Till Harbaum wrote: > Hi, > > isochronous transfers are imho currently not supported by libusb. So it needs > a new kernel driver (which isn't that much of a problem, agreed) and wouldn't > be portable (also not a problem since emc isn't portable at all). So yes, this > would be possible. > > On the other hand i am not quite sure of isochronous provides a low enough > latency. Low latency typically is what usb interrupt transfers were made for, > especially > if you want to report something back from the device to the host. This is > probably > what you want to report end switches. I'd be very interested in such a driver. This may be useless for sending software step/direction out to a USB parallel port, which seems insane anyway.
The general scheme of my 3 PPMC board products is to read all inputs in a burst, compute new velocity and digital outputs, and then send that in a burst. As long as the transfer is completed within the servo update interval, then the latency is not a problem. The PPMC and stepper controller don't actually update the velocity output until the NEXT read encoder position command is given, so the time jitter caused by latency should be limited. It seems if the USB restriction of 1000 requests a second is lifted, this would work with a very simple and cheap USB controller chip. Jon ------------------------------------------------------------------------- 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
