Newbie: Benchmarking Squeak vs. Flash, Director and IE

Andreas Raab andreas.raab at
Fri Jan 17 02:03:58 UTC 2003

> Hi Group, 

Hi Ralf,

> I just tried to evaluate its performance and i'm wondering
> if i did it correct, since it seems to be pretty fast. At least 
> from my point of view, since i'm mostly working with flash.

Your measures resemble my experience pretty closely. A little while ago I
did a course with some design students ("game programming" - the idea was to
give them a hands-on experience in programming game like environments) and
since I wasn't sure how far they'd go (we needed collision detection -
that's in general an O(n^2) exercise) I ran similar tests. The result were
quite similar with one exception - Flash was much slower than in your
comparison (about ten times slower than Director/Lingo) but that was Flash 5
so there's a chance something has improved with MX. But on the general scale
the results were similar.

> squeak/smalltalk       result ms: 7 (62 with 100000 loops)
> director/lingo            result ms: 33 (300 ms with 100000 loops)
> javascript/ie5.5         result ms: 100 (610 ms with 100000 loops)
> actionscript/flash mx result ms: 1382 

I think I remember JS a little faster (about half of Director/Lingo) but
this probably depends on the exact system you were using (I used Flash 5 vs.
Director 8.5 vs. JS/IE6 vs. Squeak 3.2).

> Maybe you can take a short look at my smalltalk script
> and tell me, if it is correct ? It should create an object and 
> call its method func 10000 times. 

Yes, it all looks correct. And like I was saying, I saw similar results in
my own measures so that's pretty much the way it seems to be.

Happy Squeaking,
  - Andreas

More information about the Squeak-dev mailing list