24 Jul
2010
24 Jul
'10
7:58 p.m.
On 7/24/2010 8:47 PM, Liam Proven wrote:
On 25 July 2010 02:32, Sridhar Ayengar<ploopster at
gmail.com> wrote:
If there is interest, I am positive a solution could be created that
would allow use of all 25 pins of the parallel port, but, it would have
two drawbacks/showstoppers:
Most code that uses the parallel port as a GPIO port hacks at it using
the actual IO address (I forget the actual base addy for the PC port
right now, but everyone knows what I'm talking about. It *might* be
possible to trap all direct access calls in that space and emulating the
IO store/read calls, but that carries a very high overhead.
Latency. I'm not sure you could "flip" a line on the emulated parallel
port nearly as fast as you can do on a real parallel port. If some
implementation used the various lines as a serial line, for example, it
would be very hard to get the IO packet for a "pin HI" and then another
one for "pin LO" to the USB device in the required timeframe.
The latter is most likely the showstopper. As an example, for years,
CBM folks have used a cable called the X1541 cable to communicate with
IEC-based drives (1541, 71, 81, etc.) from the parallel port. Due to
the transition frequency, it was impossible to directly translate that
to USB. The solution was to move more of the low level work to a small
uC attached to the PC via USB.
Jim
--
Jim Brain, Brain Innovations (X)
brain at jbrain.com
Dabbling in WWW, Embedded Systems, Old CBM computers, and Good Times!
Home: http://www.jbrain.com
Liam Proven wrote:
I don't know, but AFAICR I have not /seen/ anything that works like
that. I suspect it'd be a lot more work to make a hardware-level
compatible USB<=>Parallel adaptor than a cheap'n'cheerful one that
just does enough for printing.
I have no idea on technical grounds, but inasmuch
as there are, or
were, USB<=>SCSI adaptors, I think it's /technically/ possible in
terms of electronic capabilities. What I suspect on no particular
evidence is that software-wise it probably isn't, 'cos I am /guessing/
that a USB device can't present itself as a port on 0x378 on IRQ 7
etc.
Well, wasn't it actually 0x3bc on the original PC? And I don't
see why a
USB device couldn't do that, given the proper drivers.