🚀 go-pugleaf

Hey bub, where do you get off tryng to inject common sense into a perfectly
good flamebait thread?

;>)

"Tony Hill" <hilla@uoguelph.ca> wrote in message
news:i6979uc0qf7egp4piqctpr72om8fop8p3i@4ax.com...
> On Fri, 15 Mar 2002 17:32:11 +0200, "Ari Oppenheimer"
> <ak79@iafrica.com> wrote:
> >I've noticed that the new AMD Proccessors are call XP 1600, XP 1700, XP
1800
> >etc, but the actual speed of the XP 1600 for example is only 1.4 GHz.
> >
> >Now does this compare to a P4 1.6Ghz or a P4 1.4Ghz.
>
> It's more or less comperable to a P4 at 1.8GHz in most applications.
>
> > Also, consider the fact
> >the the P4 runs at a 400Mhz bus. To me it seems as if Intel comes out as
the
> >sure winner over here. After all it seems as if P4 1.6Ghz is better than
AMD
> >1.4Ghz,.
>
> The P4 also has only an 8K L1 data cache and ~21K L1 instruction cache
> vs. 64K of both for the AMD chips, and the P4 has a very long pipeline
> which takes longer to recover from branch misprediction, and the P4
> has fewer execution units for both integer and floating point, etc.
> etc.  Long story short, there is a LOT to consider here.
>
> What AMD did with their new model number system is to exchange one
> meaningless measure of performance (clock speed) with another
> meaningless measure of performance (model number).  If you want to
> know how fast one chip will be as compared to another, you're best bet
> is to check for benchmarks that closely relate to what you use your
> computer for most.  Here's some links to a few recent benchmarks:
>
> http://www.anandtech.com/showdoc.html?i=1543
> http://www.aceshardware.com/read.jsp?id=45000277
>
>
> >Please let me know if my deduction is right, or should I go for the AMD
>
> As you can probably see from the above benchmarks, there is no
> clear-cut winner for every single application out there, in some cases
> even the slowest AthlonXPs beat out the very fastest P4s, while in
> other applications the opposite is true.
>
> It should also be pointed out that the processor is only one part of
> the system, and often a relatively small part when it comes to actual
> performance.  If you look at this article:
>
> http://www.aceshardware.com/read.jsp?id=45000212
>
> You can see that the type of memory used and the motherboard chipset
> can have a very serious impact on performance (though the above is
> probably the most extreme case of this).  Video cards can also have a
> tremendous impact on performance, particularly for games and for
> workstation-type 3D graphics.  Hard drive performance is also
> extremely important in determining how fast your system will be, and
> even things like network cards and sound cards have an impact on the
> overall system speed.
>