Initial customer testing has shown what they are calling “pretty impressive” results. Comparing a 1265 MHz Tesla C1060 running Jacket 1.3 with CUDA 2.3 and a 1025 MHz Fermi C2050 running Jacket 1.4 with CUDA 3.0, the results are as follows:

Multiply-add (MAD), double precision on GPU, Fermi is 1.4x faster than Tesla for 4000x4000 matrices (2s vs 2.8s)

2D 2-point (GRADIENT), single precision on GPU, Fermi is 3x faster than Tesla for 4000x4000 matrices (0.3s vs 0.9s)

2D FFT (FFT2), double precision on GPU, Fermi is 2x faster than Tesla for 4000x4000 matrices (0.1s vs 0.2s)