DFI NFII Ultra Infinity - Page 7

..:: SiSoft SANDRA ::..

The first benchmark we will be using today is SiSoft SANDRA, one of the most popular synthetic benchmarking programs. In the cpu arithmetic test, you can see that there is virtually no difference between the two boards. This synthetic benchmark thrives on the core clock of the cpu, and since both motherboards clock the processor to virtually the same frequency, each platform should show that they are equal in performance, which they do.

Moving on to the SANDRA cpu multimedia tests, the NFII Ultra Infinity is able to nudge ahead of the 7VT600-L, but not by much. So far it seems as if a budget KT600 motherboard can keep up with the NFII Ultra Infinity. But the story unfolds, however, when we take a look at the SANDRA memory bandwidth tests. The 2% average gain in performance is provided by a boost in theoretical memory bandwidth that the nForce2 dual-channel DDR memory controller can offer over the KT600 chipset. Hopefully the NFII Ultra Infinity will be able to increase its' lead in the rest of our benchmarking suite.

..:: MBReview PriBench v1.04 ::..

Using our very own in-house benchmark, MBReview PriBench uses a computationally intensive algorithm to compute several hundred millions of prime integers. Unlike other synthetic benchmarks which relies on additional aspects within a computer system like hard drive speeds, PriBench does not rely on any of these exterior devices for the tests. PriBench is a system level benchmark, and solely utilizes the performance of the processor and memory subsystem. Our benchmarking program is extremely accurate as we have witnessed time fluctuations of approximately .03 seconds.

With a quick look at the results, PriBench seems useless. But our benchmark focuses more on showing the power of comparing processor architectures and technologies, and then seeing how much a core clock and/or FSB speed increase can be beneficial to the subsystem. Since we are comparing platforms that use the exact same processor technology and core clock speed, there will be virtually no difference. But it is interesting to see how extremely accurate our program is, knowing that the NFII Ultra Infinity only has a 1.39MHz core clock speed advantage.

..:: SuperPI ::..

In the Super PI tests, we run the program four times; once at one million, once at two million, once at four million, and once again at eight million. The numbers in the graph above show the time in seconds that it took the system to calculate prime integers to the set number of digits. In this benchmark, we can see without a doubt that the NFII Ultra Infinity is able to show an average performance gain of 8% over the fastest KT600 solution we have ever tested.

..:: Specviewperf 7.0 ::..

So far everything looks good for the NFII Ultra Infinity, but the SPECviewperf benchmark will tell you another story. The KT600 chipset based motherboard shows that it is able to outperform DFI's new nForce2 offering by a slight margin, but the Infinity took a huge stride forward in the drv08 test. The scores for the NFII Ultra Infinity are around where other nForce2 boards have demonstrated, so at least the DFI board is being consistent.

..:: FutureMark 3DMark2001SE ::..

3DMark2001 SE is first off on today's list of multimedia application benchmarks. The DFI NFII Ultra Infinity is able to power ahead of the Gigabyte 7VT600-L by a sizeable margin. In this benchmark and in those remaining, we'll really begin to see how much performance the dual-channel DDR nForce2 Ultra 400 chipset can offer in comparison to the single-channel VIA KT600 solution. For the 3DMark2001 SE tests, we are seeing a performance difference of roughly 4% in favor of the NFII Ultra Infinity. This number is right around where we expected it to be.

..:: Quake III Arena ::..

The next benchmark we have for you is embedded within my favorite computer game, Quake III Arena. There is no question that the relevance of this benchmark is coming to an end soon, but overall performance can improve dramatically with increased system memory bandwidth, even with the fastest video cards on the market. It is no surprise that the NFII Ultra Infinity is again able to walk away with another win because of the superior memory performance it can deliver when using the nForce2 Ultra 400 chipset over the competition.

..:: Unreal Tournament 2003 ::..

The last gaming benchmark we will test our motherboards with is Unreal Tournament 2003 Demo v2206 with the HardOCP 2.1 version benchmarking add-on program. I have found that the add-on can save a lot of time when benchmarking, has a very simple interface, and is already pre-configured to test a variety of platforms equally instead of having to manually adjust any of the in-game settings myself. I chose to benchmark our motherboards using the Direct3D engine with low quality tests at a 640x480 resolution. DFI's new nForce2 motherboard finally puts the last nail in the coffin of the KT600 when it comes down to gaming performance. We are seeing a rough 7% increase in performance from DFI choosing to use the nForce2 Ultra 400 chipset alone.