Navigate: Advice Beginners BIOS Guide CPUs Links Mainboards Memory Network Storage Video/Sound Cards Contact Forum SiteMap Sponsors WebNews Home	.	.
Prices: 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 Search Prices:

Codecreatures is mostly limited by the graphics card and, thus, we didn't expect much of a difference between different settings as PC2700 vs. PC3200, AGP X8 or AGP X4. Suffice it to say that our expectations did not let us down, that is we did find the same tiny performance drop with DDR400 compared to DDR333, changing the AGP transfer mode went completely unnoticed by the benchmark.

After running CodeCreatures about 10 times in 1024x768x32bpp resolution we finally figured that the 0.1 fps performance difference was real. No surprise, though, in view of the other benchmarks shown. If changing the AGP mode made any difference, it is lost in the "lack of" granularity of the results.

UnrealTournament 2003

The demo is finally available and features a number of integrated benchmarks (thanks to Scott Wasson for pointing me in the right direction). UT2003 supposedly moves rather huge chunks of textures through the APG interface and if any of the current games will show a difference between AGP x 4 and AGP x 8, we figured, this would be the one. We ran the flyby-antalus benchmark at 1280 x 1024 x 32bpp with all details set to the maximum levels (Texture and Character Details=normal, World Detail=Highest, Physics Detail=high), everything else enabled.

After disregarding a minor performance hit for the DDR400 setting, UT2003 is the first benchmark to show a minute performance increment for AGP X8 but it still took 10 runs each and averaging to even see that one and only in PC2700 mode. Bottom line is, there is still hope that we will see some benefit some day, remember, it took about 2 years to see any difference of AGP X4 over AGP X2 and now there is very little doubt about the benefit of higher AGP transfer rates.

Content Creation Winstone2002

Let's shift gear a bit and look at some different types of benchmark. One of our main criticisms of Winstones, Sysmarks and similar benchmarks has always been that the limiting factor is not the performance of any subsystem but the bottleneck constituted by the I/O interface. In other words, using a fast HDD will give more performance increment than any upgrade in speed of the CPU or memory subsystem and we have another set of data to show it.

The system configuration is identical, however, we are comparing the Seagate Barracuda SATA V with its 8 MB cache and serial ATA 150 interface with the Maxtor D740X-6L (Parallel ATA 2 MB buffer). A roughly 7.5% difference in score would translate into some 15% difference in CPU speed at least.

Business Winstone2001

A similar situation is found with Business Winstone 2001. A Score of almost 79 is not bad at all but it is dwarfed by the score achieved using the Seagate Barracuda.

Again, what do we need faster CPUs for if a faster drive does it.

Time to wrap up this review...

next page:    => Overclocking, Conclusion =>