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:

*we were using the demo version since we found up to 5% performance variation between individual runs of the retail version benchmarks. The variability appeared to be caused mainly by low response / seek times of the CDROM (having the CD in the drive is a requirement for launching the retail version of UT2003 including the benchmarks). The demo version did not show the same fluctuation in results and even though it is hampered by other limitations it was perfectly suited for the purposes of this review.

ALL benchmarks shown were run using 2 x 256 MB DDR3200 and the Athlon XP2800+

Overclocking

Usually, we are putting our overclocking experiences towards the end of the review, in this case, the situation is somewhat different in that there are a few issues that need to be addressed in more detail.

Multiplier Selection

We already mentioned that the multiplier selector works pretty much "unconditionally", that is even with the Thoroughbred core-based XP2800+, with the sole restriction that settings above 22x did not show. To be fair, it needs to be said that these settings only showed on the ASUS A7N8X after modifying the board to unlock the Thoroughbred..

It has been shown in numerous occasions that the 333 MHz FSB processors do not like to run at higher FSB frequencies, something like a bus-threshold known from the earliest samples of the K6-2 CXT core appears to hold back the processor unless one of the L12 lines is cut or blown.

None of this appears to exist on the NF7, we were running the XP2800+ (unmodified) happily at 200 MHz FSB without stutter or barfing, however, this was only possible at multiplier settings below 9x. Since the resulting frequencies of e.g. 10 x 200 MHz are still by far not challenging the default clock frequency of the processor, something else must play into the problems here.

Increasing the chipset voltage to 1.7V added stability at high frequency whereas increasing the AGP voltage, once again, decreased system stability. Given everything we just mentioned, it would be easy to blame the board power for the issues we saw, but most likely, the real culprit is something we have seen and documented for the past 5 years and that is the strange phenomenon that high clock speed on the processor end of things works against high FSB frequencies. The only possible explanation is that higher CPU clock rates and the associated performance increase causes higher data demand which, in turn, causes more data traffic which becomes the limiting factor. Granted that this should not happen but we have seen it over and over again and so we assume it does.

next page:    => Performance: Sandra =>