Well, here we are. This is the part where it comes time to separate the men from the boys. The time to see what tales we've heard so far are truth and what are myth. Is that enough suspense?
We'll be doing a direct comparison today between the Soltek SL-KT600-R and another Soltek board, the SL-75MRN-L based on the nForce2 chipset. And keep in mind that this isn't even the newest nForce2 400 board, but a normal, run-of-the-mill nForce2 chipset. By all rights, the new KT600 ought to give it an old fashioned spanking, but we'll let the number do the talking.
Here is a quick rundown of the test rig:
Processor: AMD Athlon XP 1700+ (Thoroughbred "B")
Video Board: Sapphire Radeon 9700 Pro "Ultimate" (Supplied by Sapphire)
Memory: 2x 512MB Crucial PC2700 DDR
Hard Disk: Western Digital 80GB "SE"
Operating System: Windows XP Professional w/ SP1
Drivers Used: ATI Catalyst 3.5, nVidia 2.03 (nForce2), VIA Hyperion 4.48 (KT600)
Software Used: SiSoft Sandra MAX3, PCMark2002, 3DMark2001SE, 3DMark03, Code Creatures, Comanche 4, Unreal Tournament 2003, Quake III Arena
After the systems were installed for each motherboard tested, a fresh install of Windows XP was performed to make sure that all tests were run on equal footing. After updating the OS and installing the testing utilities, the benchmarks were run. This will allow us to see where each of these two platforms sit in terms of raw performance. Memory timings were run at 2-3-3-6 on both boards with a command rate of 2T for maximum stability and still providing acceptable performance levels.
SiSoft Sandra MAX3
SiSoft Sandra (System ANalyser, Diagnostic and Reporting Assistant) MAX3 is a synthetic Windows benchmark that features different tests used to evaluate different PC subsystems.
Sandra - CPU
Being a synthetic benchmark, SiSoft Sandra doesn't really compare to any real-world programs. But since it uses a constant set of instructions, it does manage to allow us to see a reliable comparison between test systems.
The CPU test is designed to show processor speeds while performing mathematical calculations. Since we have the same processor being used in both systems, any large disparities in this area will show a problem with the CPU pipeline on the motherboard. Since all numbers show a difference of roughly 1%, the KT600 based board does a fine job handling the processor installed.
Sandra - Memory Bandwidth
Since our KT600 board is still using a dated single channel memory controller, I had expected to see a bit larger difference in the memory department. But even with the nForce2 using a dual channel memory controller, it only managed a meager 2% higher score. But then considering that the KT600 is the newer chipset, it should have been able to put the dated nForce2 in its place. Not so, though, since the older brother manages to take the memory benchmarks.
Sandra - Hard Drive
While not specifically a motherboard test, I wanted to make sure that the hard drive was able to communicate through the Southbridge to the rest of the system with the necessary speed to be considered for an enthusiast rig. While not a huge difference, the KT600 did manage to pull off a 1.5% lead in this area.