Benchmarks I
Pifast, our first practical test, simply calculates the constant Pi to the desired number of decimal places. I've chosen 10 million using the fastest method possible. Memory bandwidth has historically played a major part in this benchmark.
First thing to note is that the nForce2 EPoX is slightly slower than the KT400 variety when running a single module of RAM asynchronously at DDR333. Running 2 modules of RAM synchronously at the same FSB, though, gives the nForce2-based motherboards around a 2-second lead. By having a potential 4.2GB/s on tap from the twinbank memory running at DDR266 speeds, and the Athlon only able to use 2.1GB/s or so at 133FSB, we have enough left over to maximise performance.
Next we'll turn our attention to MP3 encoding. We're benchmarking by encoding a 600MB+ custom WAV file (U2's Pop album, incidentally) into 192kb/s MP3 using the LAME 3.92 encoder and Razor-Lame 1.15 front-end.
The Athlon, due to its greater FPU ability, takes a decisive lead over the comparison P4. MP3 encoding loves pure MHz.
On to the SETI benchmark. I'm an avid SETI runner now so this was of particular interest for me. This takes a while to complete.
The nForce2-based motherboards show very similar performances. They both beat out the KT400 and P4-based systems comfortably. 5 minutes faster on a work unit may not sound like much, but it does all add up. The gain simply comes from better bandwidth usage.
It's nice to see that the EPoX nForce2 motherboard can put 1.3FPS into the KT400 variety when running the CPU at the same speed. Again, the extra bandwidth on offer with no synchronisation penalties pays dividends here.