RE: [BCB] FORTRAN vs. C++

[Kregg Kemper <kkemper@nospam.anelva.com>]

  I hear continuously c is faster and more efficient than C++ but have
never seen a test or heard of any test done to prove that. Where are the
graphs and percenteges that to back up these claims? 

  Last year I did my own test to see how much slower C++ was by writing
a sorting program in c style code and compiled as a bcb console app.
After recording different sort times I rewrote the program as a class
with inheritance, virtual functions, and the like. I found that it ran
15% faster than the c style code and more than 30% faster after turning
on the compiler optimizations. I know that was a lame test but know I
don't hesitate to ask someone to prove it when they say c++ is slow. 

   so show me some graphs comparing c, c++, fortran99, and delphii and
then lets talk.
 
-----Original Message-----
From: Robert Manning [mailto:Robert.Manning@nospam.trw.com]
Sent: Thursday, September 14, 2000 12:19 PM
To: bcb-list@nospam.orionlink.net
Subject: Re: [BCB] FORTRAN vs. C++


In my personal opinion...

The letter is 5-1/2 years out of date, so I'm not sure what good is to
be gained from debating such an outdated, obviously biased document.
C++ is no longer taking over the world... Java is, and for reasons
different than cited in this letter. At any rate, let's not overlook
the obvious:

- Dr. Prentice presents a biased paper in favor of a language his
company apparently uses to make money. Thus, he has a financial
motivation to justify his preference.

- Comparison tests are cited in the document, but no test data is
given, nor the source of any test programs. I would be highly suspect
of claims that Fortran 90 is superior to C++, without any proof to
back them up.

- His letter includes advertisement for his companys services. To me,
this entire letter looks more like advertising than anything else.

- The doctor's opinions regarding the shortcomings of C++ are just
that, opinions. Personally, I would prefer not to abuse myself with
Fortran, and would take C++ over it any day of the week. In fact, I'd
rather be forced to program in BASIC than Fortran.

In short, I would challenge every assertion made in this article, for
one reason or another (I don't have the time to create a detailed list).
If I were you Leslaw, I'd take this article with a grain of salt, and
remember that it's an ancient document (that someone probably forgot to
remove from their website long ago). This letter does not change my
opinion regarding C++; rather, it reminds me that some people who call
themselves programmers would prefer to learn only one language, and use
that language for everything (which often comes from people not trained
as computer scientists). Different tasks require different tools; as I'm
sure I've said before, using the right tool for the job is the proper
thing to do.

-Robert

Leslaw Bieniasz wrote:
> 
> I don't want to start holy wars, but has anybody read a paper
> at wwww.lahey.com/PRENTICE.HTM ?
> I'd be glad to know your comments.
> 
> L.B.