[FFmpeg-user] numerous bugs-80117-N gdac030d version

[Peter White]

Again with the full quote?!

juan carlos Rebate wrote:
> 2016-08-05 23:07 GMT+02:00 Peter White <peter.white at posteo.net>:
>
>> Am 05.08.2016 um 22:29 schrieb juan carlos Rebate:
>>
>>> 2016-08-05 21:29 GMT+02:00 Peter White <peter.white at posteo.net>:
>>>
>>> juan carlos Rebate wrote:
>>>>
>>>> 2016-08-05 19:46 GMT+02:00 Paul B Mahol <onemda at gmail.com>:
>>>>>
>>>>> On Friday, August 5, 2016, juan carlos Rebate <nerus492 at gmail.com>
>>>>> wrote:
>>>>>
>>>> Full uncut ffmpeg output missing.

Still missing! Some pointers follow. I am less and less inclined to
reply.

>>>> You should
>>>> provide the full, simplified command and its output. Simplified means
>>>> that you should provide a command line that is just enough to
>>>> demonstrate the bug. Remove all options that are irrelevant to
>>>> triggering it.

>> https://ffmpeg.org/bugreports.html

> line >ffmpeg -i video.flv -c:v libx264 -b:v 1024k -r 25 -c:a mp3 -ar 44100
> video2.avi

Where is the *full* *uncut* *output* of above command? I cannot even
see the version of ffmpeg there. Also, the rate control is done by
libx264 which is not a native ffmpeg encoder.

> results on ffprobe
>
> Input #0, avi, from 'video.avi':
>   Metadata:
>     encoder         : Lavf57.44.100
>   Duration: 00:21:02.86, start: 0.000000, bitrate: 205 kb/s
>     Stream #0:0: Video: h264 (High) (H264 / 0x34363248), yuv420p, 480x360
> [SAR 1:1 DAR 4:3], 64 kb/s, 25 fps, 25 tbr, 25 tbn, 50 tbc
>     Stream #0:1: Audio: mp3 (U[0][0][0] / 0x0055), 32000 Hz, stereo, s16p,
> 128 kb/s

And, most importantly, it works for me. That, of course, may be due to
me using the latest ffmpeg git master and the latest stable libx264.
I did not use mp3 as audio encoder, though. But the relevant parts are
'-ar 44100' and '-b:v 1024k' anyway, right? Also, no video corruption.

Oh, and I just now see it. Your encoding target is named video2.avi, yet 
you ffprobe video.avi which is a different file and might explain
why it has an audio sample rate of 32 instead of 44.1 kHz.
And yes, even such trivial mistakes are easily spotted, provided one 
actually posts their *full* *uncut* *output* of commands.