Foxconn 975X7AB-8EKRS2H - Page 7

..:: 975X7AB-8EKRS2H Benchmarks ::..

  • Intel Core 2 Duo E6700
  • Intel D975XBX
  • 1GB DDR-II 800MHz SuperTalent
  • 250GB Maxtor MaxLine III HDD
  • NVIDIA 6 Series - 6800 GT w/Latest Drivers
  • Windows XP Professional (SP2, DirectX 9.0c).

..:: SiSoft SANDRA ::..

To start off today’s performance testing results, we’re going to cover SiSoftware’s SANDRA benchmarking suite. The first benchmark that we ran was to determine Arithmetic performance of the 975X7AB-8EKRS2H in comparison to Intel’s D975XBX. As one would expect from the fact both boards are running off of the i975X chipset, the Whetstone and Dhrystone results virtually mirror each other, especially once error is taken into account. The 975X7AB-8EKRS2H appears to put up numbers a little over 1% better than that of Intel’s own D975XBX. This same pattern holds true again when the Multimedia results are examined, and again for the Memory bandwidth readings. In each of these results we see the 975X7AB-8EKRS2H holding an ever so slight lead.

..:: MBReview PRIBench ::..

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 rely on additional components within a computer system like the hard drive, PriBench does not rely on any of these exterior devices for the tests. PriBench is a system level benchmark, solely utilizing the performance of the processor and memory subsystem. It also focuses on comparing processor architectures and technologies and then seeing how much a core clock and/or FSB speed increase can be beneficial to the system. Our program is extremely accurate as we have witnessed time fluctuations of approximately .05 - .10 seconds.

On the PRIBench front, we can see that the 975X7AB-8EKRS2H is able to best Intel’s D975XBX by a fair margin all things considered. Typically, these results are very close, so while a performance gap might seem miniscule, it can in fact turn out to appear as a broader gain in later benchmarks. This is not always the case, however, so let’s keep moving and find out.

..:: ScienceMark ::..

To continue on with the synthetic benchmarks, we have a series of results from the ScienceMark benchmarking suite. First off let’s examine the results obtained with MemBench. The bandwidth results that were obtained should come as no surprise. Again, we see nearly identical scores placed by both the 975X7AB-8EKRS2H and D975XBX. This time, the D975XBX manages to overtake the 975X7AB-8EKRS2H by a slight margin. This flip-flop of results is not uncommon between boards, and actually appears to be very common in our testing. When it comes to overall latency and cycles required for a data packet of 512KB, an inverse relationship forms. Latency wise, both of the boards post nearly identical times, and also require the same number of cycles to complete the 512KB operation.

If you’ve ever happened to use the ScienceMark suite, you’ll likely be familiar with two of the other benchmarking applications, Cipher and Primordia. While I won’t go into the details of what exactly each of these processes accomplishes, we will quickly cover the results garnered from each. Starting off with the Cipher results, the 975X7AB-8EKRS2H manages to overtake the D975XBX by a slight margin. These results again repeat themselves in both the Primordia and Moldyn benchmarks. The performance gap between boards is still virtually nil.