MPEG2 format with ffmpeg, I got a poor performance with only about 5 fps with my I5-3470 processor. Is it right? the encoding speed is quite slow
[12:01] <GyrosGeier> hm
[12:01] <GyrosGeier> I have a set of VOBs, can I concatenate these during encoding?
[12:32] <technocf> I am trying to convert images to a video but I get "Output file #0 does not contain any stream"
[12:33] <technocf> Using the command "avconv -i frame_%d.png -r 30 video.mp4"
[12:40] <ubitux> GyrosGeier: vob are mpeg you should be able to concat them, see the faq
[12:40] <ubitux> technocf: avconv is not supported here, it's a fork
[12:42] Action: GyrosGeier checks
[12:44] Action: GyrosGeier tries
[12:47] Action: GyrosGeier will see in 4 hours, when the encoder reaches the first boundary
[13:50] <GyrosGeier> hm
[13:50] <GyrosGeier> is there machine-readble output from the encoding process?
[13:51] <GyrosGeier> or should I use the library API then?
[13:51] <GyrosGeier> (plan is to have a simple fronend that watches a directory and starts encoding, with progress reports to XBMC)
[14:40] <BlackDream> Hello, I'm trying to restream a stream using the following command: /usr/local/bin/ffmpeg -i http://4.31.30.145:9080 -acodec copy -vcodec copy -f flv rtmp://127.0.0.1:88/live/Test123
[14:41] <BlackDream> The ffmpeg responds:
[14:41] <BlackDream> Output #0, flv, to 'rtmp://127.0.0.1:88/live/Test123': Metadata: encoder : Lavf55.33.100 Stream #0:0(): Video: h264 ([7][0][0][0] / 0x0007), yuv420p, 720x576 [SAR 1:1 DAR 5:4], q=2-31, 1k tbn, 90k tbc Stream #0:1(): Audio: mp3 ([2][0][0][0] / 0x0002), 44100 Hz, mono, 56 kb/s
[14:41] <BlackDream> which says that it has audio and video codecs and the restream starts successfully!
[14:41] <BlackDream> Then i open my player i have video image, but no sound though it says it has sound codec
[14:41] <BlackDream> why is tha that happening? Thank you :)
[14:41] <chris1988> hmmmmmm
[14:47] <brontosaurusrex> BlackDream, muted audio?
[14:48] <BlackDream> no no the audio works just fine, for example if i restream it using VLC i have audio
[14:48] <BlackDream> with ffmpeg i dont have though it says it has audio codec
[14:58] <BlackDream> btw: i am using rtmp nginx module
[16:22] <farito> how does -acodec aac compare with -acodec libfaac ?
[16:22] <sacarasc> Both are crap!
[16:23] <sacarasc> Check out https://trac.ffmpeg.org/wiki/Encode/AAC
[16:24] <sacarasc> And https://trac.ffmpeg.org/wiki/GuidelinesHighQualityAudio
[16:26] <farito> says "libfdk_aac > libfaac > Native FFmpeg AAC encoder (aac) > libvo_aacenc" ok
[16:28] <farito> ta
[16:38] <farito> How can I keep the current video nd audio codecs and rotate a video 90 degrees? is it possible?
[16:41] <farito> ffmpeg -i V.mp4 -c:a copy -c:v copy -vf "transpose=1" r.mp4 <-- says Filtergraph 'transpose=1' was defined for video output stream 0:0 but codec copy was selected. Filtering and streamcopy cannot be used together.
[16:41] <sacarasc> -c copy copies the stream, you would have to reencode.
[16:43] <fabltd> Hi i am trying to transcode an http stream http://pastebin.com/CzyzW3Rc
[16:43] <fabltd> ffmpeg is seeing the stream url as agruments to the application and the stream fails to transcode.
[16:54] <fabltd> hello is there any one online?
[16:55] <c_14> fabltd: If you don't paste your complete console output we can't help you.
[16:55] <fabltd> ok sorry i added it to the pastbin
[16:55] <fabltd> here it is
[16:55] <fabltd> ffmpeg -i http://192.168.100.101/dlna/?type=DVB-S-AUTO&src=1&freq=10773&pol=h&sr=2200… -ss 00:00:05 -acodec libfdk_aac -vbr 3 -b:v 3000k -ac 2 -vcodec libx264 -preset superfast -tune zerolatency -threads 2 -s 1280x720 -flags -global_header -fflags +genpts -map 0:0 -map 0:1 -hls_time 2 -hls_wrap 40 /var/www/test.m3u8
[16:56] <c_14> The console output, ie everything ffmpeg says. Not just the command.
[16:56] <fabltd> oh sorry
[16:57] <fabltd> here it is http://pastebin.com/rKZnHLb2
[16:59] <sacarasc> Put \ before all the &s.
[17:00] <farito> What is a good way to get ffmpeg working with libfdk_aac? I am on Windows
[17:00] <farito> I understand I need to compile it somehow?
[17:00] <c_14> fabltd: there's something wrong with those quotes...
[17:01] <fabltd> i have tried with any with out the quotes.
[17:03] <fabltd> with out quotes http://pastebin.com/ULqr1WvS
[17:04] <sacarasc> Put \ before all the &s.
[17:04] <c_14> you either need to escape all the '&' chars like sacarasc said or quote the entire string in u+0022 quotes instead of the u+201C u+201D quotes you were using.
[17:08] <fabltd> great different error now :-( Unrecognized option 'vbr'.
[17:08] <fabltd> Error splitting the argument list: Option not found
[17:12] <farito> Is compiling ffmpeg with libfdk_aac more difficult in windows than in *nix?
[17:13] <sacarasc> Yes.
[17:13] <sacarasc> But compiling everything is more difficult in Windows.
[17:19] <fabltd> getting this on the cli [http @ 0x20526c0] HTTP error 406 Not Acceptable
[17:19] <fabltd> http://192.168.100.101/dlna/?type=DVB-S-AUTO&src=1&freq=10773&pol=h&sr=2200…: Input/output error
[17:27] <fabltd> c_14 are you still here ?
[17:32] <c_14> ye, that error is rather strange though. Does that url provide some sort of stream when you open it
[17:32] <c_14> ?
[17:36] <garyserj2> if I want to get ffmpeg via apt-get, how do I ensure that I get the proper ffmpeg and not the libav fork?
[17:37] <c_14> IIRC the debian repos do not contain FFmpeg. Depending on whether you're on debian or ubuntu you can use an external repo both of which are linked here: https://ffmpeg.org/download.html
[17:50] <GyrosGeier> hm
[17:51] Action: GyrosGeier used vobcopy to copy a DVD to harddisk
[17:52] <GyrosGeier> I would have expected a single episode, but rather all of them are there concatenated
[17:52] <GyrosGeier> can I somehow select that during encoding?
[17:52] <fabltd> c-14 yes it does works fine in vlc
[17:56] <c_14> Do you know what headers vlc sends to http servers? you can test by running a netcat on localhost on some port and then openning http://localhost:someport with vlc
[18:02] <fabltd> GET / HTTP/1.1
[18:02] <fabltd> Host: 192.168.100.10:25
[18:02] <fabltd> User-Agent: VLC/2.1.4 LibVLC/2.1.4
[18:02] <fabltd> Range: bytes=0-
[18:03] <fabltd> Connection: close
[18:03] <fabltd> Icy-MetaData: 1
[18:03] <fabltd> sorry
[18:04] <c_14> Hmm, the only difference is the Icy-MetaData part. Not sure why the server is throwing a 406
[18:05] <fabltd> grrr guess i might have to stick to vlc?
[18:06] <c_14> Although, wait. ffmpeg sends an accpet header, vlc does not...
[18:07] <fabltd> ok can this be switch off?
[18:07] <c_14> Try setting the headers option to exactly what vlc sends
[18:07] <GyrosGeier> 406 is "Unacceptable"
[18:07] <GyrosGeier> i.e. it didn't find a configuration matching the Accept header
[18:09] <fabltd> how do i do that c14
[18:18] <fabltd> ?
[18:19] <c_14> Maybe just try with ffmpeg -headers "Accept: *" -i "http://[..]", not sure how to completely delete the Accept header since it always sends it by default and only sometimes wants to overwrite it.
[21:16] <ChocolateArmpits> Has anyone had any luck compiling tools/ismindex on windows ?
[22:05] <spectralsun> what is this avconv nonsense
[22:05] <spectralsun> :\
[22:06] <spectralsun> which is better?
[22:06] <spectralsun> why use one over the other
[22:07] <c_14> That depends on who you ask. I prefer FFmpeg because it actually works.
[22:07] <c_14> Also, this channel is only for FFmpeg support, so if you want to use avconv you'll need to go to #libav.
[22:08] <sacarasc> spectralsun: ffmpeg merges in all of the new libav stuff, but libav doesn't merge in all of the new ffmpeg stuff.
[22:08] <spectralsun> hmm
[22:08] <spectralsun> ive always used ffmpeg, was going to continue to
[22:09] <spectralsun> one of our ubuntu servers updated to 14 and i got thrown into this nonsense
[22:09] Action: spectralsun is compiling ffmpeg again on the server
[22:09] <spectralsun> just did it from the ubuntu instructions which has you use a 2.2 snapshot
[22:10] <spectralsun> i had followed the other linux instructions which has you clone the latest from sources
[22:10] <spectralsun> on my local computer
[22:10] <spectralsun> and i calibrated our test cases for my ffmpeg (md5 checksums)
[22:10] <spectralsun> just ran the tests on the 2.2.git version and they did not match :\
[22:10] <spectralsun> i am building the latest version on the ubuntu server
[22:10] <spectralsun> i am running arch locally
[22:13] <spectralsun> the only other differnce i can find is i have gcc 4.9.0 and the server has 4.8
[22:13] <sacarasc> GCC 4.9.0 is broken (if unpatched) with ffmpeg.
[22:16] <sfan5> I wonder how the people developing gcc manage to break programs with every new release
[22:16] <spectralsun> oh
[22:17] <spectralsun> really
[22:17] <spectralsun> hmm
[22:17] <spectralsun> ugh
[22:17] <spectralsun> the one from my pacman is built with 4.9.0 also
[22:18] <spectralsun> sacarasc: how do you patch or know if its not working?
[22:21] <sacarasc> Seems it was only flac that was affected.
[22:26] <spectralsun> ok
[22:27] <spectralsun> im not using that
[22:31] <spectralsun> wow
[22:31] <spectralsun> new hashes, again
[22:32] <spectralsun> ran my tests on the ubuntu server, this time using latest source compiled by gcc 4.8
[22:32] <c_14> hashes of what exactly?
[22:32] <spectralsun> different hashes from the 2.2.git version and also different from my local copy which should be identicle execpt for being compiled by gcc 4.9
[22:33] <spectralsun> well one is a slowmo test for example
[22:33] <spectralsun> it takes a video file, at 4 seconds in it takes the next 5 seconds and makes it slowmo
[22:33] <c_14> Is the test deterministic? ie if you do it twice does it generate the same hashes?
[22:33] <spectralsun> yep
[22:34] <spectralsun> i calibrated it to my local arch computer
[22:34] <spectralsun> it doesn't match on the ubuntu server and its different from the last time i compiled ubuntu on the server (which was a 2.2 snapshot)
[22:45] <spectralsun> hmm looks like theres a gcc update for my local computer, im gonna recompile on my arch
[23:12] <spectralsun> wow
[23:12] <spectralsun> different hashes again
[23:12] <spectralsun> locally this time
[23:12] <spectralsun> and they dont match the server
[00:00] --- Mon Jun 30 2014
1
0
[00:35] <cone-849> ffmpeg.git 03Michael Niedermayer 07master:4d1fa38d984c: avcodec/mpeg12dec: Limit progressive_seq reinitilaization to where the resolution changes
[03:02] <BBB> jamrial: new patchset on my github, common resample functions are now quite a bit faster than gcc
[03:03] <BBB> linear not yet updated, so not yet ready to merge, but it shows progress
[04:04] <jamrial> BBB: nice!
[04:05] <jamrial> linear should also be faster considering you optimized the horizontal sum after the inner loop
[04:06] <BBB> right, but I havent updated them to this new style/methodology yet
[04:07] <BBB> so its still the old function, which is probably slower than gcc
[04:07] <BBB> (I really wanted to try to get the fullest out of the common one first)
[04:28] <fionag> BBB: https://github.com/rbultje/ffmpeg/commit/1f7d5e5af5ae41e3eb1e4fe423797ba924… ?
[04:30] <BBB> fionag: yes
[04:31] <fionag> are suggestions/comments okay?
[04:31] <BBB> sure
[04:32] <BBB> dont comment on the linear function yet, Im about to re-do that
[04:32] <fionag> which one is the linear function?
[04:32] <fionag> oh, the second one, I see.
[04:32] <BBB> (resample_linear_float, line 236 and below in resample.asm)
[04:32] <fionag> darn, there went my comment :P
[04:32] Action: fionag looks only in the upper function then
[04:33] <BBB> sorry :]
[04:33] <BBB> what was the comment? Im curious
[04:33] Action: fionag was thinking that if line 339's speed matters at all, maybe cache 1 / c->src_incr and multiply by it?
[04:34] <fionag> I'm also not quite sure what line 180 is doing, it feels like it serves no purpose but I must be missing something
[04:36] <BBB> look at addps 3 lines down
[04:36] <BBB> a = b; shuffle b around; b += a
[04:38] <fionag> it's doing a horizontal sum though, right?
[04:38] <fionag> the one on lines 344-348 does that without an extra moss.
[04:38] <fionag> *movss
[04:38] <BBB> its a horizontal add, basically x=abcd; y=movhlps(x) /* i.e. cd */; x += y /* now, x=a+c, b+d, & */; y = x; x = shufps(x) /* now x is b+d, & */; y += x /* y is a+b+c+d */;
[04:38] <fionag> so I was wondering why they needed to be different
[04:38] <BBB> well theres a movss, but its hidden
[04:38] <BBB> + shufps xm1, xm0, xm0, q0001
[04:39] <BBB> is actually movaps xmm1, xmm0; shufps xmm1, xmm0, q0001
[04:39] <fionag> yeah, but that doesn't work on line 180-181 for some reason?
