[Taqwus]

It would likely be more realistic if there were nonlinearities and synergies -- too little or too much attention paid to one research area having lower efficiency, and possibly the occasional benefits from cross-pollenation -- but it'd also lead to optimization-hell where players run programs to find maxima.

I sort of like the ol' non-determinism of MOO2:BAA (probability of breakthrough per round once threshold achieved), but I rarely went as far as to take Uncreative...