Re: Snappy GUI response

On Thu, 2003-07-03 at 18:10, Keith Packard wrote:

>  1)	InternAtom -- Qt interns a bunch of atoms at startup time in
> 	one round trip; Gtk+ doesn't bother.  That's actually a significant
> 	time savings.

"doesn't bother" - well, umm, that's one way of putting it. I do have
some code for a similar hack now, that maybe I'll finish up today.

>  2)	Extensions.  Each extension takes at least two round trips to 
> 	initialize.

Both this and InternAtom are startup times, things so don't really
affect interactivity.

>  3)	GetGeomtry.  Applications are curious about where windows are and
> 	the ICCCM doesn't give them enough information about borders and 
> 	such.  Mozilla spends a lot of time asking where its windows are
> 	on the screen.

Figuring out a window manager protocol extension so that GTK+ could
always track where a window on the screen is without synchronous
roundtrips would definitely be nice, though one round trip per configure
isn't going to make a *huge* difference.

> While you might think all of these are irrelevant in a non-networked 
> environment, the fact is that any context switch, whether across a 
> network, between two processes or between a process and the kernel takes  
> time.  Batching operations between context switches is a good idea in all 
> environments, and that's where X can actually *beat* kernel window systems 
> that have very little batching available.
> 
> The only X server graphics operation which has any noticeable effect on 
> 2D application performance is screen->screen copies -- nothing else even 
> shows up on the radar in any objective system-wide measurements I've done.

One exception to this is terminal output speed, which depending on the
terminal implementation can end up being pretty much a pure benchmark
of how fast the system can composite AA characters.

Regards,
						Owen