They obviously are replacing shaders with Fixed Function as many times as possible.
I just dont understand the angle some of you guys take on these issues. Especially some of you with programming exprience Because you know darn well whats being done here.
I am Still Curious how the Two score so closely together at all. Aquamark is obviously using _PP hints in there base code already. Which as we have seen does not make up entirely for the Difference in performance. You will still see some nice looking HDR effects even at FP16.
What is happening here is they (Nvidia) are Replacing shaders in AQ3 with PS1.4 created shaders. Something they plainly said they were going to use in there Rebuttle to HL2.
Which is why i say any comparrison between ATi and Nvidia with Det50's are Completely invalid. As the Game will be Reporting Max Features, Or DX9 opr whatever but Nvidia is using PS1.4 all over the place. If the game is not reporting some kind of Mixed mode resut then what you ahve is not Driver Optimization.
But outright Fraud.
Overam Mirage 4700
3.2ghz P4 HT
SIS 748FX Chipset 800mhz FSB
60Gig 7200RPM HD
Radeon 9600M Turbo 128 (400/250)
Latest good read. [url]http://www.hardocp.com/article.html?art=NTc4LDE=http://www.hardocp.com/article.html?art=NTc4LDE=[/url]