For extra credit ask those java folks to calculate some large  
factorial, something in the order of a lot of digits (million?) Don't  
be shy.

Then see how long things take.

PS A year or so back I looked at changing the internal math logic to  
handle 32bit math directly since once you go over the SmallInteger  
boundary
it backs off to a message send. What I found was that you could then  
count from +/- 2 billion really fast, but some bugs lurked and in  
looking at
the macro-benchmarks why none of this mattered.



On Nov 2, 2004, at 1:42 PM, Andreas Raab wrote:

> Heh - I was going to write message saying that I'd like to see a  
> measure of the actual difference in speed if we change this to:
>
>    [x := 1000 * 10000000] timeToRun.
>
> :^) Besides, one of the more interesting things to do is to not use  
> "long/int" but rather the Object integer type.
>
> Cheers,
> - Andreas

--
======================================================================== 
===
John M. McIntosh <johnmci at smalltalkconsulting.com> 1-800-477-2659
Corporate Smalltalk Consulting Ltd.  http://www.smalltalkconsulting.com
======================================================================== 
===