13 Oct
2002
13 Oct
'02
10:30 p.m.
gil smith <gil(a)vauxelectronics.com> wrote about HP-8x ROMs:
The roms have +12V, +6V, and -5V
power. There is an 8-bit bi-directional bus, and four non-overlapping
12V clock signals. There is a "load-memory-address" line, a
"power-on"
line, "read" and "read-control" lines, and even a "write"
line (I don't
know why write is available on a rom).
This all leads me to believe the roms are pretty specialized.
Yup, custom for the HP-8x, and CMOS versions for the HP-75. AFAIK they
aren't used in anything else, although sometimes Corvallis division
custom parts wound up in instruments, like the HP-35 chipset (with
different ROM code) in the 1722A oscilliscope.
Since
there are no lines to enable a specific rom socket, I think the 85 must
poll for roms using fixed address ranges or something -- this implies
that the roms contain address qualification circuitry of some sort.
They all map to the same address range, but have unique IDs that enable
the software to select one and deselect the others.
I'm just
speculating, but it would make the roms very difficult to duplicate,
since this is not a standard address-bus/data-bus (or even a typical
multiplexed addr/data bus).
Yes. I'd be interested in hearing how your buddy did it. I suppose
one could write an assembly-language program to dump ROM data, but
otherwise one would have to build some specialized hardware.