John Coffey wrote this note to his customers, in part to explain why it was a few months late.
I became interested in computers in 1974 when I saw some of the first Apples demonstrated at a mall in my hometown of Columbus, Indiana. I learned a little BASIC then. Later I went to the library and read books in more depth even though I was only 14 at the time. (In those days “computer graphics” meant a printout on a teletype.)
At that time I wasn’t able to afford my own computer since they cost thousands of dollars. But in 1978, I was very fortunate to meet Bud Phillips, who owned an RCA “ELF”. Bud was kind enough to lend me his machine for a week and I taught myself 1802 machine code.
While I was a high school senior in 1979, our school purchased seven TRS-80s. I was then able to master BASIC more fully and I started learning Z80 machine code.
College temporarily postponed my computer hobby. My major in college was biology and I had no access to any computer for my first year. But in 1981 I met someone who owned an OSI-C2P and this gave me an opportunity to learn 6502 machine code. But after a while I no longer had access to his machine.
And then a “miracle” happened. In January of 1982 I met Mike Moore who owned a computer that really impressed me. It was a ZX80 made by some yet unheard of English company called Sinclair Research. I became hooked by the power and user friendliness of such a little machine.
Although I still could not afford a computer of my own, Mike gave me frequent access to his. I kept impressing him with the programs that I would write. For example, in September of 1982 I began developing an early version of the Z80 assembler that comes with this tape. I also wrote the “EXTENSIONS TO BASIC” routines that would later be published in SYNC magazine.
In January of 1983, I was finally able to buy a TS1000 at $99.95. I spent the next six months writing programs in just 2K memory. Many of these programs were in machine code and would later be included in my SUPERTAPE.
Last summer the price of 16K RAM packs went down to $35 and I bought one. It was only then 1 was able to fully debug my assembler which was a 16K program.
Something else significant happened last summer. I started to advertise the programs I had written. Initially I only sold listings
which was the sole reason for buying a 2040 printer that summer. I didn’t get much response, however, so I switched to selling tapes which proved more profitable. Over the months that followed, I brought all of my TS 1000 programs into one package of 45 programs called SUPERTAPE.
In the meantime, I anxiously awaited the arrival of the TS 2068. Timex had promised September delivery and did not come through till late December. It looked then like the 2068 was going to be a pretty big deal, so I bought one at Sears for $191. I began immediately converting most of my TS1000 programs to the 2068 and by the end of February had come up with a package for the 2068 called The GREAT GAME AND GRAPHICS SHOW which was based upon my SUPERTAPE for the 1000.
Also, about the end of February 1984 Timex announced their withdrawal from the computer business and SYNC magazine quickly folded. Despite the bad news, I started advertising that I was going to convert my TS assembler to the 2068. In March someone who had responded asked if I was also going to write a compiler. I told him I had no plans to do so, but things happened very quickly after that to change my mind.
During the spring of 1984, I was finishing my senior year as a biology major at IUS in New Albany, Indiana. The assembler was proving to be very difficult to convert, and I was frustrated at not being able to use machine code on the 2068. Just prior to my spring break, a novel idea came into my head. It was more or less a whim, but I had figured out an efficient way to develop a compiler and had the first version finished in ten days.
The first version of the compiler was intended strictly as an alternative to assembly language. It used only a very small subset of BASIC and was pretty much a low level language that was independent of the 2068 operating system. It didn’t have any of the nice features that a high level BASIC would have.
My friend Mike Moore strongly suggested that I add more commands to the compiler. I complied with his request, but it took a great deal of work. I had to link my compiler code with the 2068 operating system and even before I could do that, I had to write a mammoth “semi-dissassembler” that would find, cross reference, and label most at the subroutines in the 16K ROM. I did even own a disassembler at the time.
The end result of all of the above work was version 2 of my compiler. I sent this version to TRIANGLE SINCLAIR USER GROUP for review. I graduated from college about the same time and so I had more free time to do my assembler conversion. I also started advertising the COMPASS package.
TRIANGLE SINCLAIR USER GROUP informed me that they thought my compiler could be a little better if I added string handling. So I began working on a third version which included string handling and some additional commands. It took about 2 months longer than I thought to finish this project.
6-5-1984
A Brief History of JRC Software