This may provide some interesting feedback to WGT.
As I indicated earlier I have a dual cpu Xeon 5260 dual core;
I was curious as to its floating point power. Boy, was I astonished:
This combo which I documented is capable of 44.6 Giga Flops per second. My old motherboard could only muster 589 Mega Flops. This board is a whopping 80 times faster.
It might be wise to look at your code to determine where your bottlenecks are. I still get inconsistencies from time to time.
I am sure my computer is optimized as best as it can be.