Understanding NVIDIA's FSAA
If you're interested at all in 3D graphics tech, hit Beyond3D as soon as possible to read this excellent article on NVIDIA's anti-aliasing techniques. Yes, there's been a raft of articles on FSAA popping up around the web recently, but this one is the most lucid explanation of NVIDIA's FSAA implementation I've seen to dateby far.
Check out how they clear up the question of whether NVIDIA's FSAA is "in the drivers" (i.e., software) or "hardware":
Now the reason that super-sampling is commonly referred to as being done through software is because of the multiplication of the vertex positions, as this is generally done by the CPU. However, with NVIDIA, they are actually able to do this through hardware. As we mentioned, the T&L unit is used for this multiplication, so the operations never have to be sent to the CPU. If the T&L unit was not present, this could be considered a software method, but because of it, the anti-aliasing is actually a hardware feature.
I had no idea they were able to do the sampling conversions on the GeForce chip, although it makes perfect sense. This one's recommended reading.