Path: utzoo!utgpu!news-server.csri.toronto.edu!clyde.concordia.ca!uunet!samsung!zaphod.mps.ohio-state.edu!mips!pacbell.com!pacbell!att!dptg!ulysses!andante!alice!pereira
From: pereira@alice.UUCP (Fernando Pereira)
Newsgroups: comp.lang.prolog
Subject: Re: Prolog Efficiency and Benchmarking
Message-ID: <11122@alice.UUCP>
Date: 2 Aug 90 18:22:55 GMT
References: <708@ecrc.de>
Reply-To: pereira@alice.UUCP ()
Organization: AT&T, Bell Labs
Lines: 22

In article <708@ecrc.de> micha@acrab4.UUCP (Micha Meier) writes:
> The Pereira benchmarks used
>for the same purpose are not well suited because they  often  measure
>something different than they say.

I've just looked at the comments in the source for my benchmarks, and nowhere
is there a statement of what the various programs are intended to measure,
so I don't see what evidence you have for your statement. Now, clearly I
had in mind when preparing the benchmarks that they would exercise different
aspects of a Prolog implementation, but I was familiar with enough different
implementations to know that I didn't have the time to build a benchmark
suite whose components I would be sure would measure the same individual
aspects across implementations.

Otherwise, I entirely agree with and support your points, and I will be
digging into my archives to find programs that might be useful for
benchmarking.

Fernando Pereira
2D-447, AT&T Bell Laboratories
600 Mountain Ave, Murray Hill, NJ 07974
pereira@research.att.com