Path: utzoo!utgpu!water!watmath!clyde!att!pacbell!lll-tis!mordor!joyce!sri-unix!quintus!ok
From: ok@quintus
Newsgroups: comp.lang.prolog
Subject: Pascal/C -vs- Prolog
Message-ID: <184@quintus.UUCP>
Date: 21 Jul 88 21:29:31 GMT
Sender: news@quintus.UUCP
Reply-To: ok@quintus ()
Organization: Quintus Computer Systems, Inc.
Lines: 19

Micha Meier says that they have a compiler which is "10 times" faster
than the Quintus compiler, and suggests that this may be due to it being
coded in C.

According my measurements, when you say compile(somefile) in Quintus Prolog,
over half the time goes in I/O.  This has little to do with Prolog as such,
and more to do with Quintus Prolog's user-definable streams.  [Our data-
base interface actually exploits this to compile an encrypted file directly.]

As for the rest, remember that C is essentially the same as BCPL, which is
20 years old.  It would be a strange thing if we didn't know how to use it
by now!  The DEC-10 Prolog family is about 10 years old and we still have a
lot of learning to do.  I have been able to speed several components of the
Quintus Prolog system up (figures proprietary, but I'm pleased with them)
by rewriting them in **cleaner** Prolog.  [No, I do not know when the new
system will be ready for release.]

A fast compiler is well worth having, but I think it is better for Prolog
users in the long run if vendors try to make faster compilers *in Prolog*.