Shaders for Multi-core processors

What are challenges for writing new code for a multi-core processor
What are challenges for converting existing shaders to multi-core processor

Transactional Memory

TM is software feature being researched for parallel programming.

For a multi-core computation environment, mutliple threads work on a shared memory space. Often have to deploy RMW where programmers use lock, mutex etc. For large number of parallel threads it becomes very hard for SW writers to keep track of locks. Its even harder to debug any concurrency issue that could result in deadlocks.

TM offers an atomic RMW operation without explicit need of lock for programmer.

FPU/ Double Precision in Graphics Processing

Nvidia is pushing use of GPUs for scientific computation. Scientific applications require higher floating point precision. Nvidia Tesla Product line now supports Double Precision. GT200 FPU sucked! Currently in workloads there is 8:1 (single precision: double precision), in future this ratio will get closer to 2:1.

Almost all consumer apps are single precision today.

Game physics doesn't need high precision FPU, as picture needs to look good. It might not have to be totally accurate.

Science/ Physics need to be very accurate, so higher precision FPU is required.

DX11 Rendering Pipe

DX10 Rendering Pipe

A Very Good RPM:

Input Assembler

Vertex Shader

Geometric Shader

Rasterizer

Pixel Shader

Merge

Graphics Benchmarks

3D Mark

VantageP

Futuremark is the company that released 3DMark06. After DX10 & Vista releases in 2008, instead of releasing 3DMark08, they released Vantage. These benchmarks are geared towards checking HW capabilities for latest games development. Vantage requires to have Vista OS.

Key Technologies

SLI: GPU to GPU interface introduced by Nvidia. This interface allows freeing up PCIe channels & allow GPU to GPU communication. They send Display info, ???, ???

Anti-Aliasing: filter out higher freq to remove edges. Smooth the image

Tessalation:

Ray Tracing: