Re: Turbo Pascal 1.0 trivia

David L Moore <>
30 Nov 1997 22:54:40 -0500

          From comp.compilers

Related articles
[5 earlier articles]
Re: Turbo Pascal 1.0 trivia (Stephen S. Mitchell) (1997-11-29)
Re: Turbo Pascal 1.0 trivia (laurie boshell) (1997-11-29)
Re: Turbo Pascal 1.0 trivia (1997-11-29)
Re: Turbo Pascal 1.0 trivia (Toon Moene) (1997-11-29)
Re: Turbo Pascal 1.0 trivia (Jerry L. Wahl) (1997-11-29)
Re: Turbo Pascal 1.0 trivia (1997-11-30)
Re: Turbo Pascal 1.0 trivia (David L Moore) (1997-11-30)
Re: Turbo Pascal 1.0 trivia (Martin Eskildsen) (1997-12-05)
| List of all articles for this month |

From: David L Moore <>
Newsgroups: comp.lang.pascal.borland,comp.compilers,comp.lang.misc,comp.os.cpm
Date: 30 Nov 1997 22:54:40 -0500
Organization: Netcom
References: 97-11-139 97-11-149 97-11-173
Keywords: Pascal, history

Toon Moene wrote:

> The first "native code" I saw produced by a Pascal compiler was COMPASS, i.e.
> CDC 6600/7600/175 series code.
> Not surprising, as Niklaus Wirth, and his female companion that I never can
> recall the name of, *wrote* Pascal for the CDC systems.

The CDC Pascal compiler was written by Urs Ammann. I have probably got
the number of m's or n's wrong. There was an earlier version of Pascal
which was ported to the ICL 1900 by a group in England whose names I
ought to be able to remember but cannot.

I am not sure at this point where that compiler came from - whether it
was a port or an ab initio effort. I think it was written up in
Software Practice and Experience, somewhere in the mid seventies.

Turbo Pascal did generate machine code, but a lot of it was calls to
an approximately 12K kernel which was presumably entirely hand coded
in assembler, so it was somewhere between a compiler and an
interpreter. When you output a com file, this kernel would be output
as the start of the file.

Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.