Re: [DotGNU]performance test / SWF ?

[Rhys Weatherley]

On Thursday 09 October 2003 08:56 pm, Thomas Fritzsche wrote:
> Hi Rhys and dotgnu-hackers,
>
> good to hear that my general compute-bound aspects run "pretty good".
> and I agree that speed can improved by the ways you proposed, but I have
> currently a little other problem. It is that my SWF/System.Draw
> applications are so slow. I compared this with Neil results from
> Windows+RedHat(vmware) and for some reasons his system is more than 10
> times faster. (windows even more) So this is something with my system
> configuration, but dotgnu SWF is the only application I observe this. I
> wrote to a linux mailinglist to get a hint, compiled different freetype
> versions, changes X11 settings etc. but nothing works. Everything else is
> fast as before and dotgnu is still slow as before.

Different X systems can give different behaviours.  Neil's may have a better 
accelerated X driver than yours.  Or the garbage collector performs 
differently because of different amounts of memory/swap.  It could be 
basically anything.  Performance is a combination of factors - not a linear 
measure.

Even if your X pipeline is slower doesn't mean that it cannot be optimized 
with clever algorithms.  X is fast.  Very fast.  But it is also very easy to 
use it poorly.  Rather than worrying about how slow it is, concentrate on 
writing better algorithms to feed text into the pipeline.  All things being 
equal, if we ran this on a 100 MHz machine, it should scream - *if* the right 
algorithms are used.  Which obviously they aren't at present.

What does "ilrun -M" report as the most commonly called methods when you run a 
drawing test?  That should give you a place to start.

Cheers,

Rhys.