Well, I have made a bit more progress! I printed out some more of the insane 3517-page manual, and figured out the most basic stuff in the GPIO registers. I made a program loop that flips one GPIO bit on and off, and it runs in 260 ns. That would certainly be fine for doing stepper pulses in software, and I think it is going to be OK for running the EPP emulation, too. I was hoping for even faster operation than this, since the CPU is supposed to be doing one instruction every 3.3 ns or so. But, unless the code turns into a boondoggle, it should be able to do as well as a good PCI parallel port card. Also, even with Debian Linux running, a "heartbeat" blinking the user LED, network access, etc., the timing was REALLY solid, even though this was NOT a RT thread.
I need to play around with this a bit, and try some linear program sequences with no loops or if-then conditional branching, to get an idea what effect those have on program timing. Then, I need to see what it takes to get 8 contiguous register bits sending at the same time. I have found a TI bus level translator that handles 1.8 to 5 V conversion, it is a close relative of the part I use on the USC and UPC boards, but is designed to go down to the 1.8 V levels. Jon ------------------------------------------------------------------------------ Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july _______________________________________________ Emc-developers mailing list Emc-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-developers
