Re: [Sheflug] mindcraft testing of linux!

[Will Newton <nospam [at] misconception.org.uk>]

On Sun, 6 Aug 2000, ross wrote:

> you may not thank me for raising this subject..... i'm sure you discussed
> it shortly after the event.

I try to stay away from advocacy, but it's a slow Sunday evening...
 
> part of my current studies have required me to look at the comparison of linux
> (and open source software in general) and commercial software (ie microsoft)
> with respect to performance and scalability.

Difficult to do without doing it yourself. Almost all major tests are done
with the aim of selling magazines, getting hits, promoting software,
getting endorsements etc. As the old saying goes, if you want owt doing
right, do it yourself.

> i read the mind craft report on the comparative tests and have also read the
> halloween papers. i must say that i was extremly dissapointed with the
> performance tests of linux. i had been led to believe by all the hype
> surrounding linux that it was both stable (not in dispute) and fast (compaired
> with NT). 

Mindcraft was a long time ago. Almost all the problems that caused the
poor results have been fixed. Try the recent SPEC Web results of the Tux
web server if you don't believe it. (Hint: It's Linux based and blows away
any previously tested web server)

> i reallise that some of the deficiencies in the 2.2.x kernel are being
> addressed with the 2.3.x / 2.4.x kernel. however, i would be interested to know
> if there are other (reliable) comparison tests that have been correctly
> administated which show the opposite results. if this is the case, is there a
> simple explanation for this. from what i have read of the mindcraft reports,
> redhat, and even linus, had some involvement in the testing procedure so i
> cannot suspect the validity of these tests.

>From what I heard Mindcraft contacted RedHat once and RH were not too
helpful so they just said "Oh well" and used a Microsoft labs tuned
machine against an almost stock RH and Apache setup. Not a very good test
TBH.

A lot of the bottlenecks involved are fixed in 2.3/4 and some are
backported to recent 2.2 also.

Linus stated as a goal for 2.4 "Mindcraft can't happen again".

Try the recent SPEC Web results. I don't have a link but it shouldn't be
too hard to find.


---------------------------------------------------------------------
Sheffield Linux User's Group - http://www.sheflug.co.uk
To unsubscribe from this list send mail to
- <sheflug-request [at] vuw.ac.nz> - with the word 
 "unsubscribe" in the body of the message. 

  GNU the choice of a complete generation.