[MPlayer-dev-eng] [PATCH] A new ASS renderer

[Reimar Döffinger]

On 23 Sep 2012, at 22:54, Reimar Döffinger <Reimar.Doeffinger at gmx.de> wrote:
> On Mon, Sep 24, 2012 at 12:10:21AM +0400, Vladimir Mosgalin wrote:
>> Hi Reimar Döffinger!
>> 
>> On 2012.09.23 at 21:32:07 +0200, Reimar Döffinger wrote next:
>> 
>>> Have a sample at hand? The subtitles I have around are probably usually
>>> too simple.
>>> Also a "perf record"/"perf report" trace with MPlayer debug build and debug symbols
>>> for the driver binaries might help figuring out the cause.
>> 
>> We had discussion about this already and you concluded that problem is
>> in drivers and it can't be helped and you said perf was useless.
>> http://lists.mplayerhq.hu/pipermail/mplayer-users/2011-January/081905.html
>> 
>> Despite newer drm and mesa releases, it still persists, unfortunately
>> (well, I haven't changed video card because I don't see many reasons
>> to..).
>> 
>> That issue, plus this one:
>> http://lists.mplayerhq.hu/pipermail/mplayer-users/2011-May/082587.html
>> (I continued it on the same thread)
>> are the ones why I have to use -vf ass rendering exclusively. I mean,
>> karaoke effect is one thing, but having insane slowdown on both 16/9 and
>> 4/3 videos just because my monitor is 16/10 is >.< (I finally got rid of
>> old 4/3 aspect CRT last summer, but the problem in question is worse on
>> 16/10 monitor in either case).
> 
> I suspect that is more related to the larger resolution than the aspect.
> -msglevel vo=7 can be interesting since it will print some statistics
> that can hint at the issue.
> I also made a quick-and-dirty hack that should eliminate what I guessed
> to be the issue in my tests (though it might need some tuning), but it
> made a surprisingly small difference only.

Even stranger: even without this patch, the driver does not use any relevant amount of CPU at all when using the OpenSource driver, basically freetype is using all of it (I used a trivially decodable video with those subs).
Only the closed-source driver is a bit of an issue.
I'll have to read up on the thread again, but did we ever figure out why some strange expand filter magic made a difference?
Admittedly this is using some Radeon 5700 card.