[04:39] <BBB> movaps and movss have similar complexity if its register-to-register
[04:39] <BBB> it does, I should probably change it
[04:39] <fionag> that function supports AVX, sooo
[04:39] <fionag> other than that I can't think of much of anything, it looks really well done
[04:39] <BBB> yeah itd help for avx
[04:40] <BBB> sadly I dont have an avx capable machine
[04:40] <fionag> though out of curiosity, lines 139-146: isn't the "typical" for that to name them varM instead of varD because they're in memory instead of registers? or am I remembering wrong
[04:40] <BBB> pushed new version
[04:41] <BBB> yeah it is; reason I do that is so I can share the main loop between x86-32 and -64
[04:41] <BBB> its probably slightly confusing for a casual reader but I figured it made sense
[04:41] <BBB> maybe I should rename it to something that doesnt end in d or q or m, to indicate it can be both
[04:42] <BBB> the d currently indicates that on x86-64, its a register (and on x86-32, it may be, but probably is not)
[04:43] <fionag> ahhhh, I see
[04:43] <fionag> that makes sense then, maybe just a comment explanining it would be fine
[04:44] <fionag> naming based on x86-64 (the more sane architecture) sounds reasonable
[04:45] <BBB> its just nice to have more registers ;)
[04:45] <BBB> esp. in a function like this
[04:46] <fionag> for the inner loop in the first function, is there any benefit to unrolling it? I guess maybe not because it's probably load-bound (?)
[04:46] <fionag> oh! is it just me or is there room for an fmaddps on line 167 :3
[04:46] <jamrial> there is
[04:48] <jamrial> fma3/4 and xop were part of my plans before BBB scolded me for writing inline avx :P
[04:49] <BBB> fionag: I suppose you could unroll it& I was basically planning to go the other way and do multiple output samples simultaneouls
[04:49] <BBB> *simultaneously
[04:50] <BBB> but Ill do that after the main rewrite is in
[04:58] <BBB> fionag: Ill make a note about the divss -> mulps (and outside loop preparation of that register), gcc does prepare it outside the loop but keeps using divss (I guess it doesnt really have a choice, we did specifically ask for a float div&)
[05:08] <fionag> yeah, it can't do that optimization without -ffast-math (and I'm not even sure it can do it wth)
[05:09] <fionag> what do you mean by multiple output samples simultaneously? oh gosh, is this inner loop calculating a -single- output sample? o_o
[05:17] <michaelni> jamrial, seems the mmxext emu edge code causes valgrind failures
[05:18] <michaelni> Invalid read of size 4 at 0xCC0CC0: ??? (videodsp.asm:438)
[05:20] <michaelni> jamrial, should i revert the commit ? from naive quick look fixing it would kind of undo the SPLATB_LOAD change
[05:22] <jamrial> i assume it's the "movd %1, [%2-3]" part of SPLATB_LOAD, right?
[05:23] <jamrial> anyway, yes, the whole point of this change was to get an ssse3 version, and since it didn't seem to be any faster than the mmxext then we can just undo it all
[05:30] <michaelni> jamrial, valgrind just pointed at the macro not at the instruction, so iam only 99% sure it was the movd
[05:31] <michaelni> ill double check
[05:32] <jamrial> well, not many candidates considering it's the only instruction reading four bytes in any way
[05:33] <michaelni> yes, confirmed commenting the 2 movd removed the valgrind issue
[05:33] <michaelni> ill revert the commit, thanks
[05:40] <BBB> fionag: oh yeah :)
[05:41] <BBB> fionag: all simd comes from horizontally doing the src, but dest is still scalar
[05:41] <BBB> fionag: I foresee some more speedup if that becomes simd'ed
[05:41] <BBB> anyway
[05:44] <cone-849> ffmpeg.git 03Michael Niedermayer 07master:5bca5f87d1a3: Revert "x86/videodsp: add emulated_edge_mc_mmxext"
[05:56] <BBB> so first linear I beat gcc by 2 cycles/sample on x86-64, thats not a lot
[05:56] <BBB> I was hoping for more
[05:57] <BBB> so Ill continue checking over the weekend
[05:57] <BBB> 32bit wont compile because Im not very familiar with x87 fp emu
[06:10] <BBB> oh my bad, I cant count; 32bit 66(gcc)->52 (yasm), 64bit 58(gcc)->48(yasm)
[06:10] <BBB> llvm was 66 (32bit) or 63 (64bit)
[06:11] <BBB> so thats a pretty big margin
[06:14] <BBB> pushed that to github also, please test on win64, avx and let me know if the speed measurements hold up this time
[06:14] <BBB> they should
[06:14] <BBB> but good to test
[06:14] <jamrial> win64 will probably be slightly slower than unix64 since you're using 15 gprs
[06:15] <jamrial> but still faster than inline nonetheless
[06:15] <BBB> well thats outside the loop
[06:15] <BBB> the loop iterates like 1000x or so, so the outside-loop-stuff is really very insignificant to overall cycle/sample performance
[06:15] <jamrial> true
[06:15] <BBB> (1000x per call)
[06:16] <BBB> but anyway, yes, what Im looking for is that win64 is significantly faster than inline asm with whatever compiler
[06:16] <BBB> and that it passes fate-swr
[06:16] <BBB> and that avx still works also ;)
[06:28] <jamrial> win64 (avx and not) work here
[06:29] <BBB> \o/
[06:30] <BBB> wanna speed-torture it?
[07:04] <fionag> BBB: my intuition would be that doing it the other way would be better, i.e. use N-wide SIMD to do N samples at once instead of N filter coefficients at once
[07:07] <fionag> so an M-wide filter would have an M-iteration loop that does N input samples, I think?
[07:20] <cone-849> ffmpeg.git 03Michael Niedermayer 07master:21bfed5b06f0: avcodec/mpegvideo_enc: reduce space between blocks in emu_edge in encode_mb_internal
[07:20] <cone-849> ffmpeg.git 03Michael Niedermayer 07master:504475f38ef0: avcodec/mpegvideo: dont overwrite emu_edge buffer
[07:21] <agentorange> hi so does anyone know why Im getting "missing picture in access unit with size ..."
[07:33] <BtbN> Please don't cross post question
[07:48] <agentorange> haha
[07:48] <agentorange> great answer
[10:23] <kurosu> jamrial / BBB: some times, I feel like we should have an equivalent to emms_c for x86_64
[10:23] <kurosu> for win64, store at the highest level the 10 callee-saved regs
[10:23] <kurosu> restore them when exiting (like emms_c)
[10:24] <kurosu> and never care for the callee-saved matter
[10:25] <kurosu> I haven't checked the disassembly, but seeing how hevc_mc declares fewer regs than needed and doesn't have any issue on win64
[10:26] <kurosu> it seems like this could be avoided if we now the compiler isn't going to use xmm6+ in the surrounding code (eg, no float)
[12:39] <cone-635> ffmpeg.git 03Vittorio Giovara 07master:f134b5ec53b4: apichanges: fill in changes for lavu 51.19 and 51.20
[12:39] <cone-635> ffmpeg.git 03Michael Niedermayer 07master:e925a82e2478: Merge commit 'f134b5ec53b4cb51cb69bf0c64de87687ea72b12'
[12:44] <michaelni> kurosu, a compiler could use xmm6+ in plain C code if it wanted, theres no gurantee that it doesnt, the same is of course true for mmx
[12:58] <ubitux> fate is a bit yellow
[12:59] <kurosu> michaelni, except it hardly ever does - but I agree this is not guaranteed
[13:00] <michaelni> ubitux, i fixed 2 fate issues before i went to bed
[13:01] <michaelni> some fate clients are still yellow from before that
[13:01] <ubitux> ah, ok
[13:02] <michaelni> kurosu, yes, we should just be carefull not to add code that makes assumtations about this and then forget about it, could be alot of work to debug
[13:03] <kurosu> michaelni, yeah, I didn't intend to actually try it
[13:03] <kurosu> I don't think we want to add this just for win64
[13:04] <kurosu> that was mostly a remark on how asm has to bother with it, when it sometimes actually doesn't matter
[13:47] <cone-635> ffmpeg.git 03Vittorio Giovara 07master:39975acc699c: rtpenc_jpeg: check for color_range too
[13:47] <cone-635> ffmpeg.git 03Michael Niedermayer 07master:823ea19a74ef: Merge commit '39975acc699c83af0a87a7318c0f41e189142938'
[13:47] <cone-635> ffmpeg.git 03Michael Niedermayer 07master:7448afc52b69: APIchanges: fillin some missing data
[13:47] <cone-635> ffmpeg.git 03Michael Niedermayer 07master:cc923727226b: doc/APIchanges: lengthening a hash to make it non ambigous
[13:50] <BBB> fionag: I want to do both, basically
[13:51] <BBB> fionag: so right now we do (in sse/avx) 4/8 filter samples per inner loop iteration, and then horizontally add the final sse/avx sum register to get the total sum in scalar for that one value
[13:52] <BBB> fionag: so Im thinking, why not do that inner loop 4/8x, so that the hadd can be over multiple registers (haddps) and the store is simd also
[13:52] <BBB> fionag: I think its more efficient that way by probably one or maybe several cycles/sample, which is pretty huge
[14:00] <BBB> also sent patches to ML, I think this is good enough to merge
[14:00] <BBB> we can work on top of this later
[15:41] <cone-635> ffmpeg.git 03Ronald S. Bultje 07master:ddb7b4435a84: swr: move dst_size == 0 handling outside DSP function.
[16:11] <cone-635> ffmpeg.git 03Michael Niedermayer 07master:a348f4befe23: avfilter/x86/vf_pullup: fix "invalid combination of opcode and operands" with nasm
[16:44] <BBB> weird...
[16:45] <BBB> michaelni: I have no idea what the build issue is, can you get more specific macro line numbers?
[16:56] <michaelni> when copy & paste inlining i get errors in xorps m0, m0 ; mulps m1, [filterq+min_filter_count_x4q*1] ; addps m0, m1 ; xorps m0, m0 ; xorps m2, m2 ; mulps m1, [filter1q+min_filter_count_x4q*1] ; addps m2, m3 ; addps m0, m1
[16:57] <michaelni> i wonder if theres a easier way to get line numbers from inside macros
[16:58] <BBB> maybe they need to be 3-operand?
[16:58] <BBB> can you try changing each instruction a, b to instruction a, a, b?
[16:59] <BBB> (in that sequence of things that fail)
[16:59] <BBB> weird that it works for me...
[17:00] <michaelni> seems to build with 3op
[17:00] <BBB> &
[17:01] <BBB> Ill re-submit
[17:04] <kurosu> BBB / michaelni: generally I do it in 2 steps
[17:04] <kurosu> call yasm only to preprocess the asm file, output the result
[17:04] <kurosu> filter out the debug/line number info
[17:04] <kurosu> compile
[17:04] <kurosu> see what the actual issue is
[17:05] <kurosu> could maybe be automated though some PREPROCESS=1 parameter to make
[17:06] <BBB> ok re-sent
[17:06] <BBB> kurosu: thatd be nice yes
[17:08] <michaelni> undoing macros automatically would indeed be very nice, and this is true for C too
[17:08] <BBB> benjamin once said that DEBUG=5 or so as a configure option would help unfold C macros nicely and give good debug info
[17:09] <BBB> like, there was some gcc option for it
[17:09] <BBB> I dont recall exactly
[17:27] <BBB> google is fun, I randomly wonder over this pastebin entry: http://pastebin.com/QZ2pP70F
[17:27] <BBB> so apparently touching someone elses code makes it yours& very interesting legal opinion, Im mildly shocked
[17:28] <nevcairiel> Should just ignore their twisted world
[17:37] <cone-635> ffmpeg.git 03Ronald S. Bultje 07master:faa1471ffcc1: swr: rewrite resample_common/linear_float_sse/avx in yasm.
[17:42] <BBB> yay \o/
[17:56] <iive> "gcc -E Stop after the preprocessing stage; do not run the compiler proper. The output is in the form of preprocessed source code, which is sent to the standard output."
[17:57] <iive> I think it used to be -P before, but at some version it was changed.
[20:49] <cone-635> ffmpeg.git 03Michael Niedermayer 07master:f9f8491ddf6d: avcodec/cavs: make cavs_chroma_qp non static
[20:49] <cone-635> ffmpeg.git 03Yao Wang 07master:546491667755: avcodec/cavs: improve conformance with rm52j reference decoder
[21:30] <cone-635> ffmpeg.git 03Michael Niedermayer 07master:c599c211fb71: avformat/mux: fix flush_packets flag with flushing buffers
[21:30] <cone-635> ffmpeg.git 03Michael Niedermayer 07master:e6aba1be4cb0: avformat/options_table: Fix flush_packet flag flags
[21:44] <cone-635> ffmpeg.git 03Martin Storsjö 07master:7b0c7c9163fe: arm: Detect 32 bit cpu features on ARMv8 when running on a 64 bit kernel
[21:44] <cone-635> ffmpeg.git 03Michael Niedermayer 07master:01983e50c024: Merge commit '7b0c7c9163fe3dd0081696befde28617119d2590'
[21:55] <cone-635> ffmpeg.git 03Michael Niedermayer 07master:4d3072ada3d5: doc/examples/muxing: remove unused variable
[00:00] --- Sun Jun 29 2014
1
0
[01:35] <rhg> warning
[01:35] <rhg> you may be watched
[01:35] <rhg> do usa&israel use the internet(facebook,youtube,twitter, chat rooms ..ect)to spy??
[01:35] <rhg> do usa&israel use the internet 2 collect informations,,can we call that spying??
[01:35] <rhg> do they record&analyse everything we do on the internet,,can they harm you using these informations??
[01:35] <rhg> ÊÍÐíÑ
[02:29] <jlewin> Hi all, running Windows and the lastest build of ffmpeg. Attempting to switch from UScreenCapture to screen-capture-recorder but having problems with the crop filter; it crops to the target size but then scales to the source resolution. Any thoughts or tips appreciated.
[02:52] <jlewin> Well, it looks like crop followed by scale achieves the results I was looking for. Not sure if it's the best approach but it seems to be working
[02:52] <jlewin> -vf "crop=1024:576:295:161,scale=1024:576"
[03:45] <Technicus> Hello, I am looking for some help with a < http://www.startech.com/AV/Converters/Video/HDMI-to-ExpressCard-HD-Video-Ca… >. It is a very messed up story. I found this device yesterday and it was listed on the StarTech website as having Linux support. So I ordered it with overnight delivery, it showed up before I was even awake.
[03:45] <Technicus> After work, I started messing with this device. The instruction manual has only one reference to Linux, "Compatible Operating Sustems Linux kernel 2.6.14 or later". I could not find any tutorials, forums or examples of how to interact with this device. I am competant with Linux, but I've never messed with a video capture device before.
