On 3/8/07, Eero Tamminen <[EMAIL PROTECTED]> wrote:
Hi,
ext Kalle Vahlman wrote:
> I haven't confirmed this, but according to the (non-exhaustive)
> benchmarking currently it probably should be faster to work with image
> surface in Cairo and only push the result to X, since it is
> accelerated in neither.
It might not be faster[1], but at least rest of the system (using X
server) remains more responsive while a single client uses CPU instead
of server used by all the applications being busy.
This is interesting. I hadn't thought about keeping heavy calculations
up at the client level in order to keep the X server responsive. Even
though it goes against the whole point of the X Render architecture,
it does make sense for a single-threaded X server. Have to think about
that some more...
[1] There's a bug in Cairo bugzilla about slowness on 16-bit display
when using Render though...
Although the compositing code could still be better optimized for
16-bit displays, the real "bug" there was that they were recompositing
large surfaces at each scroll step, which will probably never be fast
unless compositing is hardware accelerated.
Dan
_______________________________________________
maemo-developers mailing list
[email protected]
https://maemo.org/mailman/listinfo/maemo-developers
