GO-88 and the development of computer graphics

Although I built my first computer at the Royal Aircraft Establishment in the UK in 1952, it was not until the development of the microprocessor in the 70's that it became feasible to construct a computer at home. At that time the S100 bus became popular as the basis of a microcomputer. This was a motherboard into which one could plug cards, each with 100 contacts. Because the connectors were standardised you could buy cards from a number of manufacturers and they would work together (with a bit of persuasion) to provide a computer. Or you could design your own cards. Thus you had CPU cards, memory cards, etc . and you could, within the limitations of the technology of the time, tailor a computer to your own requirements.

I built such a computer in the late 70's and Simon Biggs * , a painter, became interested in at the development of computers for art. So we collaborated in this project; I developed the hardware and the operating system software, while Simon developed the graphic routines and his application programs. Then most computers were powered by the Zilog Z80 chip. Although the Z80 was a powerful chip for its time, it lacked one important instruction for graphics work, namely integer multiply, and it could address only 64k of memory. A new chip became available from Intel, the 16-bit 8088, which did have an integer multiply instruction, and could address one megabyte of memory. My task was to marry the 8088 to the S100 bus Z80 system.

The solution was to develop a coprocessor card, and associated software, featuring the 8088 chip, for the S100 bus. This I did and GO-88 was the result. Having gone so far I decided to increase its capacity so that it also had a stand-alone capability; thus GO-88 became a dual mode board. The new computer was much more powerful, and Simon was able to demonstrate advanced colour graphics at Adelaide Festivals of Art.

GO-88 extended the capacity of S100 bus computers. At one stage I was able to run three different operating systems: CPM, CPM86 and MS DOS. However in the early 80's the IBM PC became available. My GO-88 board was nearly twice as fast as the IBM PC, it was totally asynchronous, that is, an 8 Mhz 8088 could work with a 6 Mhz Z80 and do over 90,000 changeovers per second between the two processors; what really killed it were the IBM PC clones which were cheap and had many of the advantages of the S100 bus; for example, you could buy various plug-in cards and so tailor your computer to your own requirements. The disadvantage was that you were stuck with Intel, and increasingly, Microsoft technology, but that is the way markets work.

Read the GO-88 user manual.

*Was (2004) Research Professor, Sheffield Hallam University, UK and Senior Research Fellow, Computer Laboratory, Cambridge University, UK; now (2008) Research Professor in Art, Edinburgh College of Art, UK.