[03:45] <Technicus> So I called the technical support line. I was informed that Linux support has ended and there was no assistance to be offered, other than a full refund to include shipping charges. LAME! Even though the instruction manuel that came with the device in the box indicates Linux support, they refuse to honor their claims. WEAK! The tech support explained that since there is a new revision I can't be helped. Check out the different versions of the manua
[03:45] <Technicus> l page 7, which is 10 of the pdf:
[03:45] <Technicus> Manual Revision: 12/19/2011 < http://mediapool.xpressplatforms.com/product-files/pdf/816359-36545.pdf >; Manual Revision: 05/15/2012 < http://sgcdn.startech.com/005329/media/sets/ECHDCAP_Manual/ECHDCAP.pdf >.
[03:45] <Technicus> Yesterday when I found this device it was listed on the StarTech website as having Linux support, today it does not.
[03:45] <Technicus> Will someone please help me to determine if I can get this device to cooperate with Linux, or direct me to a channel that would be more approporate for this discussion? Thanks.
[03:47] <sacarasc> You'll probably need a kernel module driver thing.
[03:49] <Hello71> do other expresscards work?
[03:49] <Technicus> sacarasc: lspci reports the card as: "06:02.0 PCI bridge: PLX Technology, Inc. Device 8603 (rev aa)"
[03:49] <Hello71> and I thought you wanted a usb device
[03:50] <Technicus> Hello71: I was looking for usb devices until I found this, and StarTech had it posted with Linux support.
[03:50] <Hello71> lspci -vv -nn -s 06:02 | wgetpaste
[03:50] <Hello71> or sprunge or whatever
[03:51] <Technicus> I would not have purchased it had it not been described as having support.
[03:52] <Hello71> I wouldn't trust manufacturers on this
[03:52] <Hello71> 21:50:42 < Hello71> lspci -vv -nn -s 06:02 | wgetpaste
[03:53] <Technicus> Hello71: < http://bpaste.net/show/415150/ >
[03:54] <Hello71> that's a... PCI bridge?
[03:55] <Hello71> delete -s 06:02
[03:55] <TiZ> What's the most efficient way to add a background image to a video that is comprised of an x11grab capture and a v4l2 capture in opposite corners of the screen?
[03:56] <Technicus> Hello71: < http://bpaste.net/show/415160/ >
[03:57] <TiZ> I tried -i background.png and all the framerate kludges and input order kludges to make it adhere to the rest of my sources, but it ended up being way more taxing than just padding and leaving the background black, and if not that, it caused the x11grab to get REALLY choppy.
[03:58] <Hello71> Technicus: what distro?
[03:59] <TiZ> Here's my current ffmpeg command line: http://pastebin.com/k2yFBSNf How would I modify it to add a nice little background image without messing up my other inputs or synchronization, and without incurring too much extra CPU cost?
[03:59] <Technicus> Hello71: Arch Linux
[04:00] <Hello71> Technicus: #archlinux can probably help you better; barring that, ##linux
[04:00] <Hello71> although I'd recommend against the latter.
[04:01] <TiZ> ##linux isn't usually very useful. And often straight up hostile toward Arch users.
[04:26] <Technicus> TiZ: Most channels are hostile.
[04:28] <sacarasc> TiZ: -loop 1 -i foo.png maybe?
[04:30] <TiZ> sacarasc: With that, it forces the framerate to 24 fps.
[04:30] <sacarasc> TiZ: -loop 1 -r 60 -i foo.png maybe?
[04:30] <sacarasc> Oh, you're using -r 30
[04:30] <sacarasc> That, then!
[04:32] <TiZ> That... kinda works? It's at 29.
[04:34] <TiZ> Normally, I can get 30 FPS no problem. With -loop 1 -r 30 -i foo.png, and filters adjusted accordingly, I get 29 FPS. :|
[04:34] <TiZ> It also makes no difference what image format I use.
[04:35] <TiZ> http://pastebin.com/Rjet2qCR is what I've got right now.
[04:36] <sacarasc> Only 1 thread?
[04:36] <TiZ> Trying to keep it from stressing games.
[04:36] <TiZ> I'll bump up to 2 and see if it makes a difference.
[04:55] <TiZ> sacsarasc: 2 threads makes it work out... but then I put it in the script that's generating the command and then it gets choppy :/
[06:57] <Technicus> Back to the drawing board, check this thing out: < http://www.epiphan.com/products/frame-grabbers/dvi2usb-3-0/ >.
[07:21] <agentorange> hi so does anyone know why Im getting "missing picture in access unit with size ..."
[07:59] <ajeesh> http://pastebin.com/v6iChFsb
[08:00] <ajeesh> wmv: Error while opening file
[08:00] <ajeesh> but other test wmv files are converting
[08:03] <ajeesh> http://pastebin.com/QavUH2P1
[11:28] <Filarius> Hello, under Windows OS I use Livestreamer to record RTMP streams. Often I have issue with recorded files - players (VLC too) do not see video stream or codec (VLC show error "*undef* video codec"), and ffmpeg too "Stream #0:0: Video: none". I found I can fix files with Avidemux(at last to see some part), but I can`t find this software good for processing many files. I wonder how I can use
[11:28] <Filarius> FFmpeg or else software to fix so files.
[11:57] <kippi> Ji
[11:57] <kippi> Hi
[11:58] <kippi> I long time ago someone helped me with building a ffmpeg command that worked out motion, it was a lovley one liner using motion filter, is there any way I can search the archives for this command
[15:17] <Dae> Hiya. Does anyone know how you could use ffmpeg to find the "loudest parts" of a video? i.e. with a given input video, output a clip of footage where the audio stream amplitude was over a given threshold?
[17:22] <rymoshco> irc.rymosh.com
[20:06] <Reventlov> Yosh.
[23:48] <dahat> Any chance someone knows of an idiot proof guide to building FFmpeg under windows with the VS2013 toolchain (and MiniGW if need be))? ... as apparently, I'm a pretty good idiot in this case.
[00:00] --- Sun Jun 29 2014
1
0
[00:12] <cone-973> ffmpeg.git 03Anton Khirnov 07master:6a927d7aaf56: output example: use a macro instead of a static variable
[00:12] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:877716a79c9d: Merge commit '6a927d7aaf5625e83a674072913b9e292a303fd1'
[00:21] <cone-973> ffmpeg.git 03Anton Khirnov 07master:a7fcd4122b19: output example: store the scaling context in the stream context
[00:21] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:9a2850c26af8: Merge commit 'a7fcd4122b19b0f934020f4e261d0c44c4c32e11'
[01:01] <cone-973> ffmpeg.git 03Anton Khirnov 07master:182d3f8221c2: output example: rewrite encoding logic
[01:01] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:8469c73576f7: Merge commit '182d3f8221c23716ea4eafa460bdb94968f71f26'
[01:11] <cone-973> ffmpeg.git 03Anton Khirnov 07master:a5864e9a2531: output example: free the muxing format context properly
[01:11] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:ad081e4def36: Merge commit 'a5864e9a253143437471dba61a533e7fa9acec1c'
[01:16] <cone-973> ffmpeg.git 03Kostya Shishkov 07master:422e14f721c2: indeo2: rename stride to pitch for consistency with other Indeo decoders
[01:16] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:826726b83fab: Merge commit '422e14f721c22cf9c19a8e7aae051ba9d559f6b6'
[02:03] <jamrial> michaelni: can you check if this is faster for you? https://github.com/jamrial/FFmpeg/commit/53ea0eec06fe4117bf596470811e1f6224…
[02:03] <jamrial> also, how did you measure it the first time, or what video did you use?
[02:05] <michaelni> was one of the fate samples
[02:06] <michaelni> maybe vsynth1-flv.flv but iam nt sure i picked randomly after adding a abort() in mpegvideo_motion.c to find one that used the code
[02:07] <michaelni> jamrial, do you see a speed gain ?
[02:07] <michaelni> i mean with the original patch
[02:07] <michaelni> didnt yet look at the new one
[02:11] <jamrial> i saw a speed gain in emu_edge_hvar_ssse3, but not in the many hfixtbl_ssse3 functions. in some widths the latter seemed a bit slower, even
[02:11] <cone-973> ffmpeg.git 03Dirk Ausserhaus 07master:5ec6d152e26c: indeo4: B-frames decoding
[02:11] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:357168bcf634: Merge commit '5ec6d152e26c570c0a16ec72c1f354db95708179'
[02:12] <jamrial> this using the timer macros inside emulated_edge_mc() around a single hfix call
[02:13] <michaelni> i had the timers around the emulated edge mc calls
[02:13] <jamrial> not sure how you measured things yourself
[02:14] <jamrial> ok
[02:20] <michaelni> tried the new patch with 1 timer around the 3 calls in mpeg_motion_internal
[02:20] <michaelni> i cant really meassure a difference
[02:22] <michaelni> for testing i created a flv from matrixbench (-i matrixbench_mpeg2.mpg -an test.flv) benchmark was done with ./ffmpeg -threads 1 -i test.flv -f null -
[02:23] <michaelni> the number of calls to this code in the whole file are very few though
[02:23] <michaelni> so at least for this file it shouldnt matter at all how fast it is
[02:23] <michaelni> do you have a file that uses it alot ?
[02:25] <jamrial> no, tried a couple files and most would call it only a few times as well
[02:42] <cone-973> ffmpeg.git 03Diego Biurrun 07master:24f45c16224d: fate: Add dependencies for dct/fft/mdct/rdft tests
[02:42] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:5a636383db9b: Merge commit '24f45c16224d4c5d482e928676714766ffdda4fc'
[03:07] <Timothy_Gu> michaelni: why is matrixbench so often used as benchmarking material?
[03:09] <michaelni> when theres no reason to pick something specific then i tend to pick matrixbench or for rawvideo foreman or raw image lena
[03:13] <Timothy_Gu> so it's just a personal preference? there isn't a reason behind it?
[03:19] <cone-973> ffmpeg.git 03Diego Biurrun 07master:d2869aea0494: dsputil: Move MMX/SSE2-optimized IDCT bits to the x86 subdirectory
[03:19] <michaelni> yes
[03:19] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:09a7a4704e84: Merge commit 'd2869aea0494d3a20d53d5034cd41dbb488eb133'
[03:20] <jamrial> where can i find it for that matter? i don't see it in the fate samples folder
[03:25] <michaelni> jamrial, http://samples.ffmpeg.org/benchmark/testsuite1/
[03:26] <jamrial> thanks
[03:39] <cone-973> ffmpeg.git 03Gaullier Nicolas 07master:134beb9e0222: avformat/mxfenc: rename very recently added option to set/force channelcount in MXF D-10
[03:47] <Compnn> Timothy_Gu : for samples i use microsofts' video for windows sample.avi , its 15fps cinepak :P
[04:02] <Timothy_Gu> lol
[06:18] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:b96c9513f35a: avcodec/snow: factor ff_snow_get_buffer() out
[08:31] <anshul> I reset my ffmpeg git to previous commit and found my repository is now making avplay and avconv, is it normal?
[08:34] <debianuser> That was probably a merge from livav, and you switched to its parent from libav.
[09:55] <ubitux> i have 2 mts in h264, one is interlaced, the other is progressive
[09:55] <ubitux> both mark frames as interlaced
[09:56] <ubitux> do we have any way of fixing this with ffmpeg without re-encoding?
[09:56] <ubitux> like a way to alter packets without the decode/encode pass
[09:57] <ubitux> with bsf maybe?
[10:46] <thardin> http://www.nasaspaceflight.com/2014/06/recovering-falcon-9-ocean-landing-vi…
[10:47] <thardin> didn't know about this before
[12:53] <ubitux> http://blog.securitymouse.com/2014/06/raising-lazarus-20-year-old-bug-that.…
[13:15] <michaelni> ubitux, want to write news entry & update security page ?
[13:15] <ubitux> not really
[13:16] <ubitux> did you make a new release after the fix?
[13:16] <michaelni> yes
[13:16] <michaelni> d6af26c55c1ea30f85a7d9edbc373f53be1743ee
[13:17] <michaelni> need to logoff, will be bak soon
[13:18] <Compnn> whoa
[13:18] <Compnn> michael offline
[13:18] <Compnn> i like these security researchers, always saying how big a deal each bug is
[13:19] <Compnn> in reality, your operating system and all of its libs are full of issues like this
[13:19] <Compnn> yes each bug is important, but you do not have security and you must do something else to protect yourself :)
[13:20] <Compnn> virtual machine maybe
[13:38] <anshul> saste, regarding #153 did you got the the answer of ur own question, what is expected output
[13:43] <saste> anshul, do you mean this: https://trac.ffmpeg.org/ticket/153 ?
[13:44] <saste> not that I know
[13:45] <anshul> yes that ticket
[13:46] <saste> anshul, i'm not sure it is even a valid ticket
[13:47] <anshul> is disscussion regarding ticket on devel-mailing list valid
[13:58] <cone-692> ffmpeg.git 03James Almer 07master:a6b8aef0dc70: configure: remove duplicate ape_decoder_select line
[13:58] <cone-692> ffmpeg.git 03James Almer 07master:607c88126c14: configure: fix g729 decoder dependencies
[13:58] <cone-692> ffmpeg.git 03James Almer 07master:5fef4afb8527: configure: add missing audiodsp dependency for ra144
[14:09] <saste> anshul, what thread?
[14:13] <anshul> saste, i dont know which thread it was, I just wanted to know that the question you asked on that ticket #153, should i bring light on that stating new thread
[14:26] <anshul> What would be an better idea if subtiltle bitmap width is larger then video width. should be call an scale filter or should be remove extra alpha spcae in start and end
[14:36] <cone-692> ffmpeg.git 03Anshul Maheshwari 07master:263932c08409: avcodec/dvbsubdec: fix ignoring blank frame by updating got_output
[15:05] <saste> anshul, i was never answered my question
[15:05] <saste> i played the file, and the issue was not apparent to me, that's why I asked
[15:05] <saste> anshul, is your patch fixing the issue?
[15:07] <anshul> my patch is not regarding that issue #153 its regarding #3737
[15:07] <saste> anshul, oh now i see, it is the "help" text message which doesn't look right at the beginning of the video, right?
