Let us not mix the two PR peer-reviewed and public relations...
I tend to put a little bit more trust on the PR of ACM (even if it is just an academic conference paper) http://portal.acm.org/citation.cfm?id=1816021&coll=GUIDE&dl=GUIDE&CFID=94608761&CFTOKEN=50783980&ret=1)#opub than the one of NVIDIA (even if it is a great marketing oeuvre d'art).
http://blogs.nvidia.com/ntersect/2010/06/gpus-are-only-up-to-14-times-faster-than-cpus-says-intel.html
The reality is simple GPU was around for many years now and it did not make the impact expected by nvidia yet (I have been watching it for around 4 years). The paradigm shift in the programming model is still too big for normal programmers to jump the GPU wagon. People should realise that running spe...