Navigate: Advice Beginners BIOS Guide CPUs Links Mainboards Memory Network Storage Video/Sound Cards Contact Forum SiteMap Sponsors WebNews Home	.	.
CPU Intel P4 840 D P4 820 D P4 630 P4 640 P4 650 P4 660 P4 670 AMD Athlon64 3500+ 3700+ 3800+ 4000+ X2-3800+ X2-4200+ X2-4400+ X2-4600+ X2-4800+ 1-Way Opteron Opteron 144 Opteron 146 Opteron 148 Opteron 150 Opteron 152 2-Way Opteron Opteron 240 Opteron 242 Opteron 244 Opteron 246 Opteron 248 Opteron 250 Opteron 252 2-Way Dual Core Opteron Opteron 270 Opteron 275 nVidia GF 7800GT GF 6800GT GF 6600GT ATI R X850 XT PE R X850 XT R X800 XT PE R X800 XT R X800 XL Memory Corsair Crucial Kingston Mushkin OCZ What are you shopping for?

Test configuration

Test System I

• ASUS CUV266
• Intel PIII 500E
• ATi Radeon 64 DDR
• IBM DTLA 302070
• ASUS 40x CDROM
• Alps Floppy disk drive
• Dlink DFE 530+ fast Ethernet adapter
• Win98 first edition

• EPoX 8KHA
• AMD Thunderbird 1.2 G (266 MHz bus interface)
• MSI StarForce 822 (GeForce3)
• IBM DTLA 302070
• ASUS 40x CDROM
• Alps Floppy disk drive
• Dlink DFE 530+ fast Ethernet adapter
• Windows98 First Edition

Testing criteria

Some of the latest mainboard have certain safety features that, in case of exceeding the bootable settings, autonegotiate the highest viable BIOS configuration instead of defaulting back to default settings or else requiring a Clear CMOS. This "smart" feature is a definite advantage for the standard user, for those who try to overclock and benchmark their systems to the point where it breaks, it causes the need for constant verification of all settings, starting from the FSB to the memory latency settings. Otherwise, a 155 MHz negotiated by the BIOS might easily be mistaken for the 180+ setting specified in the BIOS. The measures taken to assure that the system was running at the settings specified were:

• WCPIUD to verify the clock / bus speed
• SiSoft Sandra Memory Benchmark to verify that the system had not defaulted back to PC1600 mode

• Expendable Timedemo
• Ultra-X RST Pro
• Sysmark 2000

Cross Platform Comparison

In most cases, when it comes down to hardware and testing results, very little can be transponed from one platform to another. As it turned out, however, the stability and FSB limits for the different DDR DIMMs were identical within 1 MHz from the ASUS CUV266 and the EPoX 8KHA and, therefore, appear to reflect accurately the limitations of the DIMMs used rather than platform-specific aberrations.

One Word on Performance

Time and again, I read that there are differences in bandwidth between different DIMMs on the same platform. I have not been able to ever replicate these findings, that is, if the BIOS settings are kept constant, there is no difference in bandwidth, regardless of what DIMM is used. There are factors, though, that more or less influence the SiSoft Sandra results. Such issues are:

• Repetitive runs: Since the benchmark occupies a certain memory space, another run will not necessarily write and read to the same memory areas. This changes the number of possible page hits and misses and, therefore will lead to different results.
• Other software that was run before SiSoft Sandra was started: See above.
• Memory and chipset temperature: Simply by increasing the VI/O (or VDD), the scores can be altered, the key issue here is the chip temperature and the thermal slowdown.

Since benchmarking is not the real issue here, we'll leave it like this and assume that all DIMMs perform near-identical at the same BIOS settings (including FSB)

next page:    => Micron vs. Crucial; Mosel-Vitelic =>