RBURKE
January 28th, 2009, 17:46
So this is my first post, i have been playing around with mental mill for quite a while and have been able to make a lot of great shaders for use in maya.
However i am now in need of some guidance in creating a particular effect for my shader.
What i want to do is create a shader where the image for the cube map to be used in reflection is calculated by a set of cameras in a scene. A long time ago i did something similar in blender where a set of cameras was placed inside of an object looking towards each end of each axis. One camera for x+ one for x- one for y+ y- etc. Every frame these cameras would take a shot of the scene however none of them can see the target sphere. These images where them combined in a cube map and applied to the model creating a simulated reflection of the scene around the object. Coolest part about his was that i could move the object and the reflection would update.
What i would like to create is a similar setup but in real-time. So my question is is it possible to create a node in MM that will build the cube map based on input from the cameras. OR is anything like this not possible in mental mill. Also would i be able to tell these cameras to be updated in real-time and stored to a buffer or something creating a real-time simulated reflection. Any suggestions or info would help, but some example networks would be even better.
However i am now in need of some guidance in creating a particular effect for my shader.
What i want to do is create a shader where the image for the cube map to be used in reflection is calculated by a set of cameras in a scene. A long time ago i did something similar in blender where a set of cameras was placed inside of an object looking towards each end of each axis. One camera for x+ one for x- one for y+ y- etc. Every frame these cameras would take a shot of the scene however none of them can see the target sphere. These images where them combined in a cube map and applied to the model creating a simulated reflection of the scene around the object. Coolest part about his was that i could move the object and the reflection would update.
What i would like to create is a similar setup but in real-time. So my question is is it possible to create a node in MM that will build the cube map based on input from the cameras. OR is anything like this not possible in mental mill. Also would i be able to tell these cameras to be updated in real-time and stored to a buffer or something creating a real-time simulated reflection. Any suggestions or info would help, but some example networks would be even better.