I stumbled upon a paper called High-performance 3D Compressive Sensing MRI reconstruction that says "Our (optimized) baseline implementation on a quad-core Core i7 is able to reconstruct a 256×160×80 volume of the neurovasculature from an 8-channel, 10× undersampled data set within 56 seconds, which is already a significant improvement over existing implementations. The latest six-core Core i7 reduces the reconstruction time further to 32 seconds. Moreover, we show that the CS algorithm benefits from modern throughput-oriented architectures. Specifically, our CUDA-base implementation on NVIDIA GTX480 reconstructs the same dataset in 16 seconds, while Intel's Knights Ferry (KNF) of the MIC (Many Integrated Core) architecture even reduces the time to 12 seconds."
It would be nice to know what clock speed the Core i7s were, but still, hooray for faster MRIs. Of course, a 6-core system runs around 200 watts max? while a Knights Ferry system would run at 500 watts max? Performance per watt for the pure Core i7 system would be (1p/200w=0.005) while the Larrabee would be (1.625p/500w=0.00325). Sure it's less efficient, but considering the GTX 480 is a scorcher as well, Intel at least beats Nvidia here in pure performance and efficiency.
Curiously, Google Analytics showed a spike in visitors on February 4, 2011, without about 30 visits from Beaverton, (Oregon), using Internet Explorer on Windows, from "intel corporation" service corporation. I mean, this information is all available on Google.