Soltek SL-K890Pro-939 Motherboard

In order to get each platform running in top notch shape, we optimized all three BIOS' for the best memory performance, including low timing and 1T command rates. We installed Windows XP with Service Pack 2, got DirectX 9.0c up and running, and turned off all of the power-saving features normally associated with the default Windows installation. Auto Updating and System Restore were similarly both turned off.

One very important point that does warrant attention is system drivers. NVIDIA is very proud of its Unified Driver Architecture, which lets you upgrade from an nForce2- or nForce3-based motherboard directly to nForce4 without having to format. You turn the system on and all of the devices just work. Should you find it necessary to install just the nForce4 standalone driver kit the drivers can be uninstalled, minimizing potential software conflicts.

VIA's driver package is another story entirely. Switching from K8T890 to K8T880 Pro and vice versa results in a failure to boot. Moreover, installing Windows from the K8T800 Pro board requires a driver floppy disk, while the K8T890 and its VT8237R don't. We'd certainly like to see VIA dedicate the same level of attention as NVIDIA to developing a more robust driver package.

Test System Specifications "AMD Inside!"

Preliminary Benchmarks with SiSoft SANDRA 2005 Synthetic Testing

We began our testing with SiSoftware's SANDRA, the System ANalyzer, Diagnostic and Reporting Assistant. SANDRA consists of a set of information and diagnostic utilities that can provide a host of useful information about your hardware and operating system. We ran three of the built-in subsystem tests that partially comprise the SANDRA 2005 suite (CPU, Multimedia, and Memory).  All of these tests were run with our processor set to its default clock speed of 2.4GHz (12x200MHz).  (Note: Some of these boards ran the processor with a default FSB higher than 200MHz. Where necessary, we manually set the FSB to 200MHz to keep the playing field level.)

For the most part, our processor tests come out fairly equal--as they should, given the consistency of our test systems. Even memory bandwidth numbers are somewhat flat, which makes sense since the Athlon 64 processor controls that link as well and all of our systems were set to identically low memory timings.