|
More Performance
ScienceMark
2.0 ~ Memory
Although
as expected we see L2 cache performance on a par with
the similarly clocked 2800+ I was rather surprised
to see L1 cache performance improve by so much.
ScienceMark
Cache Performance Comparison
ScienceMark
2.0 ~ Primordia Benchmark
If
we throw up this result against the 2400+ which we
should remember runs at a healthy 2.0GHz, a mere 160MHz
behind the 3000+ we get an idea of the performance
that's achieved by combining a 333MHz FSB with a 512k
L2 cache. The 2400+ takes around 16.5% longer to complete
the simulation.
Primordia
Simulation Times, Lower Is Better
HotCPU
Tester Pro 3
HotCPU
Pro is a great little piece of software written by
Damon Chitsaz of Opusware who have now joined forces
with 7Byte
Computers. In addition to a diagnostic routine
which pumps CPU usage to 100% for a definable number
of hours it also features a quick and effective CPU/memory
benchmark.
Here
we see a perfect example of frequency dependant results.
Although we've consistently seen the 3000+ outperform
the 2800+ in a huge variety of applications in testing
here, taking this single benchmark in isolation would
lead you to conclude (wrongly in most cases) that
Barton is the slower CPU.
Barton
3000+
Thoroughbred
2800+
It
doesn't really matter how fast your processor is if
it isn't stable so I fired up the diagnostic routine
in HotCPU and let it ramp up CPU utilization to 100%.
Stability Test Running
Using
a 6 hour burn in over night last night I'm pleased
to say that both our test system and the 3000+ emerged
victorious!
|
