3DMark, 3DMark11, Unigine Heaven

That we have taken a look at our gaming tests using our Frame Rating metrics and testing methods, we are including a set of tests from more standard benchmarks like 3DMark 11, Unigine Heaven and the newer 3DMark benchmark. 

I consider these tests to be somewhat of a "best case" for all the cards in our comparison.  We aren't using our frame capture system, we aren't measuring frame latency, nothing like that; I think this should give you an idea of graphics performance if each vendor had the best result for each game.

3dmark11-e.png

This result puts the HD 7990 just behind the HD 7970s in CrossFire, what I more or less expected based on the specifications and clock speeds of the cards.  With the HD 7990 running about 100 MHz slower than the HD 7970s we do see lower benchmark results in our "best case" testing.  The GTX 690 is faster than the HD 7990 and the GTX Titan is slower, though the GTX 690 is much closer to the new AMD option.

3dmark-fs.png

3dmark-fse.png

Wow, now there is a significant gap between the AMD and NVIDIA options.  The HD 7990 is running 11% faster than the GTX 690 and 35% faster than the GTX Titan at the Extreme Fire Strike preset!

heaven.png

Our Unigine Heave results are actually interesting as well, with the HD 7990 outpacing the GTX 690 at the "normal" tessellation levels but actually coming in behind it when the tessellation is cranked to the "extreme" level.  Obviously the tessellation performance of the Kepler architecture is paying dividends in some areas.

« PreviousNext »