Testing For Memory Interface Limitations: 1920x1080
Our first comparison benchmark doesn't employ any MSAA, but instead is run with FXAA enabled. The finishing order goes: GeForce GTX 670, Radeon HD 7950, GeForce GTX 660 Ti, and Radeon HD 7870.
The two Radeon HD 7750s fell into last place by quite a bit, but they still performed well enough considering power consumption less than 100 W, combined. Together, they cost about $220.
It’s interesting that the 2 GB GeForce GTX 660 Ti consistently beats the 3 GB model. So far, it looks like 3 GB of GDDR5 are too much for the 192-bit interface to handle efficiently.
The results start changing as soon as we apply 2x MSAA. The GeForce GTX 670 and Radeon HD 7950 are still on top, but the Radeon HD 7870 now beats Nvidia's GeForce GTX 660 Ti.
With 4x MSAA applied, the Radeon HD 7950 passes Nvidia's GeForce GTX 670, putting AMD in the lead in both segments. Of course, we pushed things even further, though.
Nvidia’s GeForce cards simply do not do well with 8x MSAA applied. The Radeon HD 7950 and Radeon HD 7870 not only beat their competition, but AMD's Radeon HD 7870 even beats the GeForce GTX 670.
The GeForce GTX 660 Ti can’t even keep up with two passively-cooled Radeon HD 7750s.
We’d also like to say a few words about minimum frame rates in this benchmark. Nvidia’s GeForce cards lose the race big time, no matter what anti-aliasing settings we use. Subjectively, a single GeForce GTX 660 Ti under the effects of 8x MSAA is worse than the already-bad CrossFire setup. That's right: Nvidia's stuttering under those settings is more annoying than the micro-stuttering typical of many CrossFire arrays. And this is in spite of our efforts to pick settings that yield playable frame rates on all cards at every setting. We really can’t recommend Nvidia's GeForce GTX 660 Ti if you plan to use 4x or 8x MSAA; even two Radeon HD 7750s are a better choice.