Re: About our effort at NoMachine

Am Mittwoch, 26. März 2003 18:45 schrieb Alan Cox:

Hi,

> but stop on a failure. That gives you the ability to do more than just
> batching, since you can batch a sequence of commands for the general
> case and respond to the failure. My gut feeling (and I've not been
> doing analysis unlike you folks) is that this would improve things
> immensely for startup

Yeah, in general all traditional latency hiding techniques would help 
tremendeously. I especially think about hiding latency and bandwidth problems 
with a possibly persistent cache (e.g. using a mmap hashed ondisk file with 
checksums for individual items like pixmaps etc.) By putting the data in nice 
suitable precomputed formats it should be even possible for the X server to 
directly map them.  

Paradoxically this can also help to reduce the memory footprint of XFree86. 
(Which is unfortunately in a lot of cases higher than proprietary 
implementations)

IMHO the really important thing is to use the optimal layout and format of the 
data in this cache so that the X server can really benefit without extra 
conversion work.

Regards,
-- martin

Dipl.-Phys. Martin Konold

e r f r a k o n
Erlewein, Frank, Konold & Partner - Beratende Ingenieure und Physiker
Nobelstrasse 15, 70569 Stuttgart, Germany
fon: 0711 67400963, fax: 0711 67400959
email: [email protected]