Stress Testing the Soyo KT400 Ultra

Overall, we were very impressed with the stability and reliability of the Soyo KT400 Ultra motherboard using different types of memory and hardware configurations.

Looking at the ASUS A7V8X (KT400) and ECS L7VTA (KT400) motherboards, it's clear that DDR400 offers nothing in the way of additional performance. Therefore, it's understandable that some users are indifferent or simply don't care to use DDR400 with any Athlon XP motherboard. However, we believe that with consistent BIOS updates, several KT400 motherboards will eventually show a performance improvement with DDR400 memory versus DDR333. We think this could be the case with Soyo's KT400 Ultra. This motherboard simply performed very poorly with DDR400, which could be blamed on both Soyo and VIA (although Soyo seems to be more at fault here, as you're about to see in our benchmark results). You take a pretty massive performance hit when enabling DDR400 with the KT400 Ultra, so we suggest staying with DDR333 for this motherboard.

This is where the good news comes into play. As you're about to see in our benchmarks shortly, the Soyo KT400 Ultra is more or less the fastest DDR333 Athlon XP motherboard we've tested here at AnandTech using an Athlon XP 2000+ CPU, GeForce4 Ti4600 video card, and Corsair CL2 DDR400 memory. This isn't the fastest (stock) Athlon XP setup in existence, but it puts up some pretty good numbers nonetheless.

However, what's even more impressive about the Soyo KT400 Ultra's DDR333 performance are the memory timings it's able to run at. We tested five different types of memory (Samsung, Kingston, Twinmos, Mushkin, and Corsair) all set to operate at 333MHz. Running our usual set of benchmarks with each memory type, we encountered no stability or reliability issues, and performance was always top-notch compared to competing KT333 and KT400 motherboards. In addition, we were able to run this memory at the fastest timings we've ever seen.

Of course, we then decided to see how well the KT400 Ultra would stand up against our stress tests by populating all 3 memory slots with DDR333 (we opted to use high quality memory, and that meant going with 3 sticks of Corsair CL2 DDR400 downclocked to DDR333). Running 3 sticks of CL2 Corsair memory at DDR333, we were able to achieve the following timings running Prime95 for 24 hours and rerunning our benchmark suite twice:

Stable Memory Timings
(all banks populated)

Clock Speed
166MHz
Timing Mode
Turbo
CAS Latency
2
Bank Interleave
4-bank
Precharge to Active
2T
Active to Precharge
5T
Active to CMD
2T
Command Rate
1T

These are very impressive timings to say the least. The ECS L7VTA (another solid KT400 motherboard) was able to achieve these same timings using Corsair CL2 DDR400, however bank interleave had to be disabled for the L7VTA to operate flawlessly, whereas the Soyo KT400 Ultra was easily able to handle 4-bank interleave.

Knowing that DDR400 performance took such a huge performance hit, we didn't think it was necessary to even bother running 3 DIMMs of the stuff with the KT400 Ultra. Perhaps Soyo will release a new BIOS that will improve on DDR400 support in some shape or form, but that's just unconfirmed speculation on our part.

The Test

Performance Test Configuration

Processor(s):
AMD Athlon XP 2000+
RAM:
256MB Samsung DDR333 CAS2.5 module (M368L3223CTL-C (L) B3)
256MB Mushkin DDR400 CAS2.5 module
256MB Corsair DDR400 CAS2 module
Hard Drive(s):
Western Digital 120GB 7200 RPM Special Edition (8MB Buffer)
Bus Master Drivers:
VIA 4-in-1 4.43
Video Card(s):
ASUS V8460 Ultra NVIDIA GeForce4 Ti 4600
Video Drivers:
NVIDIA Detonator 30.82
Operation System(s):
Windows XP Professional
BIOS Version
1R

The AnandTech Motherboard Testbed was Sponsored by Newegg. You can buy the components we used to test at www.newegg.com.

BIOS and Overclocking Content Creation & General Usage Performance
Comments Locked

0 Comments

View All Comments

Log in

Don't have an account? Sign up now