Dr. H. Nikolaus Schaller wrote:
Am 25.09.2007 um 17:20 schrieb Dani Anon:
But hey I like how you conveniently left unanswered my comment about
how the FBUI and DirectFB projects exist solely to remove the X server
overhead. Can you explain to me and them why they are wrong and how
they have wasted all those months of development to solve a problem
that just doesn't exist?
It exists theoretically but the practical relevance depends on several
factors.
I have done some experiments on the OpenMoko and I believe that the
framebuffer itself is the slowest part (don't know why). But X11 *is*
already
quite fast.
I could achieve a lot of speed up in my GUI Framework (yes it is another
choice!)
by double buffering X11 (i.e. drawing offscreen first and then copying the
modified block in a single bitmap copy command).
In my GUI toolkit the main speed killer is the missing FPU in ARM
processors
which has to rely on FPU emulation.
So, why optimize something that already works sufficiently? It is like
washing whiter
than white. But if there is someone who wants to optimize it further -
why not?
So - IMHO - the X11-Overhead is a neglectable problem. And the benefits
of X11
overweight any drawbacks.
There may not be very much performance differences, but there is one
more drawback of using Xlibs. File size and memory.
--
Lorn 'ljp' Potter
Software Engineer, Systems Group, MES, Trolltech
_______________________________________________
OpenMoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community
