Re: [TCLUG:5285] NTS4 vs. Linux study from Mindcraft.

[pani@frontiernet.net]

Maybe the benchmark would make more sense if it was run on 
machines that an average user would buy for that kind of an
application.  How many users really have these Xeon boxes being
used for web/file serving or planning to invest in one for the
same purpose anyway ?  

If linux beats NT on cheaper hardware, which it most certainly
does, which more people can afford to buy then who cares that
NT runs faster if you add 4Gigs of RAM and 10 more processors ?

Aah well, I am just pissed off that I do not have one of those
xeons, that is all *greeen*
--
sandipan


Scott Dier wrote:
> 
> Heh, what about us pre-college but-going-to-be-CS-majors-next-fall? :)
> 
> Anyways, I have peeked through the smart reseller stuff.  Pretty neato
> benchmarks, but nothing like the data from the mindcraft survey.  Really,
> it would be cool to set it up, hammer some machines and have a managed
> port that we could hook MRTG to or something and make some graphs.
> Problem is, where do we get these sweet-ass boxes? :)  I only know one
> person with a dual-celeron 504.  And thats not even a xenon.  Oh well.
> Even if we had a ppro system we could tell what it was like comparitevly.
> (linux vs nt vs netware etc.)  I have access to NT Enterprise through a
> MSDN Universal subscription. (yay!) Too bad we couldn't convince my school
> district to let us conduct an experiment at school, Maple Grove Sr. High
> ;( (500+ pentium win95 boxen.)   Anyways.  That sparked more of a
> discussion than I expected.  And it was on-topic.  Sweet. :)  (just
> digging the signal-to-noise ratio here.)
> 
> Perhaps at the next meeting there should be a side-discussion
> (after-major-stuff, i guess i dont know the format of the meetings (yet))
> of interested parties to see if there is a real interest and avalibility
> of hardware to do this.  Write it up as a report, and publish it to the
> world at large.
> 
> I guess this also means that I am going to start going to tclug meetings.
> :)