ara.t.howard / noaa.gov wrote:
> On Sat, 5 Aug 2006, Peter Hickman wrote:
>
>> Perhaps stepping up to a 6 x 6 grid would allow more meaningful timings?
>
> showing how it performs on a 2, 4, and 8 grid would show how it scaled...
>
> -a
I'll test the 6 x 6 in Ocaml, C and Java but I'm not sure that I will be 
going for the best of 10 runs on this. 2 and 4 are simply too fast to 
measure accurately except for the slowest code. Someone calculated that 
some values higher than 6 would take months to run and 8 might be a 
little too far into that ballpark.

I do have a very noisy and hot AMD 64 4400 with 2 Gb ram running Ubuntu. 
Perhaps I will set this as the new benchmark machine.