Kenneth Lerman wrote: > John, > > When I think of that approach, I don't think rewrite; I think port. Of > course, depending on what processor the USB device uses, that might range > from difficult, to impossible. > > If the USB device has floating point and is reasonably fast, it should be > doable without much rewrite. Without floating point, it becomes > significantly more difficult. > > Ken
Perhaps. But still a non-trivial task. There are quite a few Kbytes of shared memory containing data that passes back and forth from user space (task, etc) and kernel space (motion controller). The motion controller is also quite complex, and takes full advantage of the speed of modern desktop CPUs. Just because it runs OK on a 200MHz P2 doesn't mean it will run on a microcontroller. I'm away from my EMC machine at the moment, but if somebody would run EMC and look at the HAL parameters motion.controller.time, motion.controller.tmax, motion.command-handler.time, and motion.command-handler.tmax, they could report how long (in clocks) each of those blocks takes. (controller and command-handler do not include the PID and/or stepgen and other HAL stuff, just the trajectory planning and core motion control.) I seem to recall something like 70,000 clocks, but I don't know if that was for controller or command-handler or both. At 200MHz, that is only 350uS, making 1KHz control very practical. But simpler processors will need more clocks to run the code, and will have fewer clocks per second. Regards, John Kasunich ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ _______________________________________________ Emc-developers mailing list Emc-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-developers
