Intel P4 3.2GHz, HT, 512 Mb RAM. TS time around 2.2-2.3s/TS for Phase I

---

<br>***********************************<br>
WinXP Home, P4 3.2GHz HT, 512Mb RAM. Running SETI, CPDN, Predictor, LHC, Einstein, Orbit, Lattice and Folding

AMD64 3400 @ 2673mhz
TS time around 1.7s for my last run that just finsihed yesterday.
New run starting off the same 1.7s Ts.

---

&gt; Just as a matter of interest, how low s/TS numbers do you get on the very
&gt; fastest Intel's and AMD's?
&gt;
&gt; I'm running an AMD64 3200 which is usually running at 2.27 s/TS.
&gt;

Interesting to see these kinds of numbers. My friend's powermac G5 (dual 2.5) generally runs at around 2.4 sec/TS but it usually is faster than his AMD64 box on floating-point intensive workloads. I would like to find out what compilers the CP team use for compiling the software. Using GCC for fortran is a joke under OSX, the IBM compiler produces 200-300% faster code on regular occasions...

I suspect even the AMD64 code generation is somewhat flawed, since I'd suspect an Opteron would easily beat even the fastest Pentiums (since the P4 only has 1 FP unit).

Intel compiler, at least for the Windows and Linux code. But, unfortunately it is not even the latest Intel compiler, which would do better for AMD. Do a forum search for compilers and you may find out more.