Acorn IEEE488 interface

[Eelco Huininga]

> Message: 5
> Date: Sun, 11 Dec 2005 19:35:51 +0000 (GMT)
> From: ard at p850ug1.demon.co.uk (Tony Duell)
> Subject: Re: Acorn IEEE488 interface
> To: cctalk at classiccmp.org
> Message-ID: <m1ElWzC-000IyBC at p850ug1>
> Content-Type: text/plain
> 
> 
>>I've also got the schematic for the Acorn System IEEE488 card somewhere - it 
>>wouldn't surprise me if the BBC unit is based heavily on this (I don't think I 
> 
> 
> It might be, it might not. The 'Acorn' IEEE-488 interface for the Beeb 
> doesn't seem to be an Acorn design.
> 
> The article to find is in 'Electronics and Wireless World' February 1984, 
> pages 24-26 ('IEEE488 interface for the BBC Microcomputer'). It imples 
> the interface was actually designed by a company called 'Intellegent 
> Interfaces'.

Indeed it is. I ran the IEEEFS ROMs which are available at the 'The BBC 
   lives' site through a disassembler and both ROMs have a copyright 
message saying '(C) Intelligent Interfaces Ltd and Acorn Ltd'.
Could you scan the article? I'd be very interested in it!

> Anyway, there is a schematic. It's based round a 9914 chip along with the 
> 75160 and 75162 buffers. There's a 5MHz clock circuit, an address 
> decoder, a data bus buffer, and the normal circuit to de-glitch the page 
> select line on the 1MHz bus. That's all. Nothing remotely odd. 
> 
> Link S1 selects system controller mode, link S2 selects active pull-ups 
> on the IEEE488 data lines.
I've just finished reverse-engineering the board and it's indeed a very 
straight-forward design.

[snip]

> I do. I think I might also have the IEEEFS ROM that goes with it, but I 
> can't promise that.
I wonder what version you've got. The 0.2 ROM seems like a beta version, 
with lots of 'unused code' in between the different subroutines, and the 
0.5 ROM hasn't got any filesystem support, just support for the OSWORD 
calls.

[snip]

> Hmm, I see what you mean. The manual looks very comprehensive with lots of 
> examples - but I can't see where it justifies why the addressing is done in 
> this way.

I think it was done so the IEEE interface could be accessed as an 
ordinary file system (file handles &F0-&FF were reserved for IEEEFS IIRC).


Cheers,
Eelco