[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Gnash-dev] youtube profiling
From: |
strk |
Subject: |
Re: [Gnash-dev] youtube profiling |
Date: |
Fri, 4 May 2007 11:36:26 +0200 |
On Thu, May 03, 2007 at 02:44:28PM +0200, Udo Giacomozzi wrote:
> s> 12:25 < strk> so there are more frames available (1550) then actually used
> (385)
>
> frames available == original frames from the video? You mean 1165
> frames are being skipped?
I took a closer look, and the only fixed value I get is
that a new frame is ready 3656 times, which take us
to 30fps (3656/121 seconds).
The move ideally runs at 25fps so some frames are surely skipped
already (only the last available frame is shown).
Based on the profiling, the bottlneck is still in the renderer,
at least for AGG (not for opengl!!).
All in all we can likely postpone performance analisys to
after the release...
--strk;
- Re: [Gnash-dev] youtube profiling, (continued)
- Re: [Gnash-dev] youtube profiling, strk, 2007/05/03
- Re[2]: [Gnash-dev] youtube profiling, Udo Giacomozzi, 2007/05/03
- Re: [Gnash-dev] youtube profiling, strk, 2007/05/03
- Re[2]: [Gnash-dev] youtube profiling, Udo Giacomozzi, 2007/05/03
- Re: [Gnash-dev] youtube profiling, strk, 2007/05/03
- Re[2]: [Gnash-dev] youtube profiling, Udo Giacomozzi, 2007/05/03
- Re: [Gnash-dev] youtube profiling,
strk <=
- Re[2]: [Gnash-dev] youtube profiling, Udo Giacomozzi, 2007/05/04
- Re: [Gnash-dev] youtube profiling, strk, 2007/05/04
- Re: [Gnash-dev] youtube profiling, strk, 2007/05/04
- Re: [Gnash-dev] youtube profiling, strk, 2007/05/04
- Re: [Gnash-dev] youtube profiling, strk, 2007/05/04
- Re[2]: [Gnash-dev] youtube profiling, Udo Giacomozzi, 2007/05/04
- Re: [Gnash-dev] youtube profiling, Rob Savoye, 2007/05/04
- Re: [Gnash-dev] youtube profiling, strk, 2007/05/06
- Re: [Gnash-dev] youtube profiling (-O2 and inlining), strk, 2007/05/04
- Re[2]: [Gnash-dev] youtube profiling (-O2 and inlining), Udo Giacomozzi, 2007/05/04