run_ga.py runs very slowly in EvoGym 2.0 compared to 1.0

[ShuleiLiu]

I noticed a significant performance difference between EvoGym versions:

When running run_ga.py in EvoGym 1.0, the program executes very quickly. However, in EvoGym 2.0, using the same script and default parameters, the runtime becomes much slower.

Is this slowdown expected due to changes in the simulation or default settings? Are there recommended parameters or configurations to achieve performance similar to version 1.0?

[jagdeepsb]

num-cores = number of robots to train in parallel. Note: the total number of processes created will be num-cores * n-envs (as specified below in the command line)

Have you played around with these params from the docs?