[Ffmpeg-devel-irc] ffmpeg-devel.log.20190111

burek burek021 at gmail.com
Sat Jan 12 03:05:03 EET 2019


[11:46:13 CET] <pross> what is the going rate for a working photocd decoder?
[11:47:26 CET] <durandal_1707> what?
[11:47:54 CET] <funman> http://ffmpeg.org/pipermail/ffmpeg-devel/2019-January/238711.html
[11:49:07 CET] <durandal_1707> it is between me and Carl
[11:49:23 CET] <pross> ack
[11:49:39 CET] <funman> I'll have it known Carl is a good paying customer
[11:49:54 CET] <durandal_1707> lol
[11:56:27 CET] <cone-133> ffmpeg 03Paul B Mahol 07master:0c0fc8896cdd: avfilter/af_anlmdn: log used parameters
[11:56:28 CET] <cone-133> ffmpeg 03Paul B Mahol 07master:b5ae61d2c0ae: avfilter/af_anlmdn: do not trim first samples
[11:56:29 CET] <cone-133> ffmpeg 03Paul B Mahol 07master:c563513fdf72: avfilter/af_anlmdn: drain samples at end
[12:44:31 CET] <durandal_1707> what solutions you will always use to stream audio & video?
[12:55:43 CET] <funman> upipe
[12:56:16 CET] <durandal_1707> upipe is just component its not full solution
[12:58:31 CET] <durandal_1707> repeat untrue statement until it becomes truth - that is new multimedia world
[12:58:34 CET] <kierank> there is uplay.c
[12:58:41 CET] <kierank> the examples do many things
[12:58:50 CET] <JEEB> yea, upipe is a framework
[12:59:59 CET] <durandal_1707> so FFmpeg now lies in web page when it claims its full solution to stream?
[13:00:17 CET] <JEEB> why?
[13:00:25 CET] <durandal_1707> ffserver is gone
[13:00:31 CET] <JEEB> uhh
[13:00:36 CET] <JEEB> I'm not even going to go there
[13:00:50 CET] <JEEB> because there's the stuff you can do with FFmpeg's libraries by itself
[13:00:52 CET] <JEEB> like UDP multicast etc
[13:01:02 CET] <JEEB> and then you have the stuff where you at the very least need something like nginx
[13:01:11 CET] <JEEB> f.ex. DASH/HLS HTTP POST output
[13:01:39 CET] <JEEB> adn generally speaking you really don't want to do the media serving with your encoder :|
[13:01:55 CET] <durandal_1707> so we lie?
[13:01:58 CET] <JEEB> no?
[13:02:22 CET] <JEEB> like if you only want to do something 100% FFmpeg's libs then UDP multicast or so is one 100% valid example
[13:02:40 CET] <JEEB> and not like you could do HLS/DASH with ffserver
[13:02:50 CET] <JEEB> like seriously, are you just effing with my brain or something?
[13:02:58 CET] <durandal_1707> no, sorry
[13:03:09 CET] <kierank> JEEB: he's just trolling
[13:04:08 CET] <JEEB> I would then recommend taking a course on that from TheFluff
[13:04:09 CET] <durandal_1707> no trolling, someone asked my privately is it possible to stream with ffmpeg
[13:04:33 CET] <JEEB> yes, yes it is
[13:05:04 CET] <kierank> define stream
[13:07:17 CET] <durandal_1707> it was not so detailed in its specification
[13:14:46 CET] <funman> NAME
[13:14:46 CET] <funman>        fread, fwrite - binary stream input/output
[13:47:06 CET] <cone-133> ffmpeg 03Peter Ross 07master:b4e6d1f597a3: intreadwrite: add AV_RL64A, AV_WL64A
[13:47:07 CET] <cone-133> ffmpeg 03Peter Ross 07master:127564b3f114: dstdec: big-endian compatiblity
[13:50:45 CET] <durandal_1707> pross: please follow style of other commits in commit log message
[16:57:17 CET] <cone-730> ffmpeg 03Paul B Mahol 07master:c65c4ce77365: avfilter/af_anlmdn: do not output more samples than received
[19:16:09 CET] <atomnuker> lots of posts today, I guess its because its a friday
[19:16:48 CET] <funman> should i re-post my patch with my salary disclosed?
[19:17:36 CET] <atomnuker> only if you middle click pasted it by accident and want to blame the primary buffer for existing
[19:24:16 CET] <durandal_1707> you will pay for mocking me
[19:27:09 CET] <durandal_1707> shit, what i caused...
[19:30:04 CET] <kierank> durandal_1707: i'm speechless
[20:03:30 CET] <BradleyS> well, that escalated quickly
[20:11:50 CET] <gnafu> Ugh!
[20:12:31 CET] <gnafu> "I want more [thing].  Do you hate [thing]?"
[20:12:37 CET] <gnafu> That is never valid.
[20:12:48 CET] <gnafu> "Think of the children" level BS.
[20:13:08 CET] <BradleyS> what do you have against the children
[20:13:41 CET] <gnafu> BradleyS: Children tend to be transparent and honest to a fault.
[20:13:47 CET] <gnafu> ;-)
[20:13:54 CET] <BradleyS> ;)
[20:16:26 CET] <BradleyS> i understand what people are trying to accomplish with disclosure but seems to me very much like a problem we have here in america
[20:16:54 CET] <BradleyS> people think that if they make a law, it will absolutely solve everything
[20:17:44 CET] <BradleyS> so we make laws that are utterly unenforceable that people ignore
[20:19:44 CET] <BradleyS> of course sponsored works can have ulterior motives and take short cuts, but imho the code review process should identify any potential code issues
[20:21:13 CET] <BradleyS> better to focus on improving code review than divisive rule-making imho
[20:22:57 CET] <BradleyS> plus it would be just embarrassing to know how much work durandal_1707 has done for only ~17 USD :P
[20:24:20 CET] Action: BradleyS gives durandal_1707 a cookie
[20:25:30 CET] <gnafu> And a patch should never be rejected just because it was "paid for" by some company or for some purpose the reviewer might disagree with.
[20:25:35 CET] <JEEB> yes
[20:25:51 CET] <JEEB> if something is not good enough in review technically that's it
[20:26:02 CET] <JEEB> there should be no grey area in that IMHO
[20:26:18 CET] <JEEB> you have your usual discussion about that with the disagreeing parties
[20:26:40 CET] <JEEB> or call out a 3rd party to alleviate problems if it feels like the two parties cannot come to an agreement
[20:28:04 CET] <kierank> kylophone: i see you've decided to join the insanity
[20:29:00 CET] <JEEB> from one side it's good that people actually responded
[20:29:26 CET] <JEEB> the patch in my honest opinion should receive NACKs
[20:29:51 CET] <JEEB> if our review process has issues that's not something you fix with something that can't be enforced
[00:00:00 CET] --- Sat Jan 12 2019


More information about the Ffmpeg-devel-irc mailing list