> The simple little test model I ran on my 500 MHz
> PIII took 37 seconds to complete. I'd be interested
> in hearing how that stacks up to a native windows
> implementation. I posted the .fem definition in
> the 'Files' area of this discussion group:
> look for hollow-20.fem.
I ran this problem on my old at-home machine with a screaming 200 MHz
Pentium and 96MB of memory. It took 1:55 to complete from the time I
pressed the "turn the crank" button until the solver was completely
done writing the results to disk. If I assume that the solution time
would scale linearly to a 500 MHz machine, 2/5 of the 200 MHz solution
time is 46 seconds.
Dave.
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/