"Bob La Londe" wrote in message
...
"Gunner Asch" wrote in message
...
On Mon, 18 Jul 2011 20:16:33 -0700, "Bob La Londe"
wrote:

"Gunner Asch" wrote in message
news On Mon, 18 Jul 2011 18:13:58 -0700, "Bob La Londe"
wrote:


USB to just about any sort of output..serial, parallel etc etc are
commonly available at any computer shop

LOL. Yeah right. Try running a stepper driver off of one.

I have a number of USB adapters that have an external power supply..so
the power needs are NOT taken from the computers USB port.

Want some links?

I want some links to some that will reliably control 3 or 4 axis of
steppers
for a million lines of code, and probably a several billion signals, and
will work with just about any step direction driver.

I think you totally missed (or chose to miss) that a cheap generic USB
parallel adaptor will not fill the need.


So USB 3.0 isnt fast enough?

Well, I'm not sure its strictly a matter of speed. It's a matter of
timing. I know plenty of people have tried adaptors with poor results.
With an "adaptor" the PC is still doing all the work and then the USB port
is "translating" it. The software would really prefer to control hardware
directly. I can ask over at the Mach Support forum if anybody has better
results with USB 3.0 adaptors, but realistically I can't see how. The
fundamental problem is the same. The SS takes a whole different approach.

There is also something to the way resources are allocated for USB as
opposed to a dedicated hardware communication port. I'll leave that to the
guys who know more about specifically USB to explain if they like. I know
if I send a signal to a pin on a hardware LPT port it (barring the speed of
electrons down wire) its virtually delivered precisely when its sent.

For a full size desktop its really a non issue if price is your only driving
factor however. I can still get a good quality hardware PCI parallel card
brand new for about $20. (I have three spares on the shelf.)