* Kristian Farren ([EMAIL PROTECTED]) [990425 01:40]:
> Anybody else see this article,
> http://www.microsoft.com/Windows/dailynews/042199.htm
> This seems to refute the bencmarking done by ZDNEt
> a few month ago. Who's right? or does it just show that,
> benchmarks like statistics, can be twisted to your advantage.
Here's a reasonable summary of events, if slanted in Raymond style.
http://linuxtoday.com/stories/5296.html
Beyond that they activated several features of Samba and Apache which
pretty much deliberately slowed down the system, claimed it was a stock
setup and had custom-compiled versions of both servers, and used a
non-current kernel with known performance interoperability problems with
Windows systems, they also have fallen apart on the 'independent testing'
bit.
The message which is understood (although not in a provably legal way) to
have been posted by the mindcraft folks requesting help from a newsgroup
was posted to usenet via dejanews from a host within microsoft's domain.
The machine used was equipped with an unreleased internal version of IE5.
That people involved in this testing had demontstated access to currently
unreleased software microsoft software as well as microsoft's internal
network throws the whole 'independent testing' out the window.
In other words, a deliberately sabotaged linux box compared against an NT
machine tuned by those both in the pay of and with the engineering
assistance of (far beyond what any normal customer could ever hope to
achieve) microsoft can be measured to perform poorly.
I hope the shock does not cause you undue emotional distress.
-josh
--
To get out of this list, please send email to [EMAIL PROTECTED] with
this text in its body: unsubscribe suse-linux-e
Check out the SuSE-FAQ at http://www.suse.com/Support/Doku/FAQ/ and the
archive at http://www.suse.com/Mailinglists/suse-linux-e/index.html
