nForce2 6-way Motherboard Roundup - December 2002

The Test and Other Notes

For this review we ran benchmarks using two CPUs: an Athlon XP 2000+ running a 266MHz FSB and DualDDR266 as well as an Athlon XP 2600+ running a 333MHz FSB and DualDDR333. The idea behind running two different CPUs was to show how the performance gap between nForce2 and the VIA KT400 changes once you move to 333MHz FSB CPUs.

266MHz and 333MHz FSB Performance Test Timings

All stock performance tests run on nForce2-based motherboards utilized two 256MB Corsair XMS DDR400 modules set to CAS2/2T/2T/4T timings and running at 333MHz (dual channel mode). Any other performance-enhancing timings (like Command Rate, Bank Interleave or Top Performance) were enabled in nForce2 motherboards that contained such timings.

All performance tests run on VIA KT333 and VIA KT400 motherboards utilized one 512MB Corsair XMS DDR400 module set to CAS2/2T/2T/5T timings and running at either 333MHz or 400MHz. Any other performance-enhancing timings (like Command Rate, Bank Interleave or Top Performance) were enabled in KT333/400 motherboards that contained such timings.

nForce2 Driver Notes

We have just a couple of things to mention about the nForce2 drivers used in our performance tests. First, lets list exactly what the nForce2 v. 2.00 drivers contain (per NVIDIA's web site):

Driver Versions

One of the great things about NVIDIA has always been their driver support for their products, and nForce2 is no exception. As you can clearly see in the list above, all you need to do is download and install one large file, and you'll have the latest drivers for your chipset, sound and video. The simplicity of this driver package is quite attractive.

Now that that's out of the way, we should mention that you must install Windows XP SP1 before you install the nForce v. 2.00 drivers, otherwise USB 2.0 and NVIDIA's own IDE drivers won't be installed. You can still install Windows XP without SP1, but NVIDIA's IDE driver won't be installed (you'll be using the Windows IDE driver instead) and USB 2.0 won't be installed either.

Lastly, we need to note that you can install the v. 2.00 drivers on all of the nForce2 motherboards tested here today except for the Chaintech 7NJS and ASUS A7N8X Deluxe. In the 7NJS's case, you can install the v. 2.00 drivers, but the audio portion of the install will be skipped entirely, meaning your sound won't be operational, but the rest of the v. 2.00 driver package will be successful. This is due to the fact that Chaintech uses C-Media's CMI8738 instead of the nForce2 APU.

In the A7N8X's case, you CAN install the v. 2.00 drivers, but the audio portion of the install will fail (i.e. it will be skipped entirely) because ASUS decided to assign different device ID's/product codes to the SoundStorm audio. However, the A7N8X's SoundStorm is still the same as any other SoundStorm audio found on any other nForce2 motherboard. We suggest you install the SoundStorm drivers that come with the A7N8X motherboard for best functionality.

Exact FSB Frequencies

All eight motherboards used in our benchmarks had approximately the same stock FSB setting. However, there was some tiny variation that we feel is necessary to point out:

FSB Frequency Comparison - 266MHz FSB w/ Athlon XP 2000+ CPU Clock Speed in MHz Chaintech 7NJS (nForce2) EPoX 8RDA+ (nForce2) EPoX 8K5A2+ (KT333) Leadtek K7NCR18D (nForce2) MSI K7N2-L (nForce2) Gigabyte 7VAXP (KT400) Athlon XP 2000+ Default Clockspeed ABIT NF7-S (nForce2) ASUS A7N8X Deluxe (nForce2) 1670.4 1670.4 1670.4 1670.4 1669.0 1668.1 1666.6 1664.1 1664.1 | 0 | 334 | 668 | 1002 | 1336 | 1670 | 2004

FSB Frequency Comparison - 333MHz FSB w/ Athlon XP 2600+ CPU Clock Speed in MHz ASUS A7N8X Deluxe (nForce2) Gigabyte 7VAXP (KT400) Chaintech 7NJS (nForce2) EPoX 8RDA+ (nForce2) Leadtek K7NCR18D (nForce2) MSI K7N2-L (nForce2) EPoX 8K5A2+ (KT333) Athlon XP 2600+ Default Clockspeed ABIT NF7-S (nForce2) 2103.0 2090.8 2088.1 2088.1 2088.1 2084.6 2083.8 2083.3 2079.6 | 0 | 421 | 841 | 1262 | 1682 | 2103 | 2524

Note that ASUS's A7N8X Deluxe had an unusually high default FSB setting. We still used this FSB setting in all our benchmark tests though, which is one of the reasons the A7N8X leads some of the benchmarks by a small margin.