"In 1980 I attended a Crusader Union Computer
Camp for two weeks during school Holidays. It was great fun and eye
opening. There were all sorts of enthusiasts, Uni students and businessmen
that were in the early IT industry and wanted to help others get started.
I really didn't know much about computers except that they were complex
electronic machines AND I loved electronics.
After the camp, my Dad purchased a System-80. I
don't know why he did that. I don't recall asking for one - probably
because I knew they were expensive. I am sure that Dad didn't know what
the end result of this purchase would be.
My brother and I quickly learnt BASIC, Z-80 machine
code and began decoding the 12K ROM. Occasionally we would have a competition
that involved agreeing on a specification for a machine language program
and then independently writing a program that would run the fastest
and be the smallest size. The winner would have the program that executed
in the fewest number of CPU states calculated manually. One competition
was to treat a byte as two nibbles and each nibble described setting
or resetting a graphic bit in one of the 8 surrounding pixels. My brother
won by one state per loop as I recall by careful assignment of registers
so that he could use the DJNZ opcode which combined two instructions
into one saving 3 states. Another competition was to clear the screen.
My brother again had the fastest code, but it was one byte large in
size so it was a draw (he doesn't think so).
We did a lot of machine code programming and experimented
with 'hidden' Z-80 instructions to do with IX and IY registers. I learnt
a lot and had a good understanding of how high level languages are eventually
decoded and interpreted or compiled into machine code.
Then I became interested in the hardware. I somehow
contacted a guy who worked for Dick Smith and he supplied me with Technical
Bulletins and even circuit diagrams which I slowly studied and eventually
understood everything about the machine from it keyboard interface,
tape deck, memory, video memory, character ROMs etc. I added 32k to
the 16k by piggy-backing RAM (4116?) chips onto the existing RAM chips
and then using spare gates to build the necessary CAS signals. Another
project was to increase the speed of the System-80. It ran at 1.7 MHz
which was derived by dividing a standard 10.something MHz crystal by
6. I used the Page button and spare logic circuits onboard to bypass
a divide by 2 stage to run the PC at 3.5 MHz. It worked but the problem
was that the tape deck signals were software timed and when the Z-80
ran at x2 it could no longer read tapes. So I had to load in the program
at normal speed, and then switch into 'turbo' mode. Half the time the
machine would lock-up which I later realised was caused (mainly) by
switch bounce.
I eventually purchased another System-80 in 1984
and also a hi-res unit. A year later I bought a IBM clone and I remember
being disappointed with it's speed.
I still have both System-80s and I think I have
most of my old tapes as well. I should get them out from storage (they
are in a friend's shed) and let my children see what the early days
were like before disk drives."
