AMD Radeon HD 7990 Review > Testing Methodology

Testing Methodology

As usual we tested each card with Fraps, which lets u.s. record the average frame rate in seconds over a set amount of time. Typically, nosotros run our tests for 60 seconds. Reporting the average fps (frames per 2d) is how things have been done for... well, forever. Information technology'south a fantastic metric in the sense that it's easy to tape and like shooting fish in a barrel to sympathize. But information technology doesn't tell the whole story, as The Tech Report and others have shown.

To get a fuller film, it'due south increasingly credible that yous need to gene in a card's frame latency, which looks at how quickly each frame is delivered. Regardless of how many frames a graphics menu produces on average in threescore seconds, if information technology can't deliver them all at roughly the same speed, you might encounter more brief jittery points with one GPU over another -- something nosotros've witnessed merely didn't fully sympathize.

Assuming ii cards deliver equal boilerplate frame rates, the one with lowest stable frame latency is going to offer the smoothest picture, and that's a pretty of import detail to consider if you're near to drop a wad of cash. As such, we'll be including this data from now on past measuring how long in milliseconds it takes cards to render each frame individually and then graphing that in a digestible manner.

We'll be using the latency-focused 99th percentile metric, which looks at 99% of results recorded within X milliseconds, and the lower that number is, the faster and smoother the performance is overall. By removing 1% of the most farthermost results, it'southward possible to filter anomalies that might have been caused by other components. Over again, kudos to The Tech Written report and other sites like PC Per for shining a light on this effect.