NetNews Offline 2/NetNews Offline Volume 2.iso/news/comp/lang/c++-part2/10601

Internet Message Format | 1996-08-05 | 1.8 KB

Path: chronicle.mti.sgi.com!news
From: [e-mail] (Matt Austern)
Newsgroups: comp.lang.c++
Subject: Re: Will JAVA kill C++?
Date: 08 Mar 1996 20:14:19 GMT
Organization: SGI
Message-ID: <[e-mail]>
References: <[e-mail]> <4hp18v$[e-mail]>
    <4hq2j6$[e-mail]>
Reply-To: [e-mail]
NNTP-Posting-Host: isolde.mti.sgi.com
In-reply-to: [e-mail]'s message of 8 Mar 1996 19:42:30 GMT

In article <4hq2j6$[e-mail]> [e-mail] (Tom Payne) writes:

> : I still think the C++ program would be quicker and more efficient -- There
> : are still some things you will not be able to do in JAVA, such as
> : pointers.  And anyone who believes arrays can do anything a real pointer
> : can does not know what they are talking about and probably does not know
> : how to program very well.
>
> When one is interested in speed, one should avoid pointers (as
> does. Fortran), since pointers kill opportunities for optimization.

The problem with Java isn't that it has no pointers, but, rather, that
it makes it impossible to avoid pointers.  All references to primitive
types in Java are by value (i.e., not using pointers), and all
references to class types are by reference (i.e., using pointers).

In C++, or Fortran, or Eiffel, you can have an array of complex
numbers.  In Java, though, you can't: you can have something that
looks like an array of complex numbers, but internally it's really an
array of pointers to complex numbers.  You have to pay a space penalty
to store those pointers, and a time penalty to do the dereferencing on
every element, and the optimizer will have to generate code on
assumption that two array elements might point to the same object.
--
Matt Austern
SGI: MTI Compilers Group
[e-mail]