Testing and Benchmarks
Now that all the pomp and circumstance surrounding introductions is out of the way it’s times to get down to the meat and potatoes of any good video card review: The benchmarks. Normally, this first bit would have a boring graph talking about acoustics, but this card didn’t really produce any results. Our case remained at the same noise level, whether at idle or under full load. The fan never once incrased speed, except for the full speed spin up done on the initial boot up. Other than that, this card was whisper quiet. As for our test rig, here’s the full specifications.
- AMD Phenom 9600 Black Edition 2.3GHz Quad Core CPU
- Asus M3A32-MVP Deluxe/WiFi-AP 790FX Motherboard
- OCZ Platinum XTC REV.2 PC2-6400 2GB 2X1GB DDR2-800 Memory Kit
- Foxconn 9800GTX-512N Extreme OC Video Card
- Seagate Barracuda 500GB 7200.10 16MB SATA2 Hard Drive
- Cooler Master Hyper 212 CPU Cooler
- Antec P182 Advanced Super Mid Tower Case
- Thermaltake Toughpower 1000W Power Supply
- Microsoft Windows Vista Ultimate SP1
Now that we’ve established our benchmarking environment, let’s get on with the numbers!
3DMark 2006
First things first, we need to make with the synthetic benchmarks. That means it’s time to load up 3DMark06 and watch some demo screens…over…and over. Testing was completed using default settings, with only the resolution changing. We chose two of the three most widely used widescreen resolutions, and turned up some interesting results.
It would seem that, at least in 3DMark06, the 8800 GTX performs better than the 9600 GT. The 9800 GTX of course trumped everyone. Anyhow, the Foxconn 9600 GT card posted a performance drop of 4% at 1440×900, and 8% at 1680×1050. Given that 8800 GTX cards are still almost as pricey as the 9800 GTX cards (and about $125 more than the typical 9600 GT card), the 9600 GT still comes out a winner in a price-to-performance comparison. It would be interesting to see how these results scale when a second card is installed and SLI is enabled.
3DMark Vantage
Now it’s time to try out the new kid on the block. 3DMark Vantage has been something that many of us have been asking for, for quite a while. With more and more DirectX 10 titles rearing their heads, a standard way of testing DX10 performance was really needed. The new version does make some changes over it’s predecessor though. The composite score is only generated when you select one of the presets. Should you wish to customize resolutions or other settings, all you get is a GPU and CPU score. That being said the GPU score has far more weighting, and tends to be much closer to the composite score.
This change necessitated the use of two graphs measuring the results of the “Performance” preset, and the “High” preset. The settings for the “Performance” preset are similar to what you would find in 3DMark06. The resolution is set at 1280×1024, and everything is set to medium without and Antialiasing or Anisotropic Filtering. Now due to availability I was unable to include the 8800 GTX in testing, mostly due to the fact that I don’t have it anymore. Anyhow, the 9800 GTX beat the 9600 GT by a not-half-bad 29% overall.
Switching over the “High” presets put a little more strain on things. This time the resolution is ramped up to 1680×1050, all settings are turned to High, and 4x Antialiasing and 8x Anisotropic filtering is added. With all these settings dialled up, the 9600 GT posted a performance delta of 33%. This is fair degree more significant, and should give you an idea of how much performance you can expect for your dollar. The Foxconn 9600GT-512NOC is definitely a good value for the performance offered, but once again I’m left wondering if the difference would be made up with an SLI configuration.
Real-Time Price and Stock Check – Shop Like a PRO!