21 Aug
2006
21 Aug
'06
3:15 p.m.
Tony Duell wrote:
Well, the keypad was terrible (I ended up soldering 20 switches from
Maplin onto the MK14 PCB, the holes and traces were there for them...).
There were some very marginal bits of logic design (hint : a '157 is a
multiplexer, a '175 is a latch, but the MK14 uses the former as a the
display output register. It works for _some_ makes of '157...). The CPU
buses are not broungt off-board, so expansion was a pain. And so on. I
wish I'd saved a little more money and bought an Acorn System 1.
I bought an MK14 back when it was current. That
put me off Sinclair for
life...
Out of interest, what was so bad about them that wasn't typical of that class
of machine around that time? Admittedly without having been involved at the time, it seems with hindsight that there were an awful
lot of similar machines
around that sort of era with odd design quirks, very limited power,
poor-quality PCBs etc.
Having collected a number of machines from that sort of period, I can
safely say that Sinclair were _by far_ the worst. And don't get me
started on his pocket calculators. I think I can get more accurate
results from a good slide rule.
More common is that the regulators have been bypassed
or removed entirely and
an external supply is used. It's rare-ish to find an Atom with regulators and
heatsink still intact (maybe 25% are like this). Rarer still is to find an
Atom using an external supply where someone's actually done as the manual says
and indicated on the back that the machine's wired for 5V regulated input
rather than 8V unregulated - I expect there are a lot of toasted Atoms still
floating around!
Releated to this was a horrible bit of Torch design. As you may know,
Torch sold an upgrade for the BBC micro which added an internal PCB
containing a Z80 running CP/N (no, that's not one of my typos...). The
Beeb power supply couldn't really supply that as well, so what you did
was remove the Beeb PSU altogether, connected a cable to the power
connections on the Beeb's mainboard, and run the whole thing off a PSU
in the (Torch-supplied) disk drive unit.
The problem was that the cable supplied was essentially mains cabled,
terminated in a Bulgin connector that was sometimes (albeit rarely) used
for mains. And of course some people, finding a machine after <n> years
and not realising the PSU had been removed, promptly connected this cable
to the mains. Appliing 240V AC to the 5V line does not improve matters...
-tony