home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.lang.c++
- Path: uu4news.netcom.com!amc-gw!curtis
- From: curtis@amc.com (Curtis Green)
- Subject: Re: Will JAVA kill C++?
- Message-ID: <1996Mar19.170927.10340@amc.com>
- Organization: Applied Microsystems Corporation
- X-Newsreader: TIN [version 1.1 PL6]
- References: <4ie2m7$5re@hn.ocbbs.gen.nz>
- Date: Tue, 19 Mar 1996 17:09:27 GMT
-
- Our C++ users group had a project manager from the Mocro$oft C++ optimization
- group and he said it was very difficult for an optimizer to follow pointers
- and structures, hence C++ optimization technology has a long way to go (classes)
- He said that most industry benchmarks for C++ compilers just run a suite of
- C tests, so you aren't seeing C++ optimization stats.
-
- Steve Detoni (steve@hn.ocbbs.gen.nz) wrote:
- :
- : : When one is interested in speed, one should avoid pointers (as
- : : does. Fortran), since pointers kill opportunities for optimization.
- : How can you say that .... Would you rather pass a large structures entire
- : contents, or say just a 4 byte address pointer, now which would be faster?
- : Or would you be so un-modula as to have such things as global variables.
- :
- : Optimisation is the art of using the language capabilities to their best,
- : and C(++) has alot to offer, and pointers are a major one, especially in
- : memory optimisation ....
-
- --
- Be seeing you...
-
- Curtis Green | Software Engineer
- curtis@amc.com | Applied Microsystems Embedded Systems
- | http://www.amc.com
- My opinions are |
- expressly mine | This year Khan escapes Earth on the Botany Bay.
- on my own. | (Cliff Clavin, 1996)
-