Control Program for Microcomputers...

.. alternatively Control Program / Monitor

Let me quote Bernd Pol:

«Even in micro computers, you can find operating systems of all types and sizes, depending on computer type and application. But none of them has achieved such an enormous spreading as CP/M for micro computers, based on the CPU types 8080, 8085 or Z80. Meanwhile, this Digital Research product can rightfully be regarded as a standard.»
("Vom Umgang mit CP/M", 1983)

In the late seventies and early eighties CP/M was the undisputed standard in operating systems for microcomputers. Furthermore, it was the very first DISK Operating System (DOS), and therefore the basis for today's Microsoft DOS/Windows systems. MS-DOS itself derives from QDOS (Quick and dirty operating system), a CP/M clone written by Tim Patterson, which was achieved by Microsoft for the first IBM PC. Many former CP/M applications have been ported to DOS later on, in first place DBase and Wordstar.
CP/M on the other hand is based upon the Operating System for the DEC PDP-10, "TOPS-10", at least concerning file and device naming conventions, as well as for some command designations. Other influences are less clear, but undisputable.
Actually, many things we now take for granted, have their origin in the work of a single man, Gary Kildall, the developer of CP/M. Honestly, did you ever think about why your hard disk is addressed as C: or where the DOS command DIR comes from?!

The CP/M Story

... in memoriam Gary Kildall

Already four years have passed since Gary Kildall, the creator of CP/M, has died suddenly and unexpectedly - being only 52 years old. On this occasion, we should remember how the operating system CP/M came into being:

In 1972 young Gary had achieved his doctor's degree in computer science and was then employed at the military university of Monterey, California. He liked the job, most of all because it offered him quite a lot of freedom. There was only one thing that disturbed him very much: Everyday, he had to cover more than a hundred miles to arrive at his working place. So it's quite easy to understand that Gary envisioned a computer all for himself that he could set up near to his beloved young wife...

In 1973, when Gary attended the presentation of the new Intel 8080 CPU, his dream appeared much nearer to reality. He was so enthusiastic about the 8080, that he offered to the Intel managers to build a compiler for PL/1 (= Programming Language Number 1). During this period, PL/1 was very often used for mainframes and therefore Intel agreed at once; the compiler was then called PL/M (= Programming Language for Microprocessors).

There was only a little problem: Gary didn't own a computer running on an 8080. He only had access to a Digital Equipment PDP-10. Therefore Gary coded his PL/M compiler in FORTRAN on the DEC PDP. When the compiler was ready, Gary still needed an 8080 computer for testing. He even succeeded in persuading the company Shugart to give him a floppy drive as a gift. Since the cables, the power supply and the controller were missing, Gary was unable to use it.

As the donated floppy was worthless for him, Gary decided to simulate the 8080 on the PDP computer. From this the first version of CP/M resulted - so to speak as a "workaround"...

In October 1973, Gary by chance came to know the inspired hobbyist John Torode, for whom it wasn't a problem to connect the antiquated floppy to a 8080 single-board computer.

When they started up CP/M for the first time, they both were speechless: It worked fine nearly at the first go! Because of this Gary was so much convinced of his CP/M that he offered it to Intel together with the PL/M compiler. Intel purchased the compiler immediately, but they weren't convinced of CP/M. They simply couldn't imagine why everybody should have a little computer totally of his own! At that time all big shots of the industry believed that only the giant monsters of time-sharing systems could have a great future...

But Gary, being truly American, wasn't left discouraged, and he decided to set up a trade for CP/M by himself and founded the company Digital Research, also known as DR (DR is no longer in business, as it was acquired by Novell more than two years ago).

By mail circulars Gary made his CP/M popular with practically all of California. Since the cost of version 1.3 (1.4?*) was only 70 $, this operating system soon became installed on every 8080 computer.

In 1979, DR released version 2.0, which soon was followed by CP/M 2.2. This version was installed most worldwide, as it offered high performance together with little memory requirements. Home computers like e.g. the AMSTRAD CPC series still make use of it nowadays.

CP/M version 3.0, also called CP/M Plus, was not released before 1982. Unfortunately this was too late, since one year before the IBM-PC had begun its triumphant advance. Besides the AMSTRAD CPC 6128 and their PCW (both using CP/M 3.1*) you can find CP/M Plus also on the C-128 (CP/M 3.0*). So, folks, next time you're working with CP/M on your PCW, just include a memorial minute for Gary! After all, we owe a fantastic operating system to him!

DangSoft, © 1998

(* Annotations of the translator)

Translation by Gaby Chaudry.
Thanks to Helmut Jungkunz for the check-through.