Error - template LAYOUT-DATA-WRAPPER not found

A configuration error was detected in the CGI script; the LAYOUT-DATA-WRAPPER template could not be found.

Error - template STYLE-SHEET not found

A configuration error was detected in the CGI script; the STYLE-SHEET template could not be found.

Error - template SUB-TOP-BANNER not found

A configuration error was detected in the CGI script; the SUB-TOP-BANNER template could not be found.
Subject:
From:
David Gillett <[log in to unmask]>
Reply To:
PCSOFT - Personal Computer software discussion list <[log in to unmask]>
Date:
Thu, 29 Apr 1999 11:22:42 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (49 lines)
On 28 Apr 99, at 18:58, M. Jelley wrote:

> I just read the following and would like to hear the responses of the list
> members.  As I wrote earlier this week,, I am having a P III built.  Being
> naive as I am I thought this was going to a breeze . . .  Software in, speed
> out !  What are your thoughts?
>
> > Before applications programs can take advantage of the P III instruction
> > set they'll have to be recompiled in a compiler whose code generator is P
> > III aware. That may take awhile, but eventually it will happen. Depending
> > on the performance gains -- they may be detectable only with benchmarks --
> > you can then upgrade your software to match your hardware.
>
> Thanks in advance -
> Marianne Jelley

  We could conveniently divide the "instruction set" provided by the PIII
into two groups:  instructions that were already provided by the PII, and
instructions that are new with the PIII.  Any software you have that runs on
a PII is likely to consist only of instructions from the first group.

Q. Does the PIII execute PII instructions faster than a PII at the same clock
rate would?

A. Not according to most of the benchmarks I've seen.  There may be a few
instructions which run faster, but these don't occur all that often in most
applications.


  There are various things that an application might need to do that required
several PII instructions, but which can be replaced by a single (faster) PIII-
specific instruction.  This replacement is only going to happen where code
has been generated to take advantage of PIII instructions.
  The good news here is that these sorts of operations are often done inside
*drivers* rather than in applications.  So there is a reasonable expectation
that, over time, specific operations (such as display performance) in common
applications may be able to be improved without requiring new application
versions.

  The message was, however, substantially correct; you will not *initially*
see much performance difference between PII and PIII in *most* applications;
differences will begin to emerge over time.

David G

                PCSOFT mailing list is brought to you by:
                            The NOSPIN Group
                            http://nospin.com

ATOM RSS1 RSS2

LISTSERV.ICORS.ORG Secured by F-Secure Anti-Virus CataList Email List Search Powered by LISTSERV