Hi Renaud,

If you are starting with the same inputs and the same targets, the result will be the same every time. In an optimization, the starting conditions can be critical, and starting from one set of inputs can lead to a very different system than you get with a different set of inputs. Like Mark suggested, you could work with a global search or use ZPL to make some randomized starting conditions. But if I understand your description, you might also just start with some different Zernike parameter values and see if they lead to a different wavefront. There are lots of Zernike-defined wavefronts that have the same RMS, so I’d say just try a few different starting values for your variables.