LoD+SGSSAA = Hardware Destruction
I'm glad to hear it has a good foundation though, but yes Nvidia needs to add LoD to DX10+ and bring out some actual GPU power so you can enable it.
Edit: My point was only to address the assertion some people had made about advanced AA methods and the IQ they present at face value. Some seem to discredit this new approach while praising the old way and shooting down the new way at the same time when they both have the same blurring effect.
This method works very well it seems, the IQ is at levels I'd expect, you said it reacted well to moving images which is a big plus, and it doesn't have near the performance impact of the older methods.
I think it's a step in the right direction, but we need a way to get the detail back in DX10+ as well as the gpu power to handle it all.