There are 4 parts to
this review, please use the pulldown below to browse sections or simply
hit "Next" at the bottom of each section to continue
Synthetic and Software/Production Benchmarking
You definitely know that through SEO, your business will benefit hugely, so you have to make sure to find the right Link Building Company that can help you gain your goals SEO on lookup motor rankings, finding the best Link Building Company for your business is a lot more crucial than ever. Link building Company will help you | Simple Website Templates. Every great website and web design needs a starting place or point. Our own newly redesigned website started from a basic template that was heavily modified to suit our needs and the same can be done for you. Using our simple sea website templates on our site you can buy website templates With the system up and running we head into our benchmarking. As we stated
earlier we have decided to do something a little different with our evaluation.
In this round of benchmarking we have pitted the 2400+ against the Barton
at its default settings as well as running the Barton with the same settings
as the AthlonXP2400+. By doing so we can get an idea of how the Barton's
newly added 512k L2 cache helps to improve performance over the Thoroughbred's
256K L2 cache. To test our processors performance we have chosen the following
benchmarks. As you can see we are not just going for the gaming angle
but have included several Production and Real World Benchmarks to help
determine how the Barton will perform in a workstation environment.
Software and Production
· SiSoft Sandra CPU
· SiSoft Sandra Multi-Media
· SiSoft Sandra Memory
· Business Winstone 2001 Ver 1.01
· Cinebench
· Clibench
· Science Mark 1.0
· SpecView
Gaming
· MadOnion's 3DMark 2001se
· AquaMark
· DroneZ
· UT2003 CPU Bench
SiSoft
Sandra is a standard for many end users/builders to compare their
systems to others. These benchmarks are quoted on every website you find
and are almost the cornerstone of bragging rights. Looking at this first
round of results you can see that the Barton running at 133 FSB edges
out the AthlonXP 2400+ in all three Sandra results. The larger L2 Cache
is showing its value in all three benchmarks. Something else to remember
is that there is only 77 Mhz separating these processors at default settings.
The Barton is really showing some muscle as it completely crushes the
AthlonXP 2400+ when run at Default settings.
Business
Winstone 2001 ver 1.01 is another system-level, application-based
benchmark that measures a PC's overall performance. Business Winstone
runs real applications through a series of scripted activities and uses
the time a PC takes to complete those activities to produce its performance
scores.

The Barton again shows its worth in this everyday business application
benchmark. The value of the 512k L2 cache allows the Barton running at
133 FSB to out score the AthlonXP 2400+ by 3.1 points. In its default
setup the Barton out scores the AthlonXP2400+ by 7.1 points. The Barton
certainly makes itself at home while running the Business Winstone benchmark.
SPECviewperf
6.1.2 parses command lines and data files, sets the rendering
state, and converts data sets to a format that can be traversed usingL rendering calls. It renders the data set for a pre-specified amount
of time or number of frames with animation between frames. Finally, it
outputs the result. SPECviewperf reports performance in frames per second.
In this benchmark, higher numbers are better.

The results of this benchmark are close across the board. Highly dependent
on a video cards ability to render the images you can see the effect the
Barton has in each of these benchmarks. Again both of these processors
are separated by only 77Mhz while running their default configurations.
CLI
Bench MK III SMP is as stated on the web page, a different benchmarking
solution. This benchmark runs a gamut of synthetic benchmarks that includes:
Dhrystones, Whetstones, Eight queens' problem, Matrix operations, Number
Crunching, Floating point, memory throughput, hard disk's throughput and
the CPU usage.
Well look at that both the Barton and AthlonXP 2400+ processors score
identical in all six tests while running at 133 Mhz FSB. Obviously the
512k L2 Cache isn't really helping the Barton out here. Now look at the
difference the 166 MHz FSB makes in the test results. Again the Barton
is giving us a look at its real power.
Cinebench 2000 is another benchmarking program that uses ray tracing
to test the CPU for performance. This benchmark also supports SMP as evident
in the multiple CPU ray tracing result.

Much like the Specviewperf benchmark the Cinebench benchmark is dependent
on the Video Cards ability to render the information being fed to it.
This benchmark does give us a glimpse of the advantages the Barton has
over the AthlonXP 2400+ in either configuration. Again the L2 cache is
helping things along.
Science
Mark 1.0 will measure the performance of your computer by the
execution of a few tests, including the simulation of liquid argon, energy
calculation of H2O, and the like. When comparing the three tests, smaller
numbers are better.

Science Mark is a pure processor benchmark, as the video card has no effect
on the DOS window these benchmarks run in. Each of these is a complete
calculation handled by the processor. Again the Barton does well against
the AthlonXP 2400+. Oh did you notice the last result seems backwards?
Well it's not. In this last comparison bigger is better. The score is
a result of a comparison to a base
system, which has a final score of 100.
So there you go a full round of Business and Software benchmarks showing
the Barton is packing quite a punch with its 166 Mhz FSB and 512k L2 cache.
The larger L2 cache is definitely helping the Barton push the numbers
in these benchmarks to new heights in Business applications. Now with
that out of the way it's time to look at the Barton in a different environment.
You know what they say "All work and NO play
" So let's
take the Barton for a run through a couple benchmarks and see how it handles
some gaming.
© 2003 Michael (Oreo) Falcon
|