[15:09] <anshul> I know about that issue its might be because not considering y =0 in yuv and trying to make it rgb
[15:09] <anshul> in dvbsub y=0 have special meaning , but dont know that is the only issue
[15:11] <anshul> when i debug the ffplay my speaker volume is much higher than i do directly ffplay, is any one else got this problem
[17:37] <michaelni> Timothy_Gu, launchpad.net/ffmpeg/ lists 2.2.3 as latest release, not sure who updated it in the past, but whoever did please update
[17:37] <Timothy_Gu> michaelni: it's automatic
[17:37] <michaelni> ahh, ok thats perfect, thanks
[17:57] <cone-849> ffmpeg.git 03Sergey 07master:6d2df3c00a78: ffmpeg: fix deadlock regression in threading error handing
[19:19] <cone-849> ffmpeg.git 03Diego Biurrun 07master:e63b818dbe60: dv: Properly split decoder and encoder initialization
[19:19] <cone-849> ffmpeg.git 03Michael Niedermayer 07master:3ac51145304f: Merge commit 'e63b818dbe6060b606ae10eb184859f8b97d2353'
[19:29] <cone-849> ffmpeg.git 03Diego Biurrun 07master:5dd8c08fd5e4: mpeg: Change ff_convert_matrix() to take an MpegEncContext parameter
[19:29] <cone-849> ffmpeg.git 03Michael Niedermayer 07master:1fd0e9a56d75: Merge commit '5dd8c08fd5e4c04d7a08d8934f0098a8a4a35c28'
[19:31] <Voicu> any idea why avcodec_decode_video2 would fail with "Invalid argument" ? AFAICT the parameters are correct
[19:32] <Voicu> the codec and packets are valid since I am already using them for something else and the frame is only suppposed to be allocated with av_frame_alloc, right?
[19:52] <Voicu> hmm, can it be from some weird timestamp?
[20:01] <thardin> woop, d10_channelcount pushed
[22:51] <cone-849> ffmpeg.git 03Michael Niedermayer 07master:8c22bea57447: avcodec/vp8: Fix multiple ff_thread_finish_setup() calls
[00:00] --- Sat Jun 28 2014
1
0
[03:50] <kingsob_> can someone help me determine why the video and audio tracks are coming out in the opposite order I want, I need video track to be #0 and audio track to be #1 --> http://pastebin.ca/2815952
[06:09] <rawrawree> Hey Everybody!
[06:10] <rawrawree> I'm running my head against a brick wall
[06:10] <rawrawree> anybody around?
[06:18] <rawrawree> I've got a command that's producing inconsistent results
[06:18] <rawrawree> sometimes it will convert a jpg to mpg
[06:18] <rawrawree> sometimes it won't
[06:58] <rawrawree> corrupted double-linked list: 0x0000000003583ec0
[06:58] <rawrawree> any idea on that?
[08:26] <anshul> Is there any way to play audio using ffplay without screen, only commandline
[08:29] <debianuser> anshul: https://ffmpeg.org/ffplay.html suggests option -nodisp
[08:29] <anshul> thanks
[12:18] <declan> have a question using the ffmpeg API, I hope this is the correct place
[12:18] <declan> i wonder if its possible to specify the size of the max_packet_size via the API
[12:19] <declan> i have large frame of 70K+, which get segement to sizes of 32768 bytes defined by the IO_BUFFER_SIZE
[12:19] <declan> is possible to increase the size of this buffer and secondly is it advisable
[12:19] <declan> thanks
[12:20] <declan> libavformat/aviobuf.c is the location for relevant source code
[13:10] <feliwir> hello, how can i make use of hardware acceleration in ffmpeg?
[13:11] <feliwir> i want my program to use the software decoder as fallback (already implemented) and use the hardware decoding when possible
[13:12] <feliwir> and the decoded frame should go into a glTexture without copying it back to the CPU, :) Help is appreciated
[13:27] <saste> feliwir, I don't think this is in general possible
[13:27] <debianuser> feliwir: klaxa|work wrote about that earlier, I'll quote:
[13:27] <debianuser> klaxa|work> there is some support for opencl i think
[13:27] <debianuser> klaxa|work> google for benchmarks, here is on with x264 http://forum.doom9.org/showthread.php?t=164960&page=11
[13:27] <debianuser> klaxa|work> although i'm not sure if this can be applied to ffmpeg
[13:27] <debianuser> klaxa|work> here is a mailing list thread http://ffmpeg-users.933282.n4.nabble.com/OpenCL-What-Effect-Does-It-Have-td…
[13:28] <saste> my understanding is that with H/W GPU acceleration the decoded frame is stored in the GPU memory
[13:28] <saste> so you need to move it to the CPU memory if you want to do something with that
[13:29] <saste> also see this: http://trac.ffmpeg.org/wiki/HWAccelIntro
[13:30] <saste> note: I'm the author of that text, feel free to improve/extend it in case you have reliable information
[13:32] <feliwir> thank you
[13:32] <Visual-Vincent> Hi! I have a question about using ffmpeg.exe with my application I am developing
[13:34] <Visual-Vincent> Am I allowed to include ffmpeg in my application's installer package so it can be used later by my application?
[13:35] <Visual-Vincent> My application is running ffmpeg.exe from command-line
[13:35] <saste> Visual-Vincent, http://ffmpeg.org/legal.html
[13:36] <feliwir> saste is there any usage example of this with ffmpeg?
[13:38] <saste> feliwir, yes, there is ffmpeg_vda/vdpau/dxv2.c
[13:38] <saste> it shows how to use ffmpeg API with H/W acceleration
[13:38] <Visual-Vincent> I've read /legal.html, but sadly I don't understand that much of it...
[13:39] <saste> feliwir, note that in that case from my memory the code is copying data from GPU to CPU
[13:39] <Visual-Vincent> I will not make any money on my application however
[13:39] <feliwir> okay
[13:39] <feliwir> where can i find the piece of code ?
[13:40] <saste> feliwir, but again, don't trust me and check it for yourself since I myself am relatively new to that stuff
[13:40] <saste> feliwir, i told you, they are .c files in the ffmpeg repository
[13:40] <feliwir> ffmpeg_vda/vdpau/dxv2.c that file isn't in the repository :-/
[13:40] <saste> Visual-Vincent, the usual restrictions about GPL/LGPL code apply
[13:41] <saste> feliwir, ffmpeg_vda.c ffmpeg_vdpau.c ffmpeg_dxv2.c
[13:41] <feliwir> ah okay
[13:44] <Visual-Vincent> Ok another question :P
[13:45] <Visual-Vincent> "Accompany it with the complete corresponding machine-readable source code" - Do I have to keep the source code with ffmpeg.exe even if I haven't modified it?
[13:45] <saste> Visual-Vincent, I think you're supposed to distribute it, yes
[13:46] <Visual-Vincent> Okay...
[13:47] <Visual-Vincent> Ah, yes I have to, it's written on /legal.html
[13:51] <Visual-Vincent> Okay thanks alot
[13:51] <Visual-Vincent> Bye
[13:53] <feliwir> saste a few more comment would be nice on that code :D
[14:23] <Visual-Vincent> Saste?
[14:44] <luc4> Hello! I read the h264 encoding guide, but it is unclear to me whether I can specify if I want a baseline, main profile etc... Is this possible?
[14:49] <debianuser> luc4: According to https://trac.ffmpeg.org/wiki/Encode/H.264 you can specify profile with -profile:v option, e.g.: -profile:v baseline -level 3.0
[14:52] <SmokeyD> hey all, I am tryingto increase the volume of a video file, without changing anything else to the video. I am trying to use the volume audio filter. But I get an error Unknown input format: 'volume=volume=1.0'
[14:53] <SmokeyD> this is the command I use (I have tried several different positions for the -f:a option: avconv -i video1_habitat_insectas.avi -f:a "volume=volume=30db:precision=fixed" -b:a 192k -c:a ac3 -c:v copy video1_habitat_insectas2.avi
[14:53] <sacarasc> The avconv channel is #libav
[14:53] <SmokeyD> ow, sorry
[14:53] <SmokeyD> I though ffmpg and libav is the same?
[14:53] <sacarasc> No, libav is a fork.
[14:53] <SmokeyD> ah, ok
[14:53] <SmokeyD> sorry, and thanks
[14:55] <luc4> debianuser: thanks, but it is unclear to me what relationship there is between profile:v and preset. Probably the profile decides when both are provided?
[14:56] <Hello71> it is codec-specific
[14:57] <Hello71> in h264, -preset, -tune, and -profile are orthogonal
[14:58] <luc4> So together they will find a setup? But assuming I set baseline, can I be sure that whatever other params I specify I'll get baseline?
[14:59] <luc4> I suppose yes as you say those are all orthogonal.
[17:37] <ajeesh> Hi
[17:38] <ajeesh> I am getting 1.wmv: Error while opening file while converting mwv files
[17:41] <ajeesh> ]# ffmpeg -i 1.wmv -f flv my-video.flv
[17:49] <ajeesh> okay .. iam sorry
[18:02] <feliwir> hm i just looked at ffmpeg_vdpau.c and that code is really hard to understand
[19:05] <brendanlong> Is there any reason reading frames from a source that FFMPEG can demux but can't decode wouldn't work (Kate subtitles in an Ogg file for example)? I don't need them decoded, I just want the raw demuxed data, but it seems like av_read_frame is returning errors (and I can't figure out why because I'm doing this in Chromium, which seems to remove ffmpeg's debug info)
[19:05] <brendanlong> I'm just wondering if this is possible and I'm just doing something wrong, or if it's not supposed to work
[19:43] <GyrosGeier> hi
[19:44] <GyrosGeier> is there an easy way to select multiple audio streams by language, leaving the automatic selection for highest quality for each language in place?
[19:45] <GyrosGeier> i.e. I have English 5.1, English 2.0 and German 2.0, and I'd like to selecxt English 5.1 and German 2.0
[19:45] <sfan5> I don't think you can do this
[19:47] <GyrosGeier> so I have to enumerate the substreams, and specify which ones I want by number?
[19:47] <sfan5> this is how I do it in the script I use to convert TV recordings
[19:49] <GyrosGeier> okay
[19:49] Action: GyrosGeier is in the process of reading all his DVDs to the NAS
[19:49] <GyrosGeier> so I don't have to get up to swap discs
[20:00] <GyrosGeier> hm
[20:00] <GyrosGeier> also, how do I specify which DVD track I want to encode?
[20:00] <GyrosGeier> wait
[20:01] Action: GyrosGeier googles first
[20:01] <sfan5> GyrosGeier: -map ?
[20:02] <GyrosGeier> as in, there are three episodes of that series on the DVD
[20:03] <GyrosGeier> and I'd like to select one
[20:03] <sfan5> I'm not sure whether you can do that with ffmpeg
[20:05] <GyrosGeier> hmm
[20:05] <GyrosGeier> how else?
[20:05] <GyrosGeier> I mean, mencoder can do that, but #mplayer told me that mencoder was deprecated
[20:05] <GyrosGeier> and I should use ffmpeg
[20:13] <GyrosGeier> ah
[20:13] <GyrosGeier> vobcopy
[00:00] --- Sat Jun 28 2014
1
0
[00:20] <cone-533> ffmpeg.git 03Michael Niedermayer 07release/1.1:4dfd4399eaca: Update for FFmpeg 1.1.12
[00:37] <cone-533> ffmpeg.git 03Michael Niedermayer 07fatal: ambiguous argument 'refs/tags/n1.1.12': unknown revision or path not in the working tree.
[00:37] <cone-533> Use '--' to separate paths from revisions
[00:37] <cone-533> refs/tags/n1.1.12:HEAD: avfilter/x86/vf_pullup: fix old typo
[01:30] <cone-533> ffmpeg.git 03Gaullier Nicolas 07master:3eae34d50fc5: avformat/mxfenc: set/force channelcount in MXF D-10
[02:09] <BBB> so the good news is that Im 4 cycles faster than gcc now (38 -> 34 on my machine)
[02:09] <BBB> the bad news is its kind of ugly and now I gotta get it working on linear and 32bit/win64 again
[02:09] <BBB> Im sort of going in circles...
[02:09] <BBB> ohwell
[02:52] <J_Darnley> For anyone who is interested in audio visualisations: http://users.telenet.be/darnley/avs.mp4
[02:52] <J_Darnley> Good night
[03:15] <cone-533> ffmpeg.git 03Diego Biurrun 07master:5ab03e41e553: x86: h264dsp: Fix link failure with optimizations disabled
[03:15] <cone-533> ffmpeg.git 03Michael Niedermayer 07master:11ba0c82076d: Merge commit '5ab03e41e553452118113d0c224fa32b325e45e5'
[03:23] <cone-533> ffmpeg.git 03Diego Biurrun 07master:593d2326ef98: dv: Replace a magic number by sizeof()
[03:23] <cone-533> ffmpeg.git 03Michael Niedermayer 07master:7e8fdf05e09a: Merge commit '593d2326ef985cdffe413df629419938f7b07c4c'
[04:47] <BtbN> Is this a problem in libavcodec or vlc? http://bpaste.net/show/408270/
[04:47] <BtbN> That's vlc with current master ffmpeg
[04:48] <BtbN> works fine with ffmpeg-2.2.3
[06:18] <michaelni> BtbN, dont know but when in doubt open a ticket on trac so that it can be kept track of, also provide all that is needed to reproduce
[06:19] <BtbN> happens each time i want to play anything that's h264 with vlc
[06:19] <BtbN> ffplay works fine
[06:23] <michaelni> since when is that happening ? any update of ffmpeg or vlc causing it ?
[06:24] <BtbN> Not sure about that, i noticed it yesterday, but i haven't used vlc in a while
[06:27] <BtbN> It's vdpau related though, if i disable hwaccell vlc plays it fine
[06:28] <BtbN> (while still using the avcodec decoder)
[06:52] <BtbN> michaelni, it's quite old aparently. I'm 30 days back in ffmpeg now, and it still happens
[06:59] Action: michaelni sleeps
[13:23] <J_Darnley> Oh my. I didn't realize just how obscure the expression compiler in this really was.
[13:24] <J_Darnley> Windows' API functions, inline asm, global variables...
[13:24] <J_Darnley> I wonder if it would be easier to replace it with something like lua
[15:40] <cone-973> ffmpeg.git 03Yao Wang 07master:e18b48c6fdfd: avcodec/cavs: fix B_SUB_DIRECT col-MB is Intra MB case.
[16:16] <Daemon404> is Micah Galizia on irc?
[16:16] <Daemon404> or is this an email affair
[16:41] <Compnn> Daemon404 : doesnt ring any bells
[16:41] <Daemon404> i emailed him
[16:53] <plepere> wow, intel sure fucked up with their AVX2 packing instructions.
