GPIB state machine info WTD

[Tony Duell]

> > I also designed a GPIB listener state machine in a GAL. Again I should 
> > have the equations somewhere...
> 
> I think I saw that somewhere.. A PET GPIB to Centronics adapter, right?

Actaully, a PERQ centronics interface, but I can't see any good reason 
why it wouldn't work with a PET (although it doesn't do any kind of code 
conversion.

> 
> > Be _very_ careful if you use designs intended for the PET. The PET was 
> > very forgiving in the timing of the peripherals. I have a commercial PET 
> > RS232 interface that contains a UART chip, RS232 buffers, a few TTL chips 
> > as SR latches to hold the talk and listen states, an EPROM as the code 
> > converter (PERSCI-ASCII) and talk/listen address detector) and another 
> > EPROM as the handshake control logic. Due to the fact there are no 
> > latches in the latter section, the timing goes all over the place. It 
> > works on the PET, it doesn't work on any HP machine I've tried it with.
> 
> Well, if it only just works on a 2MHz 6502, it's a sure bet it'll probably
> fail at 8MHz.
> I only wanted the IEEE488 part of the interface anyway - the CPU interfacing
> gubbins2

No, you misunderstood me. There is no CPU involved in this serial 
adapter. The EPROM is the logic for the GPIB handashake. It's the GPIB 
timing that's way out. It works with a PET as the other side of the link 
because the PET does all the handshaking in software, so it's pretty 
forgiving about timing problems. But this seiral adapter will _not_ with, 
say, a 9914 as the other device on the bus..

-tony