Sure, you just need some hints. You might all be interested in one of 
the projects that Andreas is doing: to "APLize" the array and matrix 
arithmetic in Squeak. APL was another system that could beat FORTRAN 
(because the APL interpreter could do algebra, and especially, e.g. 
know that an inner product was desired. This allowed really great 
machine code to be fashioned that optimized inner product.)

Cheers,

Alan

------

At 10:59 AM -0600 2/18/02, Ralph Johnson wrote:
>On 2/17/02 9:26 PM, "Scott A Crosby" <crosby at qwes.math.cmu.edu> wrote:
>
>>  Yep.. Most of this isn't more than an implementation problem.. Why were
>>  there no really good and fast LISP (~comparable to C) compilers for
>>  numerics and bit/byte manipulations before CMUCL? Because nobody had
>>  thought to write one. :)
>>
>
>I recall that MacLisp on the PDP10 was faster at numeric computing than
>FORTRAN back in the early 70s.  They took it as a challenge and kept working
>on their compiler until they beat FORTRAN.
>
>-Ralph


--