[FFmpeg-devel] [PATCH] wmapro decoder

Matej Knopp matej.knopp
Mon Sep 7 23:06:22 CEST 2009


Also if it's any help, the previous revision used to crash (though I
don't have the stacktrace right now) with -async 1 when the file was
processed from beginning (-ss 0). 19778 doesn't do that anymore,
rather than that it stutters.

-Matej

On Mon, Sep 7, 2009 at 10:57 PM, Matej Knopp<matej.knopp at gmail.com> wrote:
> Hi,
>
> the last commit of wmaprodec.c (from rev. 19777 to 19778) introduced
> heavy stutter into the decoder. Note that it only happens with nonzero
> -async ?(i.e. async 1), even if the file is played from beginning.
> When no async is specified file decodes normally.
>
> Unfortunately I don't have any files that I'm allowed to share that
> can be used to reproduce the problem.
>
> -Matej
>
> On Sun, Sep 6, 2009 at 7:56 PM, Diego Biurrun<diego at biurrun.de> wrote:
>> On Sun, Sep 06, 2009 at 07:39:57PM +0200, Sascha Sommer wrote:
>>>
>>> On Sonntag, 6. September 2009, Diego Biurrun wrote:
>>> > On Sun, Sep 06, 2009 at 11:00:10AM +0200, Sascha Sommer wrote:
>>> > > The MAX_AUDIO_FRAMESIZE chang is no longer required.
>>> >
>>> > Do you have performance numbers for this decoder compared to the binary?
>>>
>>> Pentium M 1,6 GHz, frequency scaling disabled. Not sure if there is a better
>>> way to benchmark.
>>
>> This is a recurring topic, we need to add it to the documentation. ?Mans
>> complained loudly the last time I used MPlayer to benchmark. ?So can
>> somebody please provide the material to fill a short section in the docs
>> about the topic?
>>
>> Diego
>> _______________________________________________
>> ffmpeg-devel mailing list
>> ffmpeg-devel at mplayerhq.hu
>> https://lists.mplayerhq.hu/mailman/listinfo/ffmpeg-devel
>>
>



More information about the ffmpeg-devel mailing list