Date: Thu, 9 Nov 2000 08:01:05 -0500
From: "Lawrence Young" <[EMAIL PROTECTED]>
Subject: Re: New Sony mininotebook
----- Original Message -----
From: "Daniel Smith" <[EMAIL PROTECTED]>
To: "Libretto" <[EMAIL PROTECTED]>
Sent: Thursday, November 09, 2000 4:03 AM
Subject: Re: New Sony mininotebook
> Date: Thu, 9 Nov 2000 20:06:33 +1100
> From: Daniel Smith <[EMAIL PROTECTED]>
> Subject: Re: New Sony mininotebook
>
> On Tue, Nov 07, 2000 at 11:38:35PM -0800, David Chien wrote:
> >
> > However, modern benchmarks have improved in various areas and some
simulate
> > the general day-long operations carried out by a typical user using
typical
> > office applications. If performance is poor even then, the results
suggest a
> > poor performance in other applications.
>
> Wrong, wrong wrong. How many times do you use comdlg32.dll? gdi.exe
user.exe
> After a number of passes the Crusoe architecture will run these faster
> as it profiles the code and optimises it.
These are UI codes and UI codes spend more time waiting for user input than
running. Benchmarking the UI codes are waste of time.
>
>
> > In any case, I'm certainly not typing in the same thing over and over
again
> > (nor clicking the same thing, nor running the same thing in the same
order on
> > my PC). I don't believe many others are doing the same, unless their
work is
> > extremely dull and repetitive ;)
>
> Well actually you are. The same piece of code gets run each time you
> pull down a menu/access the hard drive. These benchmarks are
> fundamentally broken in they assume the result is identical between
> runs.
Again, no benchmarks are intend to test the speed of UI codes. I will give
you the credit for the last statement however.
Lawrence
**************************************************************
http://libretto.basiclink.com - Libretto mailing list
http://libretto.basiclink.com/archive - Archives
http://www.picante.com/~gtaylor/portable/faq.html - FAQ
-------UNSUBSCRIBE-------
mailto:[EMAIL PROTECTED]?subject=cmd:unsubscribe
--------UNSUBSCRIBE DIGEST------
Use above but add DIGEST to the subject line...
**************************************************************
