Re: interesting article on XFree86 speed

We haven't taken a precise inventory.  The criminal thing is that
the X consortium never did anything about the issues raised in the
paper, and now that the more pressing things around text and graphics
are under control, we really need to revisit this and fix things as
should have happened nearly 10 years ago.

Alot of the mistakes around timestamps and event processing are clearly
fodder.

It isn't clear the stuff around window backgrounds/borders is worth
it.

A number of the issues clearly affect how quickly things get
(re)painted.

Some may be moot as systems are so much faster.

But that paper covers most of what is wrong, beyond 2D graphics
which Render generally fixes.
					- Jim

> Sender: [email protected]
> From: Soeren Sandmann <[email protected]>
> Date: 28 Oct 2002 16:15:43 +0100
> To: [email protected]
> Cc: Allen Akin <[email protected]>
> Subject: Re: [Render] interesting article on XFree86 speed
> -----
> <[email protected]> writes:
> 
> > 	3) Keith and I chatted with Owen about what to do with his
> > small extension to fix embedding, and have come to the conclusion we
> > need a "fix the protocol" extension, to fix items that matter (some
> > of what Hania/Mark/Joel complained about probably don't matter in
> > current contexts). 
> 
> That sounds interesting. Could you summarize what you plan to include
> in this extension? Which of the Hania/Mark/Joel problems do you think
> matter, and which don't matter?
> 

--
Jim Gettys
Cambridge Research Laboratory
HP Labs, Hewlett-Packard Company
[email protected]