Marius, Q1. I'm sorry to say I don't know. I do remember that thread that you started. What did Rob the TP author have to say about that?
For Q2, the motoin velocity is indeed limited to the definition of the joint speeds in non-trivial kins. During configuration you specify speed/accel limits for each axis and each joint. It appears that while jogging in joint mode, you get the joint defined speed limits, and in world mode you get the axis defined speed limits. On Sat, Jul 25, 2015 at 6:13 AM, Marius Alksnys <marius.alks...@gmail.com> wrote: > Curtis, > > great plans! I have two questions though: > Question 1. > Is there anything better in TP than main branch in respect to > simultaneous motion in multiple axes? > > Now at least with trivial kins a motion with any from XYZ + any from ABC > works in a manner to be able to stop ABC at the end of current or next G > code line. > > I raised it in earlier thread "Rotary axis motion too slow". In that > machine and its typical application G-code machining time reduced in > times after I exchanged Y and A axes in LinuxCNC. (Normally linear Y is > steady and rotary A is revolving fast with Z fast moves and slow X > move). People who look at the machine now are amazed how fast it works > :) But the workaround is not comfortable. > > Question 2. Is trajectory motion velocity limited to joints capabilities > in non-trivial kins? > > > On 07/23/2015 04:33 AM, Curtis Dutton wrote: > > I have rebased joints_axes7 onto master and renamed it to joints_axes8. > I'm > > running it in production and wanted to pick up the latest changes in the > > trajectory planner. I also included a few other commits. It compiles on > > wheezy and all tests appear to pass. > > > > It's available here. > > > > https://github.com/curtdutt/linuxcnc > > > > My next major task will be to address non consecutive axis names defined > > for lathes. > > > > I would like to get these changes merged into master ASAP and i think > that > > should happen before tackling homing multiple joints on a single axis > > because linuxcnc currently doesnt do that at this time. So there wouldnt > be > > a regression by merging. > > > > If anyone has any other things that they think should be fixed before it > > can go into master please let me know as I will try to address those > issues > > as well. > > > > -Curtis > > > ------------------------------------------------------------------------------ > > > > > > > ------------------------------------------------------------------------------ > _______________________________________________ > Emc-developers mailing list > Emc-developers@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-developers > ------------------------------------------------------------------------------ _______________________________________________ Emc-developers mailing list Emc-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-developers
