[FFmpeg-devel] [PATCH] Port mp=eq/eq2 to FFmpeg

Paul B Mahol onemda at gmail.com
Tue Jan 20 10:16:19 CET 2015


On 1/19/15, arwa arif <arwaarif1994 at gmail.com> wrote:
> On Mon, Jan 19, 2015 at 8:53 PM, Stefano Sabatini <stefasab at gmail.com>
> wrote:
>
>> On date Monday 2015-01-19 15:20:54 +0100, Clement Boesch encoded:
>> > On Mon, Jan 19, 2015 at 02:09:33PM +0000, Paul B Mahol wrote:
>> > > On 1/18/15, arwa arif <arwaarif1994 at gmail.com> wrote:
>> > > > Attached the patch.
>> > > >
>> > >
>> > > I'm for dropping eq code and rename eq2 to eq.
>> >
>> > Yes please let's not add 2 filters for this, it's insane. Also, "eq" is
>> > quite a bad name, but well...
>> >
>>
>> > What happened to the idea of having the feature in hue instead?
>>
>> I'm not against that if we agree it's a better path.
>>
>> About eq/eq2, are there really performance concerns for having both of
>> them?
>>
>> Arwa, can you show some benchmarks?
>>
>
> The benchmark result for a demo video for
>
> 1.) eq filter:
>
> frame= 4690 fps=120 q=31.0 Lsize=   16788kB time=00:03:07.64 bitrate=
> 732.9kbits/s
> video:7828kB audio:8796kB subtitle:0kB other streams:0kB global headers:0kB
> muxing overhead: 0.983091%
> bench: utime=45.871s
> bench: maxrss=19420kB
>
>
> 2.) eq2 filter:
>
> frame= 4690 fps=110 q=31.0 Lsize=   16788kB time=00:03:07.64 bitrate=
> 732.9kbits/s
> video:7828kB audio:8796kB subtitle:0kB other streams:0kB global headers:0kB
> muxing overhead: 0.983091%
> bench: utime=51.475s
> bench: maxrss=19920kB

That is strange considering they share same code.
>
>
>> --
>> FFmpeg = Fiendish Free Majestic Philosophical Ermetic Gem
>> _______________________________________________
>> ffmpeg-devel mailing list
>> ffmpeg-devel at ffmpeg.org
>> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>>
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>


More information about the ffmpeg-devel mailing list