Stephen Wille Padnos wrote: > Gene Heskett wrote: > > > >> Someone, maybe you, once made the comment that an accessory card for the >> parport was actually faster than the onboard. Have any tests been made to >> somewhat define what the possible speedups might be? 10% isn't much, 400% >> would solve real problems I'd think. >> >> >> > Jon Elson has a lot of eperience with this, as well as Peter Wallace. I > don't think it was me :) > A plain old I/O card with 8255s or similar, connected via PCI bus, > should be faster than a parport. Parallel ports have extra delay logic > to make sure they don't exceed the parallel port specs, which may be the > main reason why they all take around a microsecond (+/- a few hundred > ns) per I/O operation, regardless of bus or CPU speed. > > Right, very likely me. The on-motherboard parallel ports on many CPUs are still on a PCI-ISA bridge, using an ISA multi-I/O chip. The ISA-bus slows things down quite a lot. Contiguous transfers are about 1 us with the mobo ports. They can go down to about 640 ns with a PCI port. This includes cable round-trip delays and the delays on my board. This could be completely different using a different IEEE-1284 target device. So, that is close to a 2:1 speedup. Using a 600 MHz Pentium 3 CPU and a mobo parport, EMC2 could handle a 3-axis servo loop on my USC or UPC controllers in about 120 us. a 5 KHz servo update rate is reasonable on such a system, but EMC defaults to 1 KHz. With a PCI plug in card and a slightly faster CPU, the 3- or 4-axis servo time would be down to about 60 - 80 us. You could do 10 KHz on that, or go above 4 axes at 5 KHz. Except in special cases with a large amount of I/O or many axes, this just isn't a concern, 5 KHz is more than almost anyone needs for these kind of machine tool projects.
Jon ------------------------------------------------------------------------------ _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