[16:53] <plepere> and unpacking ones
[17:01] <plepere> I'm basically doing 3 instructions instead of 1. :/
[17:07] <nevcairiel> they didn't fuck up, its just not build for your use-case probably =p
[18:27] <megaTherion> Hello, is dshow still the prefered interface for webcam access on windows? DirectShow is pretty old, will it work on win8?
[18:28] <Compnn> ffmpeg does not have the updated windows media framework i dont think
[18:29] <Compnn> so probably not work on win8
[18:29] <Compnn> but i did not test
[18:29] <megaTherion> so if I wanted to rely on current lib's I rather use the windows api instead of DirectShow (or even older vfw)
[18:30] <Compnn> yes
[18:30] <megaTherion> thanks for the info then
[18:30] <Compnn> microsoft either killed dshow or will kill it
[18:31] <Compnn> i dont think anyone is working on new api for ffmpeg. patches are welcome
[18:31] <megaTherion> its dead afaik, since DirectX8 or so but I wont count that its still included in win8 or later
[18:31] <Compnn> right
[18:31] <JEEB> well, DShow continued to get newer stuff until quite late in the game
[18:31] <nevcairiel> directshow is still the most used framework of what microsoft offers in that regard
[18:31] <nevcairiel> outside of microsoft themself, barely anyone touches the new stuff
[18:32] <JEEB> I'm pretty much sure that during dx8 DXVA and stuff still wasn't really around
[18:32] <JEEB> although it's not really strictly DShow stuff
[18:32] <JEEB> also I'm not sure when the colorimetry stuff was added to DShow interfaces
[18:33] <JEEB> but yeah, vfw is /really/ dead (yet still included), DShow is kind of dead since Vista+ (at least MS hopes so), and MF is unused
[18:33] <nevcairiel> dxva2 is a vista thing, thats long after dx8
[18:33] <Daemon404> JEEB, dont tell that to people writing new vfw filters
[18:33] <Daemon404> even this year
[18:34] <nevcairiel> they did do one thing and did block dshow in Windows Store applications, so those have to use media foundation
[19:49] <Sergadra> well how does the irc support work here because i have some questions about the deving a code for a AVCHD .AVI Video with Dynamic Framerate to .MP4 or .M2TS
[19:50] <Sergadra> also hello first ^^
[19:52] <Sergadra> oh sorry i think im understand that false sorry have to go to the other one again
[19:53] <cone-973> ffmpeg.git 03James Almer 07master:057d2704e78b: x86/videodsp: add emulated_edge_mc_mmxext
[20:08] <cone-973> ffmpeg.git 03Kieran Kunhya 07master:9b60d9197970: hevc: Allow out of bound values for num_reorder_pics
[20:08] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:acc7cf70dc09: Merge commit '9b60d9197970658e91daf4b586397f450de9af69'
[20:16] <cone-973> ffmpeg.git 03Gildas Cocherel 07master:a0e1c3517a65: hevc: remove unused array min_cb_addr_zs
[20:16] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:932be89c26d7: Merge commit 'a0e1c3517a656dd32293f054a339e0ac73328138'
[21:01] <cone-973> ffmpeg.git 03Anton Khirnov 07master:1b04eb20f7e3: lavc: do not allocate edges in the default get_buffer2()
[21:01] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:e059ac59efb8: avcodec/snow: ensure the buffers have allocated edges
[21:01] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:a9f7972844b7: Merge commit '1b04eb20f7e3f0a71f73ba91efcc3d60a435e443'
[21:07] <cone-973> ffmpeg.git 03Anton Khirnov 07master:3f4edf012593: dump_stream: print the timebase as is, do not reduce it
[21:07] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:af75a8559969: Merge commit '3f4edf012593c73941caa0ef9b292da00225c3df'
[21:36] <cone-973> ffmpeg.git 03Anton Khirnov 07master:38d044429dab: output example: use the new AVFrame API for allocating the video frame
[21:36] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:c9083d85df46: Merge commit '38d044429dabcd28928693240e955c6d4430cca9'
[22:12] <cone-973> ffmpeg.git 03Anton Khirnov 07master:294daf71a7a1: output example: use a struct to bundle the video stream variables together
[22:12] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:8eb1174a9ed3: Merge commit '294daf71a7a1303b5ddd3cbefebed3b732d610f3'
[22:18] <cone-973> ffmpeg.git 03Anton Khirnov 07master:ac85f631c9a9: output example: set output channel layout
[22:18] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:a1c4635558cb: Merge commit 'ac85f631c9a9cc59aaca1c8dd6894fb1f701c594'
[22:50] <cone-973> ffmpeg.git 03Anton Khirnov 07master:5e7b125b6ae3: output example: use the new AVFrame API to allocate audio frames
[22:50] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:11991a7d9069: Merge commit '5e7b125b6ae36893dfd9cb5661c99b67363cbb38'
[23:07] <cone-973> ffmpeg.git 03Anton Khirnov 07master:edd5f957646d: output example: use OutputStream for audio streams as well
[23:07] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:b9bfd888e551: Merge commit 'edd5f957646dcbf1bb55718bc7bf1e5481c25bcb'
[23:35] <cone-973> ffmpeg.git 03Anton Khirnov 07master:63fd0d866c83: output example: allocate the audio frame only once
[23:35] <cone-973> ffmpeg.git 03Michael Niedermayer 07master:5a55d5e44575: Merge commit '63fd0d866c8300a8f251a15b1535e9ce40a407fb'
[00:00] --- Fri Jun 27 2014
1
0
[00:08] <ogrgkyle> Hello
[00:08] <ogrgkyle> I've set up FFmpeg + CodeBlocks
[00:08] <ogrgkyle> I can successfully compile the sample FFmpeg scripts
[00:08] <ogrgkyle> Now...
[00:09] <ogrgkyle> How can I loop through the frames and display them on screen?
[00:10] <ogrgkyle> Basically, I need to loop through the frames of the input vid and "give" them to OpenCV, which will detect certain things.
[00:11] <ogrgkyle> And then I'd like to display the video frame, with the detected boxes drawn on top
[00:41] <elisa87> is this wrong for converting mp3 to ogg? ffmpeg -i Most_Precious_Possession_Master.mp3 -acodec libmp3lame Most_Precious_Possession_Master.ogg
[00:42] <c_14> yes
[00:42] <c_14> you want -c:a vorbis
[00:42] <elisa87> c_14: were you talking to me?
[00:42] <c_14> yep
[00:46] <Hello71> that would work great if you wanted to convert from mp3 to mp3
[00:47] <c_14> Nah, pretty sure ogg won't eat mp3.
[00:47] <c_14> That'll fail.
[00:47] <c_14> Rather noisily too, iirc.
[00:47] <Hello71> maybe, i dunno if you can put mp3 in ogg
[00:47] <Hello71> nope
[00:47] <Hello71> [ogg @ 0x1f636f0] Unsupported codec id in stream 0
[00:48] <Hello71> Could not write header for output file #0 (incorrect codec parameters ?): Operation not permitted
[00:48] <Hello71> that's aac to mp3 in ogg but I'd assume mp3 -> mp3 is same
[00:48] <elisa87> what about this? Unknown encoder 'libmp3lame'
[00:48] <c_14> Your ffmpeg isn't compiled with the libmp3lame library.
[00:49] <Hello71> hm, mp3 is its own container format
[00:49] Action: Hello71 chuckles
[00:49] <c_14> elisa87: if you want ogg, you'll need to use libvorbis as the audio codec.
[00:49] <Hello71> so apparently you can put mp3 in wav
[00:49] <c_14> or vorbis, depending
[00:50] <c_14> Wasn't wav supposed to be for lossless formats?
[00:50] <Hello71> no, that's just the most common codec
[00:50] <Hello71> wav is a generic container
[00:50] <Hello71> it's just that usually people put pcm in there
[00:50] <elisa87> c_14: http://paste.ubuntu.com/7703060/
[00:51] <elisa87> any idea?
[00:51] <c_14> First of all, you're using libav not ffmpeg, either see #libav or download/compile FFmpeg.
[00:51] <c_14> ie
[00:51] <c_14> should suit you nicely
[00:51] <c_14> And it might be vorbis instead of libvorbis.
[00:51] <Hello71> vorbis is the ffmpeg one
[00:51] <Hello71> which usually outputs crap
[00:52] <c_14> The gusari static build is built with libvorbis.
[00:52] <c_14> Might use that.
[00:52] <c_14> The johnvansickle one appears to be built with it as well, so I guess it doesn't matter.
[00:54] <elisa87> this gave me error too aaron@aaron-desktop:~/Music$ ffmpeg -i Most_Precious_Possession_Master.mp3 vorbis Most_Precious_Possession_Master.ogg
[00:54] <elisa87> can you please write the correct command?
[00:55] <c_14> ffmpeg -i file.mp3 -c:a libvorbis out.ogg
[00:55] <c_14> if you got one of the aforementioned static bulids
[00:55] <c_14> *builds
[00:57] <elisa87> c_14: http://paste.ubuntu.com/7703085/
[00:58] <c_14> You're still using libav, not FFmpeg.
[00:58] <c_14> Download one of the static builds and just execute that.
[01:02] <David-A> elisa87: if the option -c:a is not recognized, you can try -acodec, but you should also specify a quality level (or bitrate) if the content is precious
[01:12] <elisa87> how can I reduce the size of an mp3 file to its half or its 3/4?
[01:12] <c_14> chop the bitrate in half or 3/4, you will, however, lose half to 3/4 of the quality even if you may not hear it.
[01:19] <David-A> elisa87: (the words "precious" and "master" in your filename suggests that you should be careful about the quality of the content)
[01:24] <David-A> elisa87: if a 50% or 75% bitrate reduction is sensible or not depends on what bitrate you start from. cuting in half from 300 to 150 kbit/s you may not hear any difference (depending on your ears and the kind of music), but going from 128 to 64 kbit can be ruining
[01:26] <David-A> elisa87: are you transcoding from mp3 to ogg because you think ogg will be more compact?
[01:48] <elisa87> naaah because of browser compatibility in my website for HTML5 audio
[01:49] <elisa87> naaah because of browser compatibility in my website for HTML5 audio David-A
[01:59] <David-A> elisa87: okay, as long as you keep the originals, nothing can go wrong :)
[02:08] <hai> hello. I'm trying to capture my desktop, precisely a full screen video - think "YouTube video on full screen situation". everything's great - got video and audio of the 'full screen video', but the output video is somewhat wobbly/laggy.. it's not right, what I'm trying to say. I think it might have to do with FPS?
[02:08] <hai> here's the command I used: ffmpeg -f dshow -i audio="Stereo Mix (Realtek High Defini":video="UScreenCapture" -r 24000/1001 -q 1 out.avi
[02:09] <hai> and I'm on Windows, if it's important. but I guess not, since the output on Linux would be the same wobbly/laggy video
[02:56] <David-A> hai: maybe it depends on the graphics driver, the graphics chip, the cpu speed, or the chosen encoding. when I grab 1600x1200 @ 25fps on xubuntu I see no lag, but lots of encoding artifacts. (probably because my cpu cannot encode in real time with high enough bitrate)
[02:57] <hai> David-A: but have you also tried recording a playing Flash video where there's movement?
[02:58] <David-A> hai: yes, but the flash player is only 640x360 pixels on the screen, so I moved around a window too.
[03:03] <hai> David-A: does it matter what FPS I'm setting when I'm starting to record? I have to specify the Flash video's FPS?
[03:06] <David-A> hai: I dont know. I suppose if capture fps is higher than display fps or flash fps, you would get duplicated frames, and if capture fps is lower you would miss video frames.
[03:07] <David-A> hai: I also suppose the processes are not synced, so it might be that if the fps's are exactly the same, there may be occational missed/duplicated frames anyways
[03:09] <Hello71> that is impossible
[03:09] <Hello71> assuming the timing is constant
[05:10] <ffmpeg953> Hey everybody
[05:10] <ffmpeg953> Got a question that I think is super simple
[05:10] <ffmpeg953> but before I destroyed my confidence I wanted to double check with you guys
[05:11] <ffmpeg953> drawtext=fontfile=/home/fonts/arial.ttf:text='.This is my text One Example'fontsize=20:fontcolor=white:x=40:y=300:borderw=4:bordercolor=black
[05:11] <ffmpeg953> Can I add another drawtext command to the same ffmpeg command?
[05:31] <relaxed> ffmpeg953: should be able to
[06:14] <ffmpeg953> relaxed what would I use to separate the two?
[06:14] <ffmpeg953> relaxed: just add another command?
[06:25] <relaxed> yes
[06:38] <ffmpeg953> sorry relaxed I keep missing you
[06:38] <ffmpeg953> relaxed do I separate with comma? I am a newb
[06:38] <ffmpeg953> relaxed: do it like this? drawtext=fontfile=/home/fonts/arial.ttf:text='.This is my text One Example'fontsize=20:fontcolor=white:x=40:y=300:borderw=4:bordercolor=black, drawtext=fontfile=/home/fonts/arial.ttf:text='.This is my text Two Example'fontsize=20:fontcolor=white:x=40:y=300:borderw=4:bordercolor=black
[06:38] <relaxed> https://trac.ffmpeg.org/wiki/FilteringGuide
[06:40] <ffmpeg953> it doesn't show multiple filters does it? I just read through it
[06:43] <relaxed> reread Filtergraph,Chain,Filter relationship
[06:43] <relaxed> but yes, use a comma
[06:45] <ffmpeg953> I'm using 2.2.3
[06:45] <ffmpeg953> and I don't ever use the [in] command
[06:46] <ffmpeg953> http://pastie.org/9325127
[06:46] <ffmpeg953> this is how my command looks right now
[06:47] <ffmpeg953> Can you show me proper format?
[06:48] <ffmpeg953> the "chains" you referenced use [in]
[06:48] <relaxed> -filter:v filter=blah,filter=bleh
[06:52] <ffmpeg953> running command now
[06:52] <ffmpeg953> it hasn't kicked me out yet, lol
[06:57] <ffmpeg953> nope, it failed :(
[13:54] <vtorri> hey
[13:54] <vtorri> in http://www.ffmpeg.org/general.html, i can see :
[13:54] <vtorri> HEVC X (for encoding
[13:55] <vtorri> and in the remarks :
[13:55] <vtorri> encoding supported through the external library libx265
[13:55] <vtorri> shouldn't it be E instead of X for encoding ?
[13:55] <vtorri> (if the remarks is still valid)
[13:56] <sacarasc> Should what be E instead of X?
