If you are a proud meat-eating Canadian who loves their ATi card, do not download and run Heaven 2.0. It's full of craptacular buggery, no x-fire support and CUDA calls up the yingyang. It may just even hose your card.

This is the special Fermi-edition. Nvidiots only, please.

And the benchmark buying begins.

Changes in version 2.0

Heavier tessellation load
Several major optimizations of the engine
Added new elaborated objects in the world
Physics-driven flags
More dynamic lights
Enhancements of some old assets
Introduced "moderate" and "extreme" tessellation modes

As much as I hate this kind of thing what I would like to see is for a benchmark to come out with separate ATI/nVidia optimized benchmarks just to see what each is capable of when highly optimized for one and the other. It could turn out that ATI cards have more raw power but the drivers/Cuda for nVidia are better suited for optimizations. And if that's the case maybe nVidia would instead throw some money at game dev's to include more optimizations for their cards (hopefully not at the expense of hindering ATI's performance but who am I kidding it would probably end up that way) because really, it should come down to game performance and whether that performance comes from hardware or software it shouldn't matter to the end user.

I been saying that for quite awhile. Benches are ok for giving you a ballpark idea on whats going on. They are no way near fair.

But they (NV) are obviously and blatently throwing resources (time, money, devs) into a benchmark. Anyone even trying to compare 2 cards at the same settings and getting useful results would either was be naive or stupid, and laughed out of town.

Sounds like the description of a typical fanboi to me

Hard to cheer for team green in this day and age.

I'm going to give it a try this weekend. See what its all about.

The first release of Heaven should be used just to compare the differences in tesselation and see what DX11 can do...

How could it possibly hurt the card?