No, none of that, it's very easy...
- Delete AppleTYMCEDriver.kext in your S/L/E (make sure your your running a mac pro 3,1 or something that allows you to install the nvidia drivers).
- Edit your "org.chameleon.Boot.plist" under extras from boot drive by opening with textedit.
- Under the line "<key>GraphicsEnabler</key>" it should say "<string>Yes</string>" which is default by Chimera.
- Change this line to "<string>No</string>" this will stop chimera fron trying to inject your card, this is because the card is natively supported under OSX.
- Save the edited "org.chameleon.Boot.plist" file
- Install NVDIA CUDA from "http://www.nvidia.com/object/cuda-mac-driver.html"
- Install Drivers from here "http://www.nvidia.com/object/macosx-304.00.05f02-driver.html"
- Then shut down, plug card into your PCI Express 2.0 x16 or PCI Express 3.0 x16, it doesn't matter because the card will never saturate the lane. Don't forget to attach pci power cord from power supply.
- Plug monitor into card.
- reboot (you will notice chimera and the booting apple symbol will only appear take up about 3/4 of the screen, this is because the drivers have not yet been loaded, as soon as its done booting it should fill the screen).
- Your done, go under about this mac and make sure OSX recognizes it.
I haven't had a single kernel panic, crashes or anything. Works 100% with full acceleration, CUDA, OpenCL/GL, QE/CI, card throttling, and sleep. Hope I helped
-Aaron