[13:57] <vtorri> X :)
[13:57] <vtorri> HEVC E
[13:57] <vtorri> instead of
[13:57] <vtorri> HEVC X
[13:57] <vtorri> look at the link...
[13:58] <sacarasc> I can't see "HEVC X" on that page at all.
[13:59] <vtorri> search just HEVC then...
[13:59] <vtorri> the copy past is wrong
[13:59] <sacarasc> Oh, that X is just showing that it can be done.
[13:59] <sacarasc> It's a check mark.
[13:59] <vtorri> because of the html table in the link, i guess
[13:59] <vtorri> no
[14:00] <vtorri> it means it can be done with libavcodec
[14:00] <vtorri> but the remarks says it's through libx265
[14:00] <sacarasc> So it does.
[14:00] <vtorri> if an external lib is used, then it should be E
[14:00] <vtorri> then the remark is wrong
[14:22] <relaxed> ffmpeg -codecs shows DEV.L. hevc
[15:40] <ffmpeg953> If you don't know the image size of an input, what is the command to make an image 640x480
[16:42] <Sashmo> has anyone been able to get VDPAU to work with FFMPEG and get a noticeable increase in encoding?
[16:42] <klaxa|work> vpdau is not for encoding
[16:42] <Sashmo> oh..... really?
[16:43] <Sashmo> what is it for then?
[16:43] <klaxa|work> decoding
[16:43] <Sashmo> ah....
[16:43] <Sashmo> can ffmpeg support GPU encoding?
[16:43] <Sashmo> getting mixed answers in google
[16:44] <klaxa|work> there is some support for opencl i think
[16:44] <Sashmo> yeah thats what I saw, but is there a noticeable increase?
[16:46] <klaxa|work> depends on your definition of noticeable
[16:46] <klaxa|work> google for benchmarks, here is on with x264 http://forum.doom9.org/showthread.php?t=164960&page=11
[16:47] <klaxa|work> although i'm not sure if this can be applied to ffmpeg
[16:47] <klaxa|work> here is a mailing list thread http://ffmpeg-users.933282.n4.nabble.com/OpenCL-What-Effect-Does-It-Have-td…
[16:56] <dirty_d> Hey, is it possible to just change the H.264 level with ffmpeg without re-encoding? I have some low-bitrate videos foolishly set to level 5.2. It seems Apple devices just look at the level and give up, because manually hex editing the level to 4.1 makes the video play fine.
[16:56] <dirty_d> not sure if there is a more sane way to fix them
[16:57] <sacarasc> No, you can't.
[17:11] <dirty_d> hmm, if i just set a similar bitrate and the level it re-encodes at like 400fps
[17:11] <dirty_d> how could that be?
[17:11] <dirty_d> encoding with libx264 is usually much much slower for me
[17:12] <Sashmo> klaxa|work: thanks for the links.
[17:21] <brontosaurusrex> dirty_d, whats the resolution of the originals?
[17:22] <dirty_d> 426x240
[17:22] <brontosaurusrex> well, thats like lowres nowadays, so 400 fps looks good to me
[17:22] <brontosaurusrex> and you are not doing any scaling i suppose
[17:22] <brontosaurusrex> *re-scaling
[17:23] <dirty_d> hmm
[17:24] <dirty_d> well what option would I use to keep all the codec parameters exactly the same as the original video except change the level?
[17:49] <xlinkz0> hey, i'm having trouble with a complex filtergraph that should mix two videos side by side
[17:50] <xlinkz0> here's my report : http://sprunge.us/NGNC
[17:50] <xlinkz0> and how i build the filtergraph ( for clarity ) : http://sprunge.us/cAVQ
[18:00] <xlinkz0> smh.. filter_complex not "filter complex"
[18:00] <xlinkz0> my bad
[18:08] <xlinkz0> can I concatenate and cut videos using filters?
[18:51] <xlinkz0> say I have 4 videos, ( a, b, c, d). I want to concat a+b and c+d and play them side by side
[18:51] <xlinkz0> so a+b , c+d should be played side by side
[18:51] <xlinkz0> i've tried this : http://codepad.org/EcVn4nYn
[18:52] <xlinkz0> however the output is "a, a"
[18:52] <xlinkz0> that is, the first video is played side-by-side with itself
[18:54] <xlinkz0> JEEB: please.. help a poor guy out :D
[18:54] <JEEB> > implying I have any idea about the filter_complex stuff
[18:54] <xlinkz0> :(
[18:55] <xlinkz0> well i guess it's not called _compled for nothin
[18:55] <xlinkz0> *complex
[18:55] <JEEB> the syntax is something I never felt like teaching myself since everything I wanted to do was possible by just chaining stuff in -vf
[18:55] <xlinkz0> i .. completely understand :p
[19:00] <sfan5> xlinkz0: You could try concatting a+b & c+d seperately and then doing it how it is written in https://lists.ffmpeg.org/pipermail/ffmpeg-user/2013-June/015662.html
[19:02] <xlinkz0> sfan5: i can merge two files with overlay.. the problem arises when I try to add the concat filter
[19:02] <xlinkz0> sfan5: by separately you mean different ffmpeg calls?
[19:02] <sfan5> yes
[19:03] <xlinkz0> i'd rather not do that since it would require transcoding twice
[19:03] <xlinkz0> and.. i kind of need to do it in bulkk
[19:03] <xlinkz0> i do need accurate seeking
[19:07] <xlinkz0> ubitux: have you got some spare time? .. or could you please send me to someone else I can pester for a few minutes about this complex filtergraph? :D
[19:08] <jleclanche> using ffmpeg 2.2.4, the following command completely hangs and is unkillable even with kill -9. ffmpeg -f v4l2 -i /dev/video0 output.mkv --- this is an strace output, it hangs right at the end there. http://dpaste.com/2JCN4H0
[19:08] <jleclanche> any idea?
[19:09] <sfan5> kernel bug?
[19:09] <jleclanche> likely, since its unkillable
[19:09] <sfan5> or maybe I/O wait
[19:09] <jleclanche> "mpv tv:///" works fine, however when I execute that ffmpeg command, mpv tv:/// will also hang
[19:09] <sfan5> you can't kill that either
[19:09] <jleclanche> yes, if i reboot it'll say "/home is busy"
[19:10] <jleclanche> mpv tv:/// when it actually works says [tv] v4l2: your device driver does not support VIDIOC_G_STD ioctl, VIDIOC_G_PARM was used instead.
[19:10] <jleclanche> i dont know if thats related
[19:11] <jleclanche> This all used to work a few months ago
[19:11] <jleclanche> maybe new gcc, new linux or new ffmpeg has the issue, who knows
[19:11] <sfan5> I don't think it's gccs fault
[19:12] <jleclanche> this is on arch linux fwiw, dunno if anyone else has the issue
[19:18] <checkit> Hey guys, I haven't had to use ffmpeg for sometime now but a project came up recently that I think might make good use of it. I need to capture a visitors audio submission as part of a song but I'm not sure how to go about it. Any guidance here?
[19:46] <Sergadra> hello
[20:36] <woof-woof> Can anyone explain to me why this happens with ffmpeg? > http://woof-woof.site11.com/files/sample.png
[20:39] <xlinkz0> woof-woof: different encoder options
[20:40] <woof-woof> ?
[20:40] <xlinkz0> what don't you understand from that statement?
[20:40] <woof-woof> What options?
[20:41] <xlinkz0> https://trac.ffmpeg.org/wiki/Encode/H.264
[20:41] <woof-woof> That's not h264
[20:41] <ChocolateArmpits> Has anyone compiled ismindex here before?
[20:42] <woof-woof> I just downscaled the image...
[20:42] <xlinkz0> !testcase
[20:42] <xlinkz0> well.. post your full command
[20:43] <woof-woof> http://paste.ubuntu.com/7707305/
[20:46] <woof-woof> No.!
[20:51] <woof-woof> That happens when videos are downscaled too.
[20:56] <klaxa> woof-woof: so i'm not exactly sure what i'm supposed to see, but i think you are talking about downscaling of the images makes it appear blockier in ffmpeg than in gimp?
[20:58] <woof-woof> klaxa, Those details are smoother in GIMP. I just want to know if ffmpeg can do it similar. Because it would be visually better.
[20:58] <klaxa> if so, it entirely depends on the video filter used. the scale video filter has different options, look at those, same goes for gimp
[20:59] <klaxa> see: https://www.ffmpeg.org/ffmpeg-filters.html#scale-1
[21:00] <klaxa> huh? wait... i thought it had more options...
[21:08] <woof-woof> klaxa, I found it!
[21:09] <woof-woof> The option to use is "full_chroma_int" (full chroma interpolation)
[21:09] <woof-woof> Now it looks the way I wanted!
[21:09] <woof-woof> Thanks!
[21:34] <ffmpeg953> omg I can't find a good watermark command for 2.2.3 at all
[21:34] <ffmpeg953> any one have one?
[21:58] <t4nk590> hi, i'd like to hardsub an mkv file that has multiple ssa subtitle streams. is there a way to pick which stream to use? i'm currently using -vf subtitles=$FILE
[22:00] <t4nk590> the :si=3 option does not seem too work as i get an "Error initializing filter 'subtitles' with args" error
[00:00] --- Fri Jun 27 2014
1
0
[01:01] <cone-96> ffmpeg.git 03Lukasz Marek 07master:a30f1b15c7ee: doc/build_system: document examples and config targets
[01:17] <BBB> jamrial: ok so I fixed the asm to be same speed as gcc-4.8.3 inline for 32bit
[01:17] <nevcairiel> \o/
[01:17] <jamrial> swr? cool
[01:17] <BBB> jamrial: your 64bit asm dump wasnt 100% useful because it works only on win64, not unix64& Id be very interested in a unix64 asm dump from gcc-4.8.3 so I can compare closer
[01:18] <BBB> I pushed to github, please test if you get a chance
[01:18] <jamrial> i think the win64 asm dump was from J_Darnley
[01:18] <BBB> oh, right
[01:18] <jamrial> i can give you a gcc 4.7.x or a 4.9.0 unix64 dump if it's useful
[01:18] <BBB> J_Darnley: can I get unix64 asm dump from gcc-4.8.3 also?
[01:18] <BBB> is it faster?
[01:19] <jamrial> well, the 4.7.x unix64 gave me those awful results i mentioned the other day. the 4.9.x unix64 was only slightly slower
[01:20] <jamrial> awful results = bogus, probably
[01:20] <BBB> hm& so apparently gcc-4.8.3 has the great touch
[01:20] <BBB> ok lets wait for that one then
[01:20] <BBB> Im just hoping theres something I can run locally to confirm speed against
[01:20] <BBB> the 32bit asm dump was quite useful for that
[01:20] <J_Darnley> BBB: not from me
[01:20] <BBB> hm...
[01:21] <J_Darnley> perhaps I should clarigy that...
[01:21] <BBB> J_Darnley: the asm dumps were from you and were gcc-4.8.3 right?
[01:21] <J_Darnley> yes
[01:21] <jamrial> ah wait, you mean a dump for the current inline version? if so then give me a moment, because for both 4.7 and 4.9 inline was faster than the yasm port
[01:22] <J_Darnley> I cannot give you a unix64 disassebly because my compiler cannot produce "unix" code
[01:22] <BBB> jamrial: ah yes
[01:22] <jamrial> ok, sorry for the confusion
[01:23] <BBB> also if you can confirm theyre still faster, thatd be great
[01:23] <BBB> but poking me the asm is fine also
[01:23] <BBB> then Ill compare myself
[01:23] <J_Darnley> Sorry, was that request direct at me?
[01:23] <J_Darnley> *directed
[01:24] <BBB> no, at jamrial
[01:25] <BBB> if you want to help testing speed, thatd be useful (I prefer to confirm across a bunch of systems that the yasm version is not slower)
[01:25] <jamrial> http://pastebin.com/Ez2fSdiV unix64 4.9.0 (with golomb bug fix backported from gcc-4.9 branch, but that's not relevant for this i guess)
[01:26] <BBB> thanks!
[01:26] <jamrial> i'll test what you uploaded to github and compare again later
[01:30] <BBB> crap your syntax is the inverse of what j_darnley gave me
[01:30] <BBB> J_Darnley: what did you use to dump the asm?
[01:31] <nevcairiel> with inverse oyu mean the operand order? thats weird, msvc tools should use the same as gcc
[01:32] <nevcairiel> this dump looks the wrong side up =p
[01:35] <jamrial> BBB: this better? http://pastebin.com/hS1HFPjT
[01:35] <jamrial> Intel instead of AT&T
[01:35] <BBB> yes!
[01:35] <jamrial> cool :P
[01:37] <nevcairiel> ah that looks better
[01:37] <J_Darnley> BBB: objdump -M intel
[01:37] <J_Darnley> ... too slow
[01:49] <BBB> nope that still beats me by a lot
[01:49] <BBB> 4 cycles/sample :(
[01:49] <BBB> bah
[01:49] <BBB> 3.5 really
[01:49] <BBB> still
[01:50] <BBB> must be register pre-loading& so I guess I have to do that next then
[01:50] <BBB> bah
[01:50] <BBB> ohwell
[02:00] <jamrial> BBB: i'm going to test using the measure macros you have on github, see if they help (i was getting lots of skips with the normal timer.h ones sometimes)
[02:00] <jamrial> just noticed them
[05:10] <debianuser> Hello. Got one more concat bug. Command `ffmpeg $(ls -1 ???.wav | sed 's/^/-i /') -filter_complex concat=n=399:v=0:a=1 0.wav` freezes with zero CPU/disk usage, it just does nothing. Output: pastebin.com/VZYew6f7 Files: ge.tt/1Eg60nl1/v/2 (150KB)
[05:17] <debianuser> Is that my build is buggy?
[05:26] <debianuser> Or have I hit some buffer overflow with my files?
[05:36] <BtbN> i'd use the concat input
[05:49] <debianuser> I can't. Files have different frame rates mono/stereo. I was told that "-f concat" can't handle those.
[07:55] <ubitux> is there an expected processing between ff_ref_[fi]dct()?
[07:56] <ubitux> it seems i can't get block==idct(fdct(block))
[08:59] <kurosu> depending on the dct, sometimes some rescaling is left to the quantization
[09:00] <kurosu> don't know about the _ref_ version, but I would try various rescaling after fdct or idct
[09:00] <kurosu> (by a constant)
[09:05] <ubitux> i tried to scale by n bits after fdct but it didn't seem to be help
[09:05] <ubitux> i'm still getting weird negative numbers in the output
[09:08] <ubitux> mmh maybe that's because i badly adjusted to 16-bit that code actually
[09:11] <ubitux> 16-width*
[10:16] <ubitux> ok, found my mistake
[10:17] <ubitux> anyway, the ref dct has a weird bias
[10:17] <ubitux> http://pastie.org/pastes/9321987/text is this expected?
