Well, there was a guy here in Denver who made a great deal of money from the
Commodore community with his Commodore-HPIB==>Epson-printer interface. He
started a pretty large company from that one single device for the
Commodore. I'm not even sure that the original interface, unmodified from
the COMMODORE version, would talk to HPIB devices.
I'll have to investigate the process of writing control routines for the C
and Pascal versions I routinely use. If there's QBASIC-compatible BASIC
software out there that may be worth a look. HP did like BASIC back in the
'80's.
I once used the serial interface on my logic analyzer to help me with
producing a "clone" of a dongle for one of the many dongle-protected
sogtware bits I wanted to investigate. This provided a means to do that,
since the software under Windows (3.1x) was considerably more trouble to
patch than the dongle was to emulate. I'm likely to find that the GPIB is a
bit faster than the serial interface. That might open some doors.
Dick
----- Original Message -----
From: Tony Duell <ard(a)p850ug1.demon.co.uk>
To: <classiccmp(a)classiccmp.org>
Sent: Wednesday, April 12, 2000 5:25 PM
Subject: Re: HPIB to HPIL adapter
>
> I was exposed to HPIL back in the '80's and, while there were some fancy
> instruments, e.g. 6-digit DVM, I've not seen it used since then. In
fact,
HPIL was one of the many things that HP did _right_ in the mid 1980s.
It appears (to the user/programmer) as almost a serial version of HPIB,
and it supports 31 peripherals (960? if you use extended HPIL addressing
IIRC) on single loop.
AFAIK it was only ever really used by HP (there were 3rd party products
that used it, but all that I've seen were peripherals to HP calculators),
although HP did document it completely and sold the custom (1LB3-0001
IIRC) chip that acted as an interface between the loop and a normal
microprocessor bus.
I believe that neither these chips nor any HPIL devices are still in
production, though.
Since the higher levels of HPIL and HPIB are almost the same, the HP82169
HPIB interface allows you to control HPIB instruments from an HPIL
controller or vice versa with few problems.
the GPIB is something I've seen surprisingly
little, except on equipment
Well, almost all digitally-controlled test equipment supports it.
As regards computers, just about anything HP would either have an GPIB
port as standard or at least have one as an option. All 'classic' PERQs
have
GPIB as standard as well. Commodore PETs (and
P500s...) have a somewhat
mangled varient of GPIB that works most of the time with standard GPIB
peripherals (although the reverse is not always true -- there are PET
peripherals that do not work on normal GPIB controllers). DEC made GPIB
cards for the Unibus (IB11) and Q-bus (IBV11). And of course there are
ISA cards for it. And _dozens_ of machines that I've missed...
> The LabView is not irrelevant, but only because there are people willing
to
> pay to have it used in their installations.
I'm perfectly satisfied to
use
> the GPIB command set as part of a program I write
myself. I've found
that
> approach quite satisfactory, and there's
certainly a shorter learning
curve
than with some
GUI-thingie that I can't manipulate comfortably because
someone else designed it.
I must admit that from what I've seen of Labview, I don't like it. If I
want an instrument front panel kind of interface then I'll reach over and
turn the physical knob on the physical front of the instrument. If I want
to automatically control the instrument then _I'd_ be happier writing a
program in some suitable language to do it.
>
> Because I've had so little experience with the GPIB, and because I've
got no
> doc's on the National Instruments GPIB card I
have, I'm at the mercy of
> those who may have and be willing to "share" whatever details and
drivers
they have.
I'm open to any suggestions and your "spiritual guidance" is
always welcome.
I wonder how hard it would be to figure out how to program the card
directly? Does it use a standard GPIB controller chip (probably a TMS9914
or an Intel 8291/8292 or an NEC 7210)? If so, get the data sheets on that
chip and then figure out how the address decoder, etc, on the NI card
works (ignore DMA if you can, for all some GPIB cards support it).
Then it's just a matter of writing functions to program the chip
appropriately to act as the system controller and to allow you to send
bytes, receive bytes and send commands. None are particularly hard to
write.
-tony