Originally posted by duby229
View Post
AMD did publish at least one paper around the time x86-64 was released showing simulation results for SPECint/fp performance as a function of number of GPR's. IIRC 16 was a significant boost from the original 8 in x86-32, but the improvement going from 16->32 was substantially less; thus they chose 16. IIRC the performance boost due to 8->16 GPR's alone was on the order of 10-20%.
Most 32bit binaries are compiled for i686.
Most 64bit binaries at least have SSE2 optimizations.
And that's exactly what you are seeing I'm completely sure of it. A natively compiled 32bit binary will perform very similar to a natively compiled 64bit binary.
We live age where 16GB of RAM is common and we are still stuck at 2bytes worth of GPRs... I think it's a damn shame. (And they can for damn sure no longer claim it's a transistor budget problem.)
Comment