[10:18] <ubitux> and adjusted for 16px: http://pastie.org/pastes/9321988/text
[10:24] <ubitux> seems i get better result if i scale before rounding in the fdct
[10:25] <ubitux> but still not perfect
[10:30] <ubitux> or just rounded scaling
[10:34] <ubitux> so basically, in dctref, instead of doing block[i + j] = floor(tmp + 0.499999999999) in fdct (and later a -rounded or not- scaling by N), i'm doing block[i + j] = lrint(tmp / BSIZE) (BSIZE being 1<<N, so 8 or 16)
[10:35] <ubitux> and i'm getting this: http://pastie.org/pastes/9322044/text
[10:35] <ubitux> i have no idea where the other inaccuracy comes from
[10:35] <ubitux> any ideas?
[10:37] <ubitux> (code: http://pastie.org/9322056 - based on dctref)
[10:47] <debianuser> Hello. I have something that looks like concat bug. Command `ffmpeg $(ls -1 ???.wav | sed 's/^/-i /') -filter_complex concat=n=399:v=0:a=1 0.wav` freezes with zero CPU/disk usage, it just does nothing. Output: pastebin.com/VZYew6f7 Files: ge.tt/1Eg60nl1/v/2 (150KB)
[10:47] <debianuser> Is that my build is buggy? Or have I hit some buffer overflow with my files?
[11:19] <plepere> wow, avx2 pack is retarded : if you want to do a 256b pack, you need to shuffle afterwards. ;.
[11:19] <plepere> :/
[11:37] <J_Darnley> There's always something retarded about x86 simd instructions
[11:39] <J_Darnley> I find the lack of some operations at some data sizes to be particularly annoying
[12:57] <michaelni> ubitux, fdct+idct isnt really guranteed to give the same integers you started with due to finite precission and rounding
[13:20] <ubitux> michaelni: right, but it seems i'm able to get exact in/out with what doing some naive double av_rdft
[13:20] <ubitux> and so i was wondering why the reference wasn't exact at all
[13:21] <ubitux> a better scale/rounding within the idct really helps
[13:21] <ubitux> but not so much
[13:22] <michaelni> the idct is specified to be as it is, if you intend to use it in mpeg/h26*
[13:22] <michaelni> the fdct could be changed in any way we see fit
[13:23] <ubitux> i was talking about the fdct
[13:23] <ubitux> block[i + j] = floor(tmp + 0.499999999999);
[13:23] <ubitux> this weird rounding
[13:23] <ubitux> i changed that to block[i + j] = lrint(tmp / 8)
[13:24] <ubitux> (so including the scaling within)
[13:24] <ubitux> i get way better results, but still not perfect
[13:24] <ubitux> i wonder why i have not a single diff with a the double av_rdft_calc
[13:25] <michaelni> because its double ?
[13:25] <ubitux> dctref is double
[13:25] <michaelni> but its output is int16
[13:25] <ubitux> (i meant av_dct_calc sorry)
[13:25] <ubitux> ah, mmh yeah right
[13:26] <ubitux> ok
[13:26] <michaelni> also tmp/8 may have effects on trellis performance
[13:26] <ubitux> can you elaborate?
[13:27] <ubitux> a post scaling with something like (lrint(tmp)+4)>>3 doesn't give a better accuracy than lrint(tmp/8) here
[13:27] <ubitux> i have more mismatches
[13:28] <ubitux> ah, i missed the "performance" ok
[13:28] <ubitux> note that dctref is only used in dct-test
[13:28] <ubitux> so it shouldn't matter much
[13:28] <ubitux> and speaking of this
[13:29] <ubitux> in dct-test, a see no rounding in the scaling:
[13:29] <ubitux> for (i = 0; i < 64; i++)
[13:29] <ubitux> block[i] >>= 3;
[13:29] <michaelni> we quantize the values after the fdct, and higher precission should allow higher precission distortion calculation
[13:29] <michaelni> hmmm
[13:29] <ubitux> i'm gonna write a patch
[13:30] <michaelni> also block[i + j] = floor(tmp + 0); in fdct and (...+round)>>3 later should give the same results
[13:30] <ubitux> http://pastie.org/9322657
[13:30] <ubitux> basically i mean this ^
[13:31] <ubitux> and no, having block[i]>>=3 changed so it has rounding doesn't give as good result as rounding within ff_ref_fdct()
[13:31] <ubitux> but yeah it's better
[13:36] <ubitux> michaelni: so basically currently i have this: http://pastie.org/9322666
[13:36] <ubitux> that is, with the floor(x + 0.4999...) and scaling >>= 3
[13:36] <ubitux> if i do a good rounding, that is x+4>>3
[13:36] <michaelni> http://pastie.org/9322657 <-- breaks dct-test
[13:36] <ubitux> i get way better result: http://pastie.org/9322668
[13:37] <ubitux> and if i removing the scaling completely and integrate within fdct (that is lrint(tmp/8)), i get the best results: http://pastie.org/9322667
[13:38] <ubitux> michaelni: yeah, haven't tested properly, that's just so you get the idea
[13:40] <michaelni> floor(x/8+0.5) and floor(x) with later +4>>3 should give the same results
[13:41] <ubitux> doesn't seem to, but maybe i'm doing something wrong
[13:42] <michaelni> so for what value of x does it differ ?
[13:43] <michaelni> anyway like i said we can change the fdct in any way if it improves things
[13:44] <ubitux> any idea why we had floor(x+0.499999...) instead of floor(x+0.5) btw?
[13:44] <ubitux> i'd use lrint(x) anyway for this but well
[13:50] <michaelni> the 0.49 comes from some referenc from iso/itu i think but would have to check
[13:50] <ubitux> ok
[13:51] <michaelni> or at least it was there as long as i can remember
[13:51] <michaelni> mateo`, is the new iteration of the "set/force channelcount in MXF D-1" patch ok to push ?
[14:03] <Gizmoomzig> Hello, I am trying to get certain information out of the H264 decoder. One of them is the ei_flag but i cant seem to find it. Can someone please point me in the right direction where i can find the error indicator if error concealment has to be used on a macroblock?
[14:39] <plepere> michaelni, should I re-do a patch removing the mmxext function prototype changes ?
[14:41] <michaelni> plepere, if you have time yes, otherwise i can do it
[14:42] <plepere> I'm busy on avx2 mc functions, but I'll do it probably tomorrow. :)
[14:50] <Luigi12_work> hello. I'm preparing updated ffmpeg packages for Mageia. http://ffmpeg.org/security.html for 1.1.x is up to date as of 1.1.9. Do 1.1.10 and 1.1.11 fix any further CVEs?
[14:53] <michaelni> Luigi12_work, i suggest you wait a day or 2 before updating from packages from branch theres some backports pending
[14:54] <michaelni> packages from 1.1 branch
[14:54] <Luigi12_work> michaelni: ok thanks. We have 1.1.x in Mageia 3 and 2.0.x in Mageia 4.
[14:56] <michaelni> ok, didnt know, maybe someone should start a wiki page on who uses what
[14:56] <Luigi12_work> not a bad idea :o)
[14:58] <Luigi12_work> michaelni: are these backports pending for 2.0.x as well?
[14:59] <michaelni> yes, ill try to get them out ASAP
[14:59] <Luigi12_work> michaelni: thanks again
[15:06] <cone-910> ffmpeg.git 03plepere 07master:942e22c65116: avcodec/x86/hevc: add avx2 dc idct
[15:07] <cone-910> ffmpeg.git 03plepere 07master:9ba6b17add28: avcodec/x86/hevc_idct: fix number of sse registers
[15:07] <plepere> oh thanks, michaelni. :D
[15:07] <michaelni> np
[15:26] <Daemon404> btw not sure if i ever posted this here: https://gist.github.com/dwbuiten/7101755
[15:26] <Daemon404> useful for fuzzing ffmpeg
[16:10] <michaelni> no, i think ive not seen that before, thanks
[17:01] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:91d024d75dce: avutil/timestamp: Warn about missing __STDC_FORMAT_MACROS for C++ use
[17:01] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:9933e06595c0: swscale/x86/swscale: fix missing xmm clobbers in yuv2yuvX_sse3()
[17:01] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:d581567e0948: avcodec/utils: fix sizeof(AVFrame) dependence in avcodec_encode_audio2()
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:6387aa94d664: avcodec/h263dec: Fix use of uninitialized memory from the bitstream buffer
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:f65501f80cbc: ffmpeg: dont call exit_program() from a signal hander
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:673f679c8ae2: avformat/mp3enc: use av_copy_packet()
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:3ae71dd33ed3: avcodec: Add padding after the remaining AVFrames
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:36fc9bff08f9: swscale/swscale: fix integer overflow
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:102df43f49cb: dox/scaler:fix bicubiclin typo
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:b556432f5a75: avcodec/x86/mpegvideoenc_template: fix integer overflow
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:c7f419efd1df: avcodec/h264_mp4toannexb_bsf: prepend global headers before any in stream parameter sets
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:02055da69ace: avcodec/libx264: move where x264opts is applied down so it isnt overridden by avctx & defaults
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:5ac53d07a276: avcodec/h264: clear cur_pic structure instead of duplicating it in ff_h264_update_thread_context()
[17:02] <cone-910> ffmpeg.git 03Lukasz Marek 07release/2.0:c61c66cfb17e: lavu/opt: validate range before dereference
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:5805c0c2b6cf: avcodec/wma: use av_freep(), do not leave stale pointers in memory
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:491e1ec95db4: avcodec/msrle: check return code for success before use
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:1c321b7c31aa: swresample/resample: Limit filter length
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:c9ed7a4fc13c: swresample/dither: use av_malloc_array()
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:e315549ffa16: swresample/resample: use av_malloc_array() where appropriate
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:218c853774a9: swscale/x86/swscale_template: loose hardcoded dstw_offset
[17:02] <cone-910> ffmpeg.git 03Anthoine Bourgeois 07release/2.0:853bbe1aed3b: avcodec/dirac_arith: Fix build with PIC and stack-check options
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:2ca68b98ba49: avcodec/x86/idct_sse2_xvid: fix non C99 inline function
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:690a3c4fb7c9: swscale/swscale: fix srcStride/srcSlice typo
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:6f6f5f9c8b3d: avcodec/mjpegen: Fix declared argument size
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:ed38ed3b629f: avformat/mpegts: Remove redundant check
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:8162b69091b6: avcodec/diracdec: fix undefined behavior with shifts
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:66ca450fcb07: avcodec/g723_1: add assert to help static code analyzers
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:98805cbac5a4: avfilter/f_select: fix loss of precission in SAD calculation
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:6882a04706b9: avfilter/vf_deshake: fix loss of precission with odd resolutions
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:4d9aad95b517: iavcodec/vc1dec: Fix missing {}
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:eaa6d7910545: avfilter/filtfmts: Support dynamically allocated in/outputs
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:d3bcc9cfa9fc: avformat/mux: Check for and remove invalid packet durations
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:ee282be99b9d: avformat/h263dec: Fix h263 probe
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:0be9a2e9d9bb: swresample: fix AV_CH_LAYOUT_STEREO_DOWNMIX input
[17:02] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:db33fb1c4866: ffmpeg_filter: fix pointer to local outside scope
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:d9d66d2d16fa: avcodec/mjpegdec: Fix undefined shift
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:99d4d3532fac: avfilter/graphdump: Fix pointer to local outside scope
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:0e18480fc3b3: sws: dont use the optimized 410->420 unscaled conversion when height%4
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:62b9c99aec0f: avcodec/diracdec: move mc buffer allocation to per frame
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:332777970c8b: avutil/cpu: force mmx on selection of higher x86 SIMD features
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:bdd2f7c7a761: avcodec/libvorbisenc: dont add the duration to AV_NOPTS_VALUE
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:2f1f4070446e: avcodec/aac: fix () in IS_CODEBOOK_UNSIGNED macro
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:db6f3be9418c: avcodec/golomb-test: fix () in EXTEND() macro
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:a83304814c28: avcodec/h264: fix () in macros
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:4965da92eff9: avcodec/ivi_dsp: add some missing () to macros
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:a144d305dd65: avcodec/mlpdec: fix () in MSB_MASK() macro
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:81a2740919ae: avcodec/mss34dsp: fix () in SOP* macros
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:3c169251eec2: avcodec/mss4: Fix () in MKVAL() macro
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:9fe7e450305c: avcodec/aic: fix quantization table permutation
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:a2b4e38729ca: avformat/flvenc: Do not allow creating h263/mpeg4 in flv without unofficial format extensions being enabled.
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:b94621ac6ba6: avcodec/alsdec: Clear MPEG4AudioConfig so that no use of uninitialized memory is possible
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:b93659db5bbe: avformat/cavsvideodec: Fix probing when the file extension is avs
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:01e2dd53a067: cavsdec: fix qp fixed slice handling
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:1e9ae5dbbd67: avformat/mpc: attempt to allocate a packet that is not smaller than the data inside it
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:083d0b7b2d92: avutil/lzo: Fix integer overflow
[17:03] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:4a237657047b: avutil/lzo: add asserts to be double sure against overflows
[17:15] <cone-910> ffmpeg.git 03Michael Niedermayer 07release/2.0:205e2264c3d5: update for FFmpeg 2.0.5
[18:24] <cone-910> ffmpeg.git 03Michael Niedermayer 07master:1ace0ca60ff8: avcodec/x86/hevc_idct: fix function name in comment
[18:24] <cone-910> ffmpeg.git 03Michael Niedermayer 07master:6dffc8f5aaab: avfilter/vf_pullup: use ptrdiff_t as stride argument for dsp functions
[18:24] <cone-910> ffmpeg.git 03Michael Niedermayer 07master:b8255a4c7096: avfilter/x86/vf_pullup: fix old typo
[22:23] <cone-533> ffmpeg.git 03Michael Niedermayer 07fatal: ambiguous argument 'refs/tags/n2.0.5': unknown revision or path not in the working tree.
