[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Sheflug] mindcraft testing of linux!
>>>>> "Will" == Will Newton <will [at] misconception.org.uk> writes:
Will> From what I heard Mindcraft contacted RedHat once and RH
Will> were not too helpful so they just said "Oh well" and used a
Will> Microsoft labs tuned machine against an almost stock RH and
Will> Apache setup. Not a very good test TBH.
Actually, the test was also tuned to NT's strengths, and a careful
inspection of the Linux configuration showed several adjustments that
actually made that system slower. (Not necessarily a conspiracy; some
of the knobs had misleading names that made them do more or less the
opposite of what you would expect, and other adjustments would have
made sense under NT but were actually harmful under Linux.)
Will> A lot of the bottlenecks involved are fixed in 2.3/4 and
Will> some are backported to recent 2.2 also.
Unfortunately, Linux VM is getting slower. This probably doesn't
matter for head to head comparisons against NT (tests will be oriented
to large memory machines, and Linux still uses less than NT), but it
does make it more difficult to squeeze decent performance out of a
machine without a huge memory.
Will's basic point, however, is correct. Lab benchmarks are like IQ
tests: "performance is what our test measures." Unless you get the
benchmark source, and understand it, you cannot say what the results
mean. Not to mention the inherent bias in having the testees hire the
testers.
If you want anecdotes, how about "NT couldn't host Hotmail"
vs. "Source Forge is all of open source hosted on VA Linux." That
blows away the Mindcraft report IMHO.
--
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Institute of Policy and Planning Sciences Tel/fax: +81 (298) 53-5091
_________________ _________________ _________________ _________________
What are those straight lines for? "XEmacs rules."
---------------------------------------------------------------------
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.