Hi guys!
I have been lucky enough to test Maxwell 3 for it's Siggraph debut! I'm not actually supposed to speak about it, so let's keep this quiet until they make an official announcement sometime after Siggraph.
Amazingly, Version 3 actually generates a finished render
before you press the render button by utilizing new render technology derived from their extensive research in the field of String Theory. This is completely hardware independent, being solely software driven. Therefore, no special hardware is required (no multiple GPU's) such as is required with so called "realtime" engines. I was thoroughly impressed when several times during testing, a finished render (at S.L. 50!) popped up on my screen before I even had the scene completely modeled, allowing me to see a possible result of where the scene was heading and make "preemptive" changes.
Below is an example of a render that was produced only 5 minutes after I started creating the scene in Max:
And here is what the scene looked like just before the render was produced:
Maxwell Version 3:
"As Easy as Building Your Own Linear Accelerator and Creating Tiny Universes"
Brian Looney
Maxwell + RealFlow
http://www.turbosquid.com/Search/Artists/Bubbaloo
Help! I'm dispersing into a haze of probability!