[22:23] <cone-533> Use '--' to separate paths from revisions
[22:23] <cone-533> refs/tags/n2.0.5:HEAD: avfilter/x86/vf_pullup: fix old typo
[23:15] <cone-533> ffmpeg.git 03Michael Niedermayer 07release/1.1:1f2f6b247aaf: avutil/lzo: Fix integer overflow
[23:15] <cone-533> ffmpeg.git 03Michael Niedermayer 07release/1.1:574bf2ce4ccd: avutil/lzo: add asserts to be double sure against overflows
[23:15] <cone-533> ffmpeg.git 03Luca Barbato 07release/1.1:503322f97c5a: lzo: Handle integer overflow
[23:15] <cone-533> ffmpeg.git 03Michael Niedermayer 07release/1.1:a8dd7fe5b994: Merge commit '503322f97c5a25a020933ed4ab510697d5f5b4af' into release/1.1
[23:35] <cone-533> ffmpeg.git 03Michael Niedermayer 07release/1.1:4533e8b30a38: avcodec/diracdec: move mc buffer allocation to per frame
[23:35] <cone-533> ffmpeg.git 03Michael Niedermayer 07release/1.1:e68114868ffd: avutil/cpu: force mmx on selection of higher x86 SIMD features
[23:35] <cone-533> ffmpeg.git 03Michael Niedermayer 07release/1.1:bd2266975cbd: avcodec/libvorbisenc: dont add the duration to AV_NOPTS_VALUE
[23:35] <cone-533> ffmpeg.git 03Michael Niedermayer 07release/1.1:80588930fffb: avformat/flvenc: Do not allow creating h263/mpeg4 in flv without unofficial format extensions being enabled.
[23:35] <cone-533> ffmpeg.git 03Michael Niedermayer 07release/1.1:71a4a9183f3e: avcodec/alsdec: Clear MPEG4AudioConfig so that no use of uninitialized memory is possible
[23:35] <cone-533> ffmpeg.git 03Michael Niedermayer 07release/1.1:f0d56d4f5542: avformat/mpc: attempt to allocate a packet that is not smaller than the data inside it
[00:00] --- Thu Jun 26 2014
1
0
[00:00] <_aeris_> for 4h video, not able to run this command :P
[00:04] <debianuser> Maybe something like extracting audio, joining audio, concatting video, and muxing them together would work for you...
[00:05] <debianuser> Or maybe (since your second audio is actually silence) you can join videos, and insert the audio with nonzero start pts so that it started where you need it to...
[00:06] <debianuser> anyway I have no idea how to do that. :)
[00:11] Action: debianuser looks at `-itsoffset offset` option.
[00:15] Action: debianuser now looks at https://trac.ffmpeg.org/ticket/1349 - The -itsoffset option does not work the way you think it does.
[00:39] <debianuser> _aeris_: This had just worked for me too: ffmpeg -filter_complex aevalsrc=0:d=3 -ar 24000 -ac 1 0.ogg; ffmpeg -i "concat:0.ogg|mono24kHz.ogg" -c copy cat.ogg; mplayer cat.ogg
[01:03] <Technicus> Hello, I am looking for suggestions of Linux compatible video capture cards that have HDMI input, and connect to USB 3.0?
[01:04] <Hello71> 1. i don't think we do hardware suggestions
[01:04] <Hello71> 2. pick anything that's uvc and has hdmi input
[01:04] <Technicus> Hello71: I am just asking.
[01:06] <Technicus> Given the nature of FFmpeg I suspect that someone may have experience with such devices. Maybe I am totaly wrong . . .
[01:08] <Hello71> just pick anything that supports uvc
[01:08] <Hello71> which should be almost all usb devices
[01:23] <Technicus> Hello71: Thanks.
[03:36] <debianuser> What are the ranges for XX in -aq XX/-q:a XX/-qscale:a XX? Also do I understand it correctly, that all three options are just aliases?
[03:41] <relaxed> debianuser: yes, and the ranges depend on the encoder
[03:41] <debianuser> Is there a way to "ask" decoder for its ranges?
[03:42] <relaxed> you mean encoder
[03:42] <relaxed> which one are you using?
[03:42] <debianuser> libvorbis right now
[03:43] <debianuser> But I'm also curious how to find it for other encoders to
[03:43] <relaxed> man oggenc | less +/' -q'
[03:44] <relaxed> look at man lame for its ranges
[03:44] <debianuser> what about e.g. fdkaac?
[03:45] <debianuser> or for ffmpeg's own vorbis/aac encoders?
[03:48] <relaxed> libfdk_aac uses -vbr, see "ffmpeg -h encoder=libfdk_aac"
[03:49] <relaxed> do not use ffmpeg's vorbis encoder
[03:49] <relaxed> I don't think the native aac encoder suports vbr
[03:50] <relaxed> you should be using libvorbis or libfdk_aac
[03:51] <debianuser> I think I've got the idea. Thank you!
[04:04] <relaxed> you're welcome
[04:13] <Hello71> info for popular codecs is also on trac wiki
[04:13] <Hello71> e.g. https://trac.ffmpeg.org/wiki/vpxEncodingGuide
[04:25] <wookienz> hi guys, just trying to use ffmpeg to re encode to aac and h264. The resulting file appears to be a lot lower bit rate. What flags can i use to ensure that the but rate is as high as possible? thanks.
[04:36] <debianuser> wookienz: You mean the file has lower bitrate (i.e. it's smaller) but still a good quality?
[04:38] <wookienz> debianuser, medininfo shows a lower bitrate, i just assume it is a lower quality. im a noob when it comes to this!
[04:42] <debianuser> Modern codecs are very complex and had lots of settings. The same codec with the same bitrate may have much better quality depending on its other settings. And the other way around: same quality may be achieved on different bitrates depending on used encoding profiles and other options.
[04:43] <wookienz> hmm ok. back to square one.
[04:45] <debianuser> A simple encoding command: ffmpeg -i in.avi -c:v libx264 -c:a libfdk_aac out.mp4
[04:45] <debianuser> can be tuned with lots of different options: ffmpeg -i in.avi -c:v libx264 -crf 22 -r 25 -preset veryslow -tune grain -profile:v baseline -level 3.0 -pix_fmt yuv420p -movflags faststart -c:a libfdk_aac -b:a 128k -ac 2 -cutoff 18000 out.mp4
[04:45] <debianuser> These links may be interesting for you: https://trac.ffmpeg.org/wiki/Encode/H.264 https://trac.ffmpeg.org/wiki/Encode/AAC
[04:50] <debianuser> But to make the long story short: usually it's not "bitrate", but "quality" is controlled. In the above example video quality is controlled by the option -crf 22 with range (lossless)0..51(worst), default is 23. From that option codec automatically calculates and adjusts the bitrate to get the requested quality.
[04:51] <debianuser> Another thing that affects bitrate/quality a lot is "profile" (baseline, main, high...) Higher profile gives better quality for the same bitrate. But you have to consider compatibility: many iphone/android players can only handle "baseline" profile.
[04:51] <wookienz> ok thanks, ill have a look into those.
[04:52] <wookienz> ok, sounds like they are good easy presets
[04:53] <debianuser> As for what quality to target... Try different options and look at them side-by-side, if the quality is too good you can lower the "-crf", if it's too bad make it higher. I don't know any better way. :(
[04:54] <wookienz> unfortunately i am looking at converted hundreds of files so wont be able to side by side them.
[04:57] <debianuser> Maybe experiment with a few, to get the idea of how much the quality is changed, then add some safety margin, and use those settings for the others. :) If you don't have time for experiments, use default values, they _should_ be safe.
[04:58] <wookienz> default as in: ffmpeg -i file.mkv -c:a libfdk_aac -c:v copy out.mp4 ?
[05:13] <debianuser> If the input is already H.264 and compatible with whatever you want it to be compatible then yes. By the way, libfdk_aac by default cuts frequencies higher than 14kHz (it's not ffmpeg's default, it's libfdk_aac's one). So if you're encoding some high quality audio you may want to add option "-cutoff 20000" after "-c:a libfdk_aac".
[05:18] <wookienz> ok, thanks.
[05:24] <thshdw> I know that ffmpeg is able to encode and decode in atrac3, and it can also read some metadata like artist, year of track etc. Can it also write this metadata to atrac files?
[11:44] <debianuser> Heh. Video encoder for 8088 CPU: http://trixter.oldskool.org/2014/06/19/8088-domination-post-mortem-part-1/
[11:47] <ubitux> there is no lowres support with h264?
[11:48] <iive> thank god, no
[11:48] <ubitux> too bad :)
[11:48] <iive> not really
[11:50] <iive> last time i did test with lowres, the 2^3 reduction of resolution resulted in astonishing 30% speedup.
[11:50] <ubitux> not that bad then
[11:50] <iive> and this also causes quite a lot of prediction errors.
[11:50] <ubitux> i don't need the output frame
[11:51] <ubitux> i'm looking at extracting just the mv somehow
[11:51] <ubitux> like avframe side data or some other mean
[11:51] <iive> well, grayscale decoding gives you same speed up and full resolution and no artifacts.
[11:51] <ubitux> but then i wouldn't want the frame buffer anyway, just the side data
[11:52] <ubitux> maybe i should just parse the avpacket myself, dunno
[11:52] <iive> there are flags for skipping decoding, they might be more useful...
[11:52] <iive> and there should also be some parsers
[11:52] <ubitux> ah?
[11:53] <ubitux> there are some decode skipping flags?
[11:53] <iive> how else would you do a frame drop?
[12:23] <cuba> can anyone tell me why my resulting file has no video length in vlc?
[12:23] <cuba> http://codepad.org/42vVeFCJ these are the arguments to create it
[12:26] <c_14> What value does (fps) have? And, does ffprobe/ffplay show the correct video length?
[12:30] <cuba> c_14: fps is the frames per seconds of the input coming in
[12:30] <cuba> Duration: N/A :(
[12:38] <tfirg> i would like some suggestions to improve the quality of my x11 grab. I use this command: "ffmpeg -y -f x11grab -draw_mouse 1 -framerate 5 -video_size 3840x2160 -i :0.0 -f alsa -i pulse -c:v libx264 -pix_fmt yuv420p -b:v 2500k -c:a aac -strict -2 -b:a 192k -preset ultrafast -tune zerolatency -crf 18 -movflags +faststart testgrab.mp4"
[12:39] <tfirg> the result is here (2 minute video) : https://www.youtube.com/watch?v=dt-WiII6Z2A
[12:39] <tfirg> it shows blue block and flashes
[12:39] <tfirg> and some other glitches
[12:40] <tfirg> is there a way to make that better?
[13:16] <c_14> cuba: I got that, what I meant is that you're setting it to a value somewhere, what value are you setting it to.
[13:18] <cuba> 20 atm
[13:18] <cuba> i think its mayabe that my ffmpeg process isnt directly terminated after recording stops
[13:18] <cuba> i terminate in destructor of the object
[13:19] <cuba> but the options that way are correct and shoudl result in a proper duration?
[13:22] <cuba> mh not really the case it seems
[13:24] <c_14> you might want to switch the input option to -framerate instead of -r, I've had some issues with that.
[13:26] <cuba> ill try that
[13:27] <cuba> thanks c_14
[13:36] <c_14> Does it work?
[13:38] <cuba> no
[14:04] <cuba> c_14: it isnt ffmpegs fault
[14:05] <cuba> thanks for your help
[14:07] <tfirg> http://fpaste.org/113009/36979891/ anyone? any little suggestion/pointer appreciated
[14:10] <c_14> Increase your framerate? Also, setting a bitrate and a crf is useless with x264 (afaik). maybe try with git ffmpeg (assuming you aren't).
[14:14] <tfirg> thanks yes i will gradually increase framerate , i suspect it might cause buffer underruns
[14:14] <tfirg> as for the bitrate and crf, i suppose it wont hurt either
[14:14] <tfirg> not using git, using 2.1.4
[14:15] <tfirg> the biggest problem is the blue blocks and flickers
[14:43] <termos> When using ffplay on a stream it will say bitrate: N/A, is this because the stream doesn't want to supply this metadata?
[14:44] <termos> I'd like to know the input bitrate to my encoder, but the av_dump_format will also tell me bitrate: N/A
[15:22] <metathink> hi
[15:55] <leeivan86> is there anyway to avoid reencoding a video when i just want to crop the video to get rid of blackboxes?
[15:55] <sacarasc> No.
[15:56] <leeivan86> ok thanks
[15:56] <relaxed> leeivan86: mplayer and friends have filters for this
[15:56] <leeivan86> ah ok
[15:57] <relaxed> in fact, most of ffmpeg's filters were ported from mplayer
[16:47] <jkli> hi guys
[16:48] <jkli> i want to batch convert videos but want also to cut / normalize to null / silence certain parts in the video
[16:48] <jkli> any idea?
[16:49] <klaxa|work> write a shell-script?
[16:50] <jkli> how do i edit the audio
[16:51] <relaxed> using sox would work
[16:51] <relaxed> I'm not sure if ffmpeg has a filter do that.
[16:52] <jkli> no, i want to silence an audio part
[16:52] <jkli> without cutting the video
[16:56] <relaxed> demux the audio, gogogadget sox, remux the audio back with the video
[16:58] <relaxed> ah, look at ffmpeg's afade filter
[17:04] <jkli> ah
[17:04] <jkli> afade is what im looking for
[17:28] <Neppy> hi there
[17:30] <Neppy> trying to use ffmpeg :D for some reason i have no avformat_seek_frame function.. so using avformat_seek_file (the docs claim thats still WIP though? so i'm confused..) however that hands me error code -1, which i can't really seem to find much about
[17:32] <klaxa|work> there is no avformat_seek_frame in the first place
[17:32] <klaxa|work> how are you calling avformat_seek_file? did you check what -1 means?
[17:33] <klaxa|work> did you try to use av_seek_frame?
[17:33] <Neppy> well tried to check what it means, didnt find any info on that and av_strerror just says "Error number -1 occured"
[17:34] <Neppy> yeah will probably try av_seek_frame and calling it wiht: int64_t timestamp = frameNumber * AV_TIME_BASE;
[17:34] <Neppy> int error = avformat_seek_file(avContext, videoStreamIndex, timestamp, timestamp, timestamp, AVSEEK_FLAG_FRAME);
[17:35] <Neppy> same for av_seek_frame... i think my timestamp idea may be off
[18:03] <Neppy> yeah thats the issue..
[20:51] <ChocolateArmpits> Has anyone compiled ismindex here ?
[23:02] <lanodan> I'm trying to convert mp4 to ogv but I got so much pixelisation even with -qscale 0 and -qmax 0
[00:00] --- Thu Jun 26 2014
1
0