GTX 480 vs. HD 5870, 8x AA Performance Analysis, Part 3
Heaven benchmark, Unigine Engine
Now we test the synthetic Heaven 1.0 benchmark based on the Unigine engine. It uses DX11 and fairly heavy tessellation which will strain any graphics card. Here are the settings we used for this benchmark (we checked ‘full-screen’).
Here is our benchmark run at 2560×1600. The HD 5870 occasionally produces a slideshow with 8xAA; the GTX 480 is definitely faster at extreme tessellation but it is no pleasure to watch it also struggle:
Now we look at 1920×1200:
And finally we run the benchmark at 1680×1050:
It won’t be until later this year that we will see our first DX11 games based on the Unigine Engine. For what it is worth, GTX 480 excels in this benchmark.
Let’s just include Heaven 2.0 with Heaven 1.0 as they are both Unigine engine; version 2.0 is using much more extreme tessellation with nicer visuals as a result.
Heaven 2.0
There will be at least two DX11 games based on Unigine that will be released later on this year. And there is the latest and even more stressful Unigine 2.0 benchmark and the settings we used:
As you can see there is a setting for “extreme tessellation”. We will tell you right now that this test chokes the GTX 480 at the highest settings but it is still better than the slide show the Radeon HD 5870 manages. However, the visuals are also extraordinary. Here are the results at 2560×1600:
The GTX 480 takes a relatively small performance hit when 8xAA is applied compared to the HD 5870 which just chokes. And now we look at 1920×1200:
And finally we check out 1680×1050:
Of course, this is a synthetic test based on a game engine that has yet to see a PC game that uses it in retail. But it is worth noting the tessellation capabilities of the GTX 480 and its performance with 8xAA.
Please take into consideration that nVidia uses a different version of AntiAliasing starting 8x and up, therefore comparisons are henceforth limited at best. Sadly I don’t have a direct link right now, but please take it into consideration before drawing (final) conclusions.
We took special care to make sure that identical AA settings were applied in all of our benchmarks including for Crysis. We even noted that in the full retail game, Just Cause 2, that we observed the benchmark results showed the Radeon was running at 8xCSAA while the GeForce was 8xAA.
However, we have since learned from AMD that the benchmark results are wrongly identifying 8xMSAA as CSAA. The Radeon is actually running 8xMSAA and this minor issue will be addressed in a future patch.
Everything we test is “apples to apple” unless it is specified in the review.
Nice article man. Cheers
good