[rescue] SMP on intel wasteful?

[Dave McGuire]

On June 25, Andrew Weiss wrote:
> > Let me ask you again (since you're finally starting to say a little more
> > than stuff like "x86 is repulsive):  For real world work, what can you
> > show me in the RISC world that compares to my $2,000 x86 box in terms of
> > productivity?
>
> A Powermac G4.  Dollar for dollar you get better performance and value 
> out of the latest G4 Quicksilver line.  Especially since out of the box 
> the thing runs a Unix based OS, and you can always run Yellowdog or 
> NetBSD if you choose.  You can also build Beowulf clusters of these in 
> Marathon racks.  I think last time I did the comparison I started with 
> what machine you get from Dell vs. Apple for exactly $1700.  The 
> machines were identical almost... even in CPU speed... except the PPC 
> had GigEthernet and one other really nice feature that I'm forgetting at 
> the moment.

  Yup, I'd have to agree with Andrew...G4.  Further I can point out
Sun's Netra X1.  As I understand it, it's been replaced by a more
recent product with which I'm unfamiliar, but at $1K list price, all
in all it's gotta be the best bang for the buck I've seen in the
computer world in a long time.  Also, you might want to ask Sridhar
(if he ever reads email again) about the upcoming POWER4 desktops that
IBM is working on.  I seem to remember pricing being targeted
somewhere in that range.

  These, of course, won't be "high productivity machines" right away if
you're working in a company full of "duhhh Windows!"  idiots, but
that's your problem.  Myself I tend to be a little more discerning
over the people with whom I associate.

  By the way...the list archives reflect the fact that I said a lot more
than "x86 is repulsive" since the very beginning of this discussion,
backing up my disparaging anti-PeeCee remarks with real information.
Stop insisting that I didn't back up my statements.  Your lack of
understanding the information I presented does not make it incorrect.
Everyone else here got it...if you didn't, perhaps you're out of your
league here.

  Also along those lines, I've given up on addressing your CISC/RISC
assumptions because it's clear that you just don't understand.  I'm
sorry, but that's just going to have to be your problem, as I have
work to do.  In the interest of helping my fellow man, I will refer
you to "High Performance Computing" by Dowd and Severance for more
information.  It offers excellent, mostly non-theoretical discussion
on the topic.

        -Dave

-- 
Dave McGuire                  "Needing a calculator indicates that
St. Petersburg, FL              your .emacs file is incomplete." -Joshua Boyd