Get almost peformance impact free Anti-Aliasing
 

There's a new AA method called FXAA, which is a shader based, post processing technique. Instead of just working on the edges of polygons, this is a full screen AA method that does the entire picture. FXAA3 adds only 0.20milliseconds post processing time to a 480gtx. There's almost no performance impact either.

The picture it creates is similar to SSAA, which is softer looking than MSAA, some people will like it, some won't. Here's the DLL wrapper to enable it, dunno if P99 will detect it as a cheat or not. If Rogean is only using buffer overflow attacks to detect MQuest, probably not:

http://neogaf.net/forum/showthread.php?t=439293

The text is a little too blurry by default. If you enable the default sharpening filter, it gets better but it's a little over sharpened. If someone can find a better sharpening algorithm to replace the default one with, the FXAA method might be pretty useful.

AMD has their own implementation of FXAA as part of the driver. You can enable it by checking Morphological AA inside Catalyst Control Center. It also suffers from the text blurring you mentioned with FXAA. The best aspect of Morphological AA and FXAA is they will work with games that have no native AA and you are unable to override standard AA methods via your video driver.

What is the point of this DLL wrapper? Does it work with cards that don't natively expose the mode within the driver?

this changes everything.

Unless you are cheating, don't worry about MQ2 detection.

Also, I used to have a version of iCEnhancer that worked with EQ. Not sure where it is now.

Only things that didn't work were motion blur. Everything else, improved shadows included, worked.

I'm worried if I use this for BF3 i'll get my ass punk busted

Likely so.

good news is Warden accepts it.

Bad news is it makes character select run like absolute shit, everything else is perfect.

I've been screwing with this one and the input lag is definitely higher than 0.20ms. Feels like at least 8-10ms. Might not be implemented correctly with this beta yet. Needs to fix that and include a better sharpening algorithm and it will be useful.