Generating Custom Deep Camera Data in Shader ?

[djo]

Hey Guys,

 

I have been playing with a raytracing shader. The surface I apply the shader on and the actual position I am computing in the shader is quite a faire distance away. It's like I'm looking through a window. All works well and good however I am now looking into implementing Deep data in this shader

 

We can read deep data with dsmpixel. But is there anyway of writing it out from the shader ? At the moment my deep data comes out with the actual plane surface data that the shader is applied to but not the objects and procedural inside the "window" which my shader generates.

 

Do any of you have any clues on how to feed this custom data to Mantra do so ? Is it just an simple array variable to export or a twin array both depth and color/alpha ? Is it even possible ?

 

Thanks for any idea you might have.

 

J

[djo]

Upon further thinking if there is no straight process to do so in Mantra I'm thinking of perhaps writing a point cloud data with all this information a point per pixel and then find a way to convert it to a DCM file.