Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!ucsd!ucbvax!hplabs!hpda!hpcupt1!jamiller
From: jamiller@hpcupt1.HP.COM (Jim Miller)
Newsgroups: comp.lang.c
Subject: Re: Ada speed vs. C...
Message-ID: <5940021@hpcupt1.HP.COM>
Date: 15 Mar 90 20:25:24 GMT
References: <8347@hubcap.clemson.edu>
Organization: Hewlett Packard, Cupertino
Lines: 23

>/ hpcupt1:comp.lang.c / dd@sei.cmu.edu (Dennis Doubleday) /  6:14 am  Mar 14, 1990 /
>hammondr@sunroof.crd.ge.com (Richard A Hammond) writes:
>>2) Ada runtime checks should be disabled for comparison to other languages!
>
>And why not?  If it's pure speed you're after, then you can have it in
>Ada by disabling runtime checks.  Why is this an unfair comparison?  C
>gives you no runtime checks, so why should its execution speed be
>compared to that of Ada with runtime checks enabled?  On the other
>hand, if you're willing to sacrifice a little speed for runtime
>security, then Ada is also a good choice.
>
>-- 
>Dennis Doubleday  (dd@sei.cmu.edu)                     _   /|

Why not? because the pupose of a benchmark should show how the languages
are going to be used.  Not how they could be used.  Mentioned previously
was that the C test called routines instead of having the code inline
-- which is what I do in the *few* spots that speed counts. 

So, either compare normal usage, or speed hack usage, but not
your speed hack to other's normal use.

   jim - it's my opinion so it is TRVTH - miller