The most surprising thing in the last issue was
P. Holy’s letter about Hisoft.
[Ed. – See
in this issue too.] You may remember me waxing ecstatic
over their Z80 editor-assembler package for use with CP/M, (“the editor is a joy to
use” – shock horror probe!) and saying what good software it is. I have written to
them twice, asking fairly difficult questions, and they answered both of my
questions promptly. They were very patient when my own carelessness resulted in my
being unable to operate the system properly, too! And they put their telephone
number at the top of their letters to me... In fact, my confidence in them is such
that I ordered their new Pascal 4 compiler for CP/M systems as soon as I saw the
advert (which I notice has their Ansaphone number in it) in
Personal Computer World.
The compiler output is fast machine code, as can be seen by the P. C. W. benchmarks
for the earlier version for use with Nas-Sys. And at £40, it is incredibly cheap. A
proper review will trip from my Naspen as soon as I have had a good go with the
compiler. Unfortunately, I saw the advert too late for this episode’s deadline.
Did you see the nice things Uncle Dusty wrote about my item in Micro-Power?
The “copyright notice” that
produces when given the Y command refers
only to the part I wrote, not the whole thing. It is there primarily so that people
can see how to add their own commands to the system – I’m a great believer in
learning by disassembling!
A particularly nasty gremlin.
I recently joined the Taunton Computer Club, which meets at the Somerset
College of Art and Technology on Tuesday evenings, from 6pm to 9pm (at which latter
time the “serious” members transfer to the staff bar for further learned discussion,
or something...) and took Marvin in to give a sort of demonstration. Everything went
well for about an hour and a half, then the system went haywire. After a few minutes
head-scratching (my head, not the ones in the disk drive, which I clean much more
carefully!) and reset pushing, the system settled down and ran for the rest of the
evening without further problems. So I assumed it to be mains noise, caused by all
the other machines on the same ring (four RM*8*Z’s, an A*pl*, and a couple of
Si**la*rs, if you must know!) and carried on demonstrating my latest bizarre
At home again, the same problem kept cropping up, and finally I had to do
something, because it was driving me daft, and it managed to erase an important file
from a disk. So I stripped the system down, and found that the cooling fan had
extracted all the chalk dust from the atmosphere of the classroom, and stuck it to
the back of the processor board. The lessons here seem to be that if you fit a fan,
a filter needs to go in too, and it isn’t always the electricity board’s fault. Mind
you, it often is their fault, as has been pointed out to me recently, in a letter
from a member of the Merseyside Nascom Users Group. They (the electricity people,
not MNUG!) are rumoured to be in the habit of constantly sending thumping great
pulses down the line to operate their remote controlled equipment. Own up, any SWEB
employees who are reading this, and feel they can comment on these libels...
Further delusions of grandeur.
Ever since I saw the article in Personal Computer World a while back on how
to interface a Z80 and memory to a 6800 based system, I have been thinking about
adding more processing power to a Nascom. Not, I hasten to add, because of any lack
of power in the basic system. Possibly you will have heard that some programmers
have developed multi-programming systems for Nascoms. Their software will run more
than one job at once, and I take my hat off to them, in a figurative way. They have
done something on a micro that some mainframe people would have us believe can not
be done. But is it something that needed to be done? I think (and this is definitely
a matter of opinion) that when the system is to run another job, given the cost of a
Z80, it is a better idea to add another processor. I picture an add-on board
carrying a processor, some memory and a simple control program. The processor, I
suppose, will probably have to be a Z80. The main reason for this is that the Z8000
is still too expensive. There are several even more exotic possibilities, such as