Croquet on ARM

Tim Rowledge tim at sumeru.stanford.edu
Mon Dec 9 18:44:41 UTC 2002


Bert Freudenberg <bert at isg.cs.uni-magdeburg.de> is claimed by the authorities to have written:

> I wonder how Croquet would perform on this XScale-based NVIDIA-equipped
> machine? http://www.drobe.co.uk/riscos/artifact500.html
It will be better than on my aging (seven years old) StrongARM but not
enormously so; there's no OpenGL or sensible analogue as yet so using
the card's 3D stuff is going to be a bit tricky.  Also we still lack a
FPU with that particular ARM. The good news about the machine is the
memory bandwidth is massively better and the cpu is somewhat better. I'm
guessing (without any evidence as yet) that Squeak ought to run about
4-6 times faster on average. I'm basing this on extrapolating from a
previous go-faster option for the Acorn RPC that put a fast (300MHz
StrongARM) on a board with fast-ish memory and managed to roughly triple
Squeak performance.

Anyong interested in some really rough and tumble argumentation about
the relative merits of machines should take a look at
comp.sys.acorn.misc and dip into the debate about the differences
between the Omega and Iyonix. Makes our discussions about
java/c-dull/whatever look quiet :-)

"May you live in interesting times" as the old saying goes....

tim
-- 
Tim Rowledge, tim at sumeru.stanford.edu, http://sumeru.stanford.edu/tim
Useful random insult:- He hasn't a single redeeming vice.  -- Oscar Wilde




More information about the Squeak-dev mailing list