Interesting, I have seeing that latest driver development has pushed the Radeon HD 3870 closer to the 8800GT than in the past, suddenly looks even more attractive than it used to be.
http://www.tomshardware.com/re...gtx-review,1800-6.html
http://www.bjorn3d.com/read.php?cID=1278&pageID=4849 <<Company of Heroes and World in Conflict performance is still anemic though...
http://www.overclockersclub.co...iews/xfx_9800gtx/6.htm <<Abnormally low scores on Crysis with the HD 3870 and abnormally high scores with the same card in Call of Duty 4, strangely the HD 3870 runs performs slighly faster in World in Conflict than the 8800GT, something that is the opposite in the bjorn3d review. Both reviews uses the same CPU but bjorn3d is overclocked, so that might be the explain the performance differences...
Is quite a feat that the HD 3870 is able to do Shader Anti Aliasing and being playable at the same time, look at Call of Juarez, it could run even better if it was DX10.1. Seems that it's complex architecture will be more suitable for the future, I would take the HD 3870 over the 9600GT anyday, specially now that it's known that the 9600GT will faint and will show it's weaknesses in future games, but not a bad card considering that's is almost a half of a 8800GT and is and it's performance is close to it, great feat.
http://www.beyond3d.com/content/reviews/47/8
So did AMD really build a better filter?
Arguably they didn't, in terms of overall image quality. The linear nature of the tent down-filter and the fact it takes into account contribution from neighbouring pixels means that image blurring is a natural side effect. There are simply better filter kernels to be applied, especially when you're looking outside the pixel, which is the key property of what AMD did. However balancing those filters against computational cost and hardware limitations is likely a key reason why tent was chosen.
That said, the key advantage comes not in terms of being able to apply these filters in the first instance; NVIDIA or anyone else is quite free to implement the same filters to give equivalent image quality, if that's what the customer is looking for.
No, the advantage comes because they built their hardware with a fast decompression path into their unified shader core. Samples and sample locations can be made known to the compute-heavy core of their latest chips in an efficient manner, in order for the filter to be computed and final pixel colour written. Current competing hardware doesn't have that luxury.