X-bit labs

PCMark Vantage: New Complex System Performance Benchmark (page 8)

Category: Other

by Doors4ever , Ilya Gavrichenkov

[ 10/18/2007 | 08:06 AM ]

Three Platforms in PCMark Vantage

Our first impression from this test would be incomplete if we hadn’t checked it out in practice, i.e. for contemporary platforms comparison. Of course, the new Futuremark benchmark will very soon occupy and important position among our testing tools that is why very soon we will be able to offer you a pretty significant database of results. And in the meanwhile we decided to perform some preliminary tests on three different platforms based on a dual-core and quad-core Intel processors and a dual-core CPU from AMD.

Our testbeds were built using the following equipment:

• CPUs:
  • AMD Athlon 64 X2 6000+ (Socket AM2, 3.0GHz, 2x1024KB L2, Windsor);
  • Intel Core 2 Duo E6750 (LGA775, 2.66GHz, 1333MHz FSB, 4MB L2, Conroe);
  • Intel Core 2 Quad Q6600 (LGA775, 2.4GHz, 1067MHz FSB, 8MB L2, Kentsfield).
• Mainboards:
  • ASUS M2N32-SLI Deluxe (Socket AM2, NVIDIA nForce 590 SLI, DDR2 SDRAM);
  • ASUS P5K3 (LGA775, Intel P35, DDR3 SDRAM).
• Memory:
  • 2048MB DDR2-800 SDRAM (2 x 1024MB, DDR2-800, 4-4-4-12);
  • 2048MB DDR3-1333 SDRAM (2 x 1024MB, DDR3-1333, 7-7-7-20).
• Graphics card: OCZ GeForce 8800GTX.
• HDD: Western Digital WD1500AHFD.
• PSU: SilverStone SST-ST85ZF (850W).
• OS: Microsoft Windows Vista Ultimate x86.

The main difference between the test platforms was the CPU. Of course, it was not the only difference, the systems were based on different mainboards and used different memory. However, our experience shows that these are secondary factors that do not have that big of an influence on the system performance in real applications after all. Graphics cards and hard disk drives were the same in all three systems.

So, first of all let’s take a look at the main score provided by PCMark Suite:

The obtained results are quite adequate. In most cases, this situation can be observed between these systems on Intel and AMD processors in other modern applications. In other words, this benchmark provides pretty adequate estimate of the average system performance. At the same time, as you can see from the results, PCMark vantage is pretty nicely optimized for quad-core platforms which may make it fit for comparative analysis of upcoming multi-core platforms performance. Now let’s see what the other scores obtained from the rest of the tests look like.

I have to say that the overall picture is pretty much the same as in PCMark Suite. However, the developers didn’t actually promise anything else. All these test suites are specifically developed not to deny the results demonstrated in PCMark Suite, but to find out if any of the computer subsystems affect its overall performance in any way. That is why it makes sense to use any of the additional benchmarking suites besides the PCMark Suite only if you are comparing similar platforms, which don’t differ from one another dramatically.

Note that despite everything we have just said, the results of TV and Movies Suite stand out a bit. Here the AMD Athlon 64 X2 6000+ processor demonstrates higher performance than normally. The optimistic AMD’s score is most likely based on Futuremark developers’ belief that this processor copes well with video content processing tasks. However real application tests that we performed in our lab do not confirm this tendency. Therefore, we can question the results in this case.

Luckily, this is the only situation when the practical scores provided by PCMark vantage seem unjustified. The previous PCMark05 test was much more doubtful from this prospective, but nevertheless, it managed to become very popular. We hope that this only issue will not spoil the overall impression from the new PCMark vantage test.