I've been doing lots of Perl-simulation runs on my laptop, investigating how the amount of recombination per cycle affects the equilibrium accumulation of uptake sequences. I had originally thought that it might just affect the time to equilibrium, but instead it dramatically affects the state at equilibrium. More recombination gives a higher score, with score being independent of mutation rate over at least the two mutation rates I've been able to test (0.01 and 0.001 mutations per bp per cycle). This will become a nice graph.
My queries about finding a faster computer to run the simulations on gave lots of replies.
(I've deleted the rest of this post because I absentmindedly posted the same info a few days later.)