@@daliasprints9798 why not, you could gain efficiency and reduce print times, use less material when unnecessary, use different types of supports when needed, all sorts of stuff, thats the entire point.
@@richardjohnson8009 Because you don't understand the tools and are just blabbering about the popular thing of the day. All of the constraints here are known and physically modeled. You don't have to apply magic black boxes trying to get a neural network to rediscover the patterns that we already know that are already consequences of sound physical models. Getting it to do that would be far more work developing criteria for training (to evaluate the quality of the print) than just using what we already know.