Good morning and evening!

I got a procedure f1, which takes two numbers and outputs one. Just a numeric computation.

NLPSolve(f1, -1..2, -2..2);
Warning, no iterations performed as initial point satisfies first-order conditions

but plotting a 3D graph shows me a good, smooth minimum. Setting starting point does not help.

It seems that NLPSolve thinks that gradient is equal to zero, but why?