PIXEL SHADER MATH PRECISION DIRECTX 9 CLASS HARDWARE

One of the most significant though subtle changes that was introduced in DirectX 9 pixel shaders is higher precision registers. The pixel shader 2.0 temporary registers and constants must have a minimum of 16-bit signed floating point. This allows for much more complicated and subtle shaders and reduces the chances of banding due to loss of precision. Some hardware implementations will support a full 32-bit IEEE float precision. Input texture coordinate registers are a minimum of 24-bit signed floating point numbers. The input color registers are unchanged from the DirectX 8 shader implementation and are still limited precision numbers in the [0,1] range.



Real-Time Shader Programming(c) Covering Directx 9. 0
Real-Time Shader Programming (The Morgan Kaufmann Series in Computer Graphics)
ISBN: 1558608532
EAN: 2147483647
Year: 2005
Pages: 104
Authors: Ron Fosner

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net