|
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 |
LOSTCIRCUITS |
|
| Shootout above 2 GHz Northwood vs. Willamette vs. Athlon XP | |
| (Review by MS, January 7, 2002) |
Sandra for D.O.I.P.V
I am not going to compare apples and oranges or use a synthetic benchmark to show either the Athlon or the P4 up front. Rather, SiSoft Sandra offers a great tool to compare the different versions of the Pentium4 with each other. The rationale behind this is that some of Intel's viewgraphs show an increase in floating point performance of the Northwood compared to the Willamette whereas the white papers pretty much convey that there have not been any changes in the FPU architecture. Lets take a look at the demonstration of isolated performance values. The version of Sandra used was 2002 which will be publicly available next week at the SiSoft home page.
The results make it pretty clear that there may be differences between the two 2.0 GHz cores but if they exist, they are certainly well hidden in the noise. Of course, anyone who wants to can see the differences, so I won't argue
Sandra CPU Multimedia Benchmark
The exact same picture emerges for the SSE2 multimedia benchmark, the differences are possibly there but not significant.
These benchmarks are actually important to show that there are no hidded architectural changes that Intel doesn't want to tell us about. Further, as we show later, in real world benchmarks, there are quite dramatic differences between the Willamette and the Northwood but they have nothing to do with FPU or ALU performance and originate solely from smarter data management and/or a bigger L2 cache. Since we got this mystery part out of the way, we can delve right into the onslaught of benchmarks.
next page: => Office Applications =>
