Hi all.
I want to find the minimum of a function which has multiple local minima. When I call amebsa the routine ends after only two function evaluations. I have tried all kinds of temperatures from plus to minus infinity and I have put the amebsa call inside a loop where temperature is steadily decreased, while all other function arguments are left untouched between calls, as the book tells you to do. Nevertheless from the second call onwards inside this loop, amebsa does no more function evaluations.
I have obtained a (likely) global minimum value with repeated applications of the standard simplex algorithm amoeba. When I put this value in amebsa, then and only then it did 68 function evaluations and returned the same minimum value.
Is the amebsa code completely useless or am I missing something?
Thanks for any help.

P.S. Any links to alternative simulated annealing/global optimization C++ code will be greatly appreciated