At Wed, 20 Aug 2008 16:43:55 +1000,
Sven Schott wrote:
>
> That is cool. I didn't even know about tinyBenchmarks.
>
> Pepper Pad 3 tinyBenchmarks:
>
> '28193832 bytecodes/sec; 922382 sends/sec'
>
> From my MacBook Pro for comparison:
>
> '524053224 bytecodes/sec; 12262411 sends/sec'
28 million bc/s is decent for a smaller device. It is usually a bad
idea to draw too much from the result of micro benchmarks, but because
Squeak's VM implementation is portable across platform (and unfriendly
enough to the memory system), the result from tinyBenchmarks has been
very "proportional" to what it feels like on the platform.
Another thing I do is something like (change the parameters as you like):
[200 timesRepeat: [Display fill: (100@100 extent: 400@400) fillColor: Color random]] timeToRun
to stress the memory band width and to see the display update
performance.
-- Yoshiki
_______________________________________________
Beginners mailing list
[hidden email]
http://lists.squeakfoundation.org/mailman/listinfo/beginners