Then again, perhaps I should have anticipated 3DMark03's immense popularity. These days, PC enthusiasts seem to be engaging in more "my system is faster than yours" banter than ever before. 3DMark03 represents the latest arena in which geeks across the globe can pit their systems against each other in a battle for performance supremacy and bragging rights.

3DMark03's release also gives me an excuse to get my graph on. I've rounded up seven contemporary graphics cards all based on different graphics chipsets and run them through this latest benchmark gauntlet. How do graphics cards from ATI, NVIDIA, and Matrox compare? Let's lace up the gloves and find out.

The briefest of introductions
I could ramble on about 3DMark03's origins, explain all its tests in great detail, and showcase all sorts of pretty screen shots to introduce this new benchmark, but I won't. Futuremark's 3DMark03 product page already offers plenty of supplemental materials, and the sheer number of 3DMark03 downloads suggests that many will have seen the benchmark firsthand already. Today, we're after the benchmarks, so let's get to it.

Our testing methods
All tests were run three times, and their results were averaged, using the following test systems.

I've gathered up a wide range of graphics cards for testing today, but unfortunately, I don't have access to a GeForce FX to throw in the mix. Then again, the public doesn't really have access to the GeForce FX, either.

DirectX 9 was installed on the test system and the latest drivers from each manufacturer were used for testing. A set of updated Detonator 42.68 drivers was used with the three GeForce4 cards, though this particular driver isn't available to the public just yet.

I did try SiS's Xabre 600 with 3DMark03 and SiS's latest performance drivers, but to my dismay, many of the scenes displayed visual artifacts and just didn't look right, so I haven't included any of the Xabre's results.

We used the following versions of our test applications:

• 3DMark03

The test systems' Windows desktop was set at 1024x768 in 32-bit color at a 75Hz screen refresh rate. Vertical refresh sync (vsync) was disabled for all tests. The 3D gaming tests used the highest detail image quality settings in 32-bit color without any antialiasing or anisotropic filtering.

All the tests and methods we employed are publicly available and reproducible. If you have questions about our methods, hit our forums to talk with us about them.