Re: History and evolution of compilers

On 8 Oct 1997 00:43:21 -0400, Peter Ludemann <ludemann@inxight.com>
wrote:
>The moderator wrote:
>
>> Fortran IV did get two compilers on the 360 series, Fortran G which was
>> fast and generated rotten code, and Fortran H which produced very good
>> code. -John]
>
>As I recall, the Fortran G code wasn't "rotten"; in most cases it was
>acceptable (maybe you're thinking about the WATFOR/WATFIV compilers).
>Wirth's Algol-W compiler produced code of about Fortran G's quality;
>and it ran as fast as the WATFIV compiler.


Ah, yes. WATFIV. Don Cowan and company. (Some of the authors were
my CS instructors; mostly retired now.)


WATFIV intentionally didn't optimize hardly at all; the point of it
was to provide a fast compiler with a goodly amount of compile error
diagnostics targeted for use by students that would compile programs
more often than they ran them, and then would probably use test sets
of such trivial size that the program would seldom run more than a few
seconds (in elapsed time) anyways.


At least half of the point must have been to provide a compiler that
actually provided "useful" error messages to students that wouldn't
necessarily have copies of IBM's error code manuals handy.


I think that I may have recompiled using FORTG or FORTH once or twice;
it was hardly a worthwhile exercise wading through the differing
options and diagnostics from my perspective as a student at the time.
--
cbbrowne@hex.net, <http://www.hex.net/~cbbrowne>
--

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