[MPlayer-users] Unsharp Mask - Faster/Multi-Threaded?
Phil Rhodes
phil_rhodes at rocketmail.com
Sat Jan 19 00:12:12 CET 2008
Let me get this clear - are we talking about programming pixel shaders to
run on the video here?
P
----- Original Message -----
From: <galenz at zinkconsulting.com>
To: "MPlayer usage questions, feature requests,bug reports"
<mplayer-users at mplayerhq.hu>
Sent: Friday, January 18, 2008 10:47 PM
Subject: Re: [MPlayer-users] Unsharp Mask - Faster/Multi-Threaded?
>
> On Jan 16, 2008, at 7:24 AM, Vladimir Mosgalin wrote:
>
>> Hi galenz at zinkconsulting.com!
>>
>> On 2008.01.15 at 06:26:11 -0800, galenz at zinkconsulting.com wrote next:
>>
>>> I'm trying to use the unsharp mask filter with 1080p video in
>>> mplayer.
>>>
>>> The problem is that the filter is too slow. It appears to run as a
>>> single thread, which is maybe 20-40% short of realtime on my system
>>
>> Maybe multi-thread unsharp mask isn't really what you need, what you
>> need is making video filters process images in separate thread (on
>> separate core) than video decoding, since decoding 1080p uses a lot of
>> cpu. On the other hand, now with 4-cores cpu becoming mainstream,
>> forcing each filter into its own thread and even maybe in-filter
>> thread
>> usage is desirable, too..
>>
>> --
>>
>> Vladimir
>
> I'm actually already multi-threading the decoding of the 1080p video.
> I have tried from 1 to 4 threads, with the fast option as well. I am
> using no other filters.
>
> Basically, it seems that the issue at hand is the rate at which the
> unsharp mask filter runs.
>
> I'd really love an OpenGL customtex file that does an unsharp mask :)
>
> -Galen
> _______________________________________________
> MPlayer-users mailing list
> MPlayer-users at mplayerhq.hu
> http://lists.mplayerhq.hu/mailman/listinfo/mplayer-users
>
>
> --
> No virus found in this incoming message.
> Checked by AVG Free Edition.
> Version: 7.5.516 / Virus Database: 269.19.6/1230 - Release Date:
> 17/01/2008 16:59
>
>
More information about the MPlayer-users
mailing list