On 3 December 2011 13:11, Yishin Li <y...@araisrobo.com> wrote: > approach #1: > * have negative request-velocity for BACKWARD motion of current TC > * replace tc-queue with circular buffer for bi-directional move > * figure out how to feed the bi-directional circular buffer > > approach #2: > * have negative request-velocity for BACKWARD motion of current TC > * a dedicate FORWARD-tc-queue for FORWARD motion > * a dedicate BACKWARD-tc-queue for BACKWARD motion
This ties in rather tightly with the "jog/touch-off while paused" issue and the difficulty with "run from line" both of which I think are being looked at by other people, though I don't know if any coding has been done. I have been thinking about it, and I think that the second idea is probably best, a fixed "history" queue that describes the path that exists in the metal, and a more flexible, re-computable "future" queue that depends on the G-code not yet run. The "future queue" has to be re-computable, as so much can change, including the whole flow of the G-code, depending on variables and offsets changing. "Run from line" is a particularly interesting puzzle, but not directly linked to your question. I think that you would probably want to link this into "adaptive feed" and "feed over-ride", simply allow these to go negative. -- atp The idea that there is no such thing as objective truth is, quite simply, wrong. ------------------------------------------------------------------------------ All the data continuously generated in your IT infrastructure contains a definitive record of customers, application performance, security threats, fraudulent activity, and more. Splunk takes this data and makes sense of it. IT sense. And common sense. http://p.sf.net/sfu/splunk-novd2d _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
