Test Setup And Benchmarks

Picking the hardware for today’s performance testing was particularly difficult. On one hand, it’s clear that Lucid put a lot of effort into optimizing and qualifying ATI’s Radeon HD 4000-series and Nvidia’s GeForce GTX 200-series hardware. And we know that most of the company’s work went into Windows Vista, only recently shifting toward Windows 7.

With that said, Windows 7 is the de facto now. It’s the OS that lets us run with Catalyst 9.12 and GeForce 195.62 installed at the same time. It’s what makes X-mode possible. It’s in for sure.

Nvidia’s GeForce GTX 200-series is really all Nvidia has right now. And while GTX 260s and 275s are probably the most common models, it’s hard to imagine anyone with less than a GeForce GTX 280 or 285 eyeing a $350 P55-based motherboard. The GeForce GTX 285 is a logical choice, too.

The more difficult choice was ATI’s Radeon HD 5870, supported by Lucid’s 1.4.1 driver as of December 31^st, 2009. It becomes clear in the benchmarks that support for the 5000-series cards is still early, and we’re willing to take this into account as we go through our analysis. We’re just glad the company’s software team was as timely as they were—again, it’s hard to imagine anybody buying a $350 motherboard to mix their $150 Radeon HD 4000-series cards with an Nvidia board on the cusp of being eclipsed.

For each set of tests, we’ll run three resolutions with no anti-aliasing or anisotropic filtering and then three resolutions with the eye candy turned on. For the no AA/no AF scores, we’ll also include a baseline single Radeon HD 5870 score for evaluating the overall effect of running in CrossFire or A-mode. One thing we're missing here is an example of dissimilar cards from the same vendor. This is something we'll possibly spend more time evaluating as new drivers arrive, shoring up compatibility in our gaming suite.

With the stage set, let’s move on to the numbers.