John Kasunich wrote: > Jon Elson wrote: > >>John Kasunich wrote: >> >>>Jon Elson wrote: >>> > I have no idea how hard it is to do better with this lookahead, >>> >>>It's hard. :-( >>> >>>That's the rub - if there is a discontinuity the machine has to slow >>>down. But it doesn't know there is a discontinuity until it gets there. >>>(Or in EMC's case, until it gets within one segment of there.) >> >>Well, then, what is the difference between G64 and G64 P0.0005 ? >>In this particular program, it makes a 4 minute difference, or >>a factor of 15:1 ! it still didn't get up to the programmed >>feed rate, but it got a lot closer. I think I must have left >>the file with the 60 IPM feedrate in it, and it did 6.28 inches >>in 17 seconds, or 22 IPM. >> >>I can understand horrible performance in G61 mode, that would be >>expected. >> > > > Maybe the lines in this program are close enough to collinear that when > you give G64 a tolerance to work with Jeff's code is combining bunches > of them into much longer segments that allow much higher velocity. > Well, that is the idea. With it taking 10,000 chords to make a full circle, they should be VERY close to co-linear. I think these chords are all within .0001" of being colinear. > I think G64 without a tolerance can only blend moves that are exactly > tangent - that would mostly apply to paths made up of lines joined by arcs. Well, at least near the "corners" there were a lot of segments that were truly co-linear due to roundoff, but it didn't seem to speed up at those spots in G64.
Jon ------------------------------------------------------------------------- This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Don't miss this year's exciting event. There's still time to save $100. Use priority code J8TL2D2. http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
