My main machine has about the same build and speed as your Pentium 4. It has superb performance. Then I have a second Pentium 4 at 2.8 GHz that cruises nicely. I round out the stable with four relatively short-legged (at least in terms of this program) CPUs huffing and puffing along including a Celeron at 1.3 MHz, an Athlon 2100+, a Sempron 3100+, and a two year-old Mac iBook (I know nothing about its CPU). Each machine has about 1 MB of RAM. The final four together appear to have about the same output as the slower Pentium 4.
I have a hunch that if I were to study the FAQ (which I just haven't gotten around to doing), I'd discover the program is really optimized for Pentium 4 based machines.
There are probably tricks for tweaking all the machines to greater output, but I'm not that dedicated of a nerd.