Thank you for quick response
Actually I'm implementing something similar to the Nvidia "Fermi Hair Demo" (
http://www.geforce.com/games-applicatio ... -hair-demo) and therefore it's absolutely necessary to access values (rather position information than colors) from buffers and textures in the domain shader.
But even for displacement, how could I ever move vertices generated by the tessellation stage based on a displacement texture? Isn't the domain shader designed for giving the new vertices their final position, based on additional information? And what would be the benefit of the tessellation engine when the shape of my geometry looks the same after tessellation like it did before (only subdivided)?
The Microsoft documentation of DX11 (shader model 5) says: "Texture2D type (as it exists in Shader Model 4) plus resource variables. This texture object supports the following methods in addition to the methods in Shader Model 4."
Further it's available for "Vertex-, Hull-, Domain-, Geometry-, Pixel- and Compute-Shader"
->
http://msdn.microsoft.com/en-us/library ... 85%29.aspx
I need to read information from buffers/textures within the domain shader based on the patch- and segment-IDs and it makes unfortunately no sense to me to provide all this data by the vertex shader (besides it would be too much data).
It sounds like you haven't ever tested this feature yet, but if it works for you or you see what the problem is, I would be very happy and thankful if you could share the information
