your best bet is to optimize as you're programming. don't wait till the end to try it.
my personal opinion is that optimize moves stuff around and if you haven't labeled your features very specifically it's very hard to track exactly what's going on. i tried it a few times and decided the few seconds I might gain was too much headache.
Yes, it's a neat idea but unfortunately makes for yet another feature on the very long list of PC-DMIS features that weren't baked long enough to be of any use.
If you're working with flat parts, clearplanes are usually much better anyway in my opinion.
I have had pretty good luck using optimize path on just small portions of a measuring routine, where the probe and probe angle doesn't change. You get some pretty wild moves and the order of the features can seem all over the place. Watching it run, you have no idea where it is going next. CMM operators had a fit
I agree with
BIGWIG7 , I didn't feel it was worth the headache. Going back and editing/revising those programs is a pain.
Back when I played with the optimize path features, I often wished I could have it just insert a clearplane command when it detected a collision, rather than insert the convoluted move points it would come up with. For all I know that is an option now. I'm still in v2020R1.
