> The NURBS(G5.2) implementation of emc2 is actually breaking a NURBS curve > into a lot of STRAIGHT and ARC segments, which keeps accelrate and > decelerate while performing a NURBS motion.
do you mean that it keeps accelerating and decelerating with every line/arc segment? Isn't there something in EMC2 which keeps the velocity when two movements are within some epsilon of being collinear? > It's not smooth. More over, if > you feed a NURBS curve with too many control points, it breaks it into way > too many segments and behaves like it was crashed. For example, please try > the attached helix3.ngc. Thank you for another test case. I already have a simple one that follows a NURB fitted to a hand drawn square. Simple, yet induces all sorts of weird movements at some of the segment boundaries. Truly weird. It will likely be awhile before I can really delve into this again -- I'm apply for GSoC and an internship that is related to my thesis research, so I'm slammed at the moment. EBo -- ------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ Emc-developers mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-developers
