GV-N295-18I-B (rev. 2.0)

Ensure top-notch compatibility and performance for all OpenGL 2.1 application.

Shader Model 4.0 adds support for indexed temporaries which can be quite useful for certain tasks.Regular direct temporary access is preferable is most cases. One reason is that indexed temporaries are hard to optimize. The shader optimizer may not be able to identify optimizations across indexed accesses that could otherwise have been detected. Furthermore, indexed temporaries tend to increase register pressure a lot. An ordinary shader that contains for instance a few dozen variables will seldom consume a few dozen temporaries in the end but is likely to be optimized down to a handful depending on what the shader does. This is because the shader optimizer can easily track all variables and reuse registers. This is typically not possible for indexed temporaries, thus the register pressure of the shader may increase dramatically. This could be detrimental to performance as it reduces the hardware’s ability to hide latencies.

PCI Express^® 2.0 –Now you are ready for the most demanding graphics applications thanks to PCI Express^® 2.0 support, which allows up to twice the throughput of current PCI Express^® cards. Doubles the bus standard's bandwidth from 2.5 Gbit/s (PCIe 1.1) to 5 Gbit/sec.

Massively multi-threaded architecture supports thousands of independent simultaneous threads, providing extreme processing efficiency in advanced, next generation shader programs.

Advanced shader processors architected for physics computation enable a new level of physics effects to be simulated and rendered on the GPU--all while freeing the CPU to run the game engine and AI.