|
Advice Beginners BIOS Guide CPUs Links Mainboards Memory Network Storage Video/Sound Cards Contact Forum SiteMap Sponsors WebNews Home 
|
. | . |
Prices: Mainboards ABIT ASUS Chaintech Shuttle Soyo Tyan CPU Intel P4 2.4C-800 P4 2.6C-800 P4 2.8C-800 P4 3.0-800 P4 3.2-800 AMD AthlonXP XP 1700+ XP 2000+ XP 2400+ XP 2500+ XP 2700+ XP 3000+ XP 3200+ Athlon64 Athlon64 3200+ Athlon64 FX-51 Opteron Opteron 240 Opteron 242 Opteron 244 Opteron 246 Memory Corsair Crucial Kingston Mushkin OCZ |
SPEC ViewPerf
ViewPerf 7.1 is the latest version out there but it is only marginally different from the 7.0 version which is the one we are still using for the time being since the update mostly addresses some compatibility issues. We have been running ViewPerf before mostly on the nVidia Quadro4 XGL980 graphics card and found it extremely sensitive to differences in the AGP protocol as well as memory latencies. Some of the individual benches are mostly limited by the graphics adapter, this has not changed but others are definitely showing the power of the CPU. All numbers are given as fps.
3DSMax
The difference is not huge but the P4 pulls ahead.
DRV-08
DRV-08 shows everybody performing the same way with the A64 FX taking a nominal lead. Note that these numbers are way above those you usually see.
DX-07
DX-07 grants the P4 the lead, however, it still appears more an issue of the CPU-bus interface than raw CPU power since there is no difference whatsoever between the 3.0 and the 3.2 GHz version. Alternatively, we might see the limitation of the graphics card itself.
Light-05
Light-04 has always favored the Intel CPUs (with a few short intermissions in between)
ProE-01
The Athlon 64 comes in between the P4 3.0 and the 3.2 version.
UGS-01
Another "nominal" random placement
In summary, SPEC ViewPerf favors the P4, that is, every benchmark that is not graphics limited is won by the P4. On the bright side, the A 64 FX clearly pulls ahead of the XP3200+ and overall, aside from Light-05, the differences are hardly anything to write home about. It is also possible that with revised chipset drivers or else with a different chipset altogether, that is, a wider and faster HyperTransport interface than the 24-bit bus we are stuck with here, things might take a very different turn.
next page: => Cinebench =>
