[tech] ARM 2 revival project

Nick Bannon nick at ucc.gu.uwa.edu.au
Wed Sep 24 11:51:02 WST 2014


On Wed, Sep 24, 2014 at 11:23:32AM +0800, Peter N Lewis wrote:
> On 24 Sep 2014, at 11:15, Nick Bannon <nick at ucc.gu.uwa.edu.au> wrote:
> > Hi! Mind if I bounce this through to the tech at ucc.gu.uwa.edu.au list?
> Not at all.

Great!

[BOB], [BG3] and I have started a little Monday night project... reviving
the ARM2! Built starting in 1988 by [JPQ], we all had a peer into it
last Friday after the 40th dinner.

I'm not sure if it ever received a good name as such..? Maybe that's a
TODO item.

If basic hardware permits, I'm sure we can get it doing _something_ -
it can't be harder than programming a Teensy, right? <grin>
Lovely clean ARM26 assembler, I'm looking forward to our first ROM dumps.

On Tue, Sep 23, 2014 at 09:59:18AM +0800, Peter N Lewis wrote:
> From Marcus:
> The ARM almost certainly it boots into a debugger talking on the serial port. There is floppy/scsi disk boot code in the ROM but I doubt it automatically triggers.
> 
> I would guess that the command is "b <something>" to trigger the boot. Either floppy/scsi or serial, no idea what options would be used to pick between them.
> The debugger might have a help with the "?" command, but if it exists it would be cryptic.

Thanks, [JMJ]! Many moons ago - you showed me it doing that very thing
and there have been no hardware or ROM changes since then. With the
possible exception of some connector wires falling off.

[BOB], [BG3] and I dedusted it and tested out the power supply over the
last two Monday evenings. It looks functional, there are definite clock
and data signals flying around it, visible on the Tektronix DSO, some
look cleaner than others.

We haven't made it spit out any identifiable serial boot messages yet.
There's a couple of potential serial ports, with a transmit line powered
at -12-ish volts.

I think:
  * one goes into the IOC; and

  * one goes into a pair of 1488/1489 RS-232 line (driver/receiver)s,
    which I can't see on the ARM.tiff circuit diagram.

Do you know which one the boot monitor/debugger might be listening to?
( [PNL] replies: No idea. )

> Hmm... I wonder did we support the Kermit protocol for downloading the boot image? http://en.wikipedia.org/wiki/Kermit_(protocol)
> 
> I do remember having to type a series of commands to ping the io ports to read/write disk blocks. We had a fancy FIFO chip that could grab the 512 byte block from the IO chips. And we had a fancy SCSI chip that handled all the scsi handshaking. And I had commands to read/write the FIFO to memory.
> 
> I know we had a keyboard on it, did we have a mouse as well? Probably not.	

Looks like a AT keyboard? There's power on the right line, haven't had
any responses back yet.

> Is the scsi disk still installed? If it's there, I doubt it's functional.

No, it's unplugged and possibly unterminated.

> I guess you could power it on and see if there a video signal. Probably best to unplug the power supply from the board first to make sure the power supply comes up OK.
> 
> Enjoy,
>    Peter.

There's a likely looking video socket, a DE-9 female, I think. Not sure
of the pinouts, we can probably work that out. How much code needs to
run successfully before there's a video display? Should there be VSYNC
and HSYNC signals running from power-on?

( [PNL] replies: Lots.  And again, no idea. ... Good luck )

<grin>

On Wed, Sep 24, 2014 at 09:40:42AM +0800, Peter N Lewis wrote:
> Another comment from Marcus:
> >If they really want it running then figuring out the custom PAL I made is going to be really interesting.

Heh. Well, it's an excuse to finally upgrade our ROM burner.  We've got
a working one already, it's just plugged into a ISA bus on a 486 running
DOS; with a floppy drive and no functional networking.

Currrently choosing between a:
  * http://www.conitec.net/english/galep5.php
  * ...or a cheapo Chinese TL866A or similar.

Nick.

-- 
   Nick Bannon   | "I made this letter longer than usual because
nick-sig at rcpt.to | I lack the time to make it shorter." - Pascal


More information about the tech mailing list