probably has nothing to do with it, just an example
neil.challinor
there is a bug that I can't reproduce (but always occurs with a constructed cylinder)
If you save a measurement program from 2019 or 2014 to 2021.2, and this measurement program contains a constructed cylinder made of circles. then from time to time there is a crash when you modify the circles from which the cylinder is built. the crash occurs when you want to save. which then always leads to the file no longer opening (with the error message "Bad archive"). Document recovery then saves the day.
If this error occurs with a program, it really crashes every time you modify this specific cylinder. It only works if you completely delete the cylinder and rebuild it.
I sometimes feel like modernizing old measurement programs, on such days pcDMIS crashes almost every 2 hours, since I know that it's the cylinder it's not really a problem, just annoying
the cylinder is evaluated with geo tol and before the "SaveAs" with FCF, maybe that has something to do with it.
anyway
mostly input are 2x "auto.circle"
its a normal cylinder "BF" and "LEAST_SQR"
the error only occurs with 1 cylinder within a program, never with several. If I ever find a program like this again, I can send it to you. but since I've modernized almost everything in the meantime, I don't think I'll find any more
Yep. That's the bad part. We could use sampling hits using autocircle, cylinder, cone prior to 2020 R2 where there was a boss in the way or not because the probe didn't go to the center of the hole & then moved back to take hits in the hole. Tech support already checked those troublesome programs but said they are fine. Their conclusion is there are bugs behind the nuts & bolts of the software. At least we know in the lab it's not us but Hexagon software developer geniuses.