Here is the White Paper for NVidia GTX680. GeForce-GTX-680-Whitepaper-FINAL.pdf.

From the White Paper, " Prior to the introduction of Kepler, video encoding on previous GeForce products was handled by encode software running on the GPU's array of CUDA Cores. While the CUDA Cores were able to deliver tremendous performance speedups compared to CPU-based encoding, one downside of using these high-speed processor cores to process video encoding was increased power consumption. By using specialized circuitry for H.264 encoding, the NVENC hardware encoder in Kepler is almost four times faster than our previous CUDA-based encoder while consuming much less power."

As of April 25, 2012, the price tags are:

• GTX680 - $499
• GTX680 SLI (Scalable Link Interface) - $999

GeForce GTX 680 includes NVENC which is able to take a number of input codecs and decode, preprocess, and encode H.264-based content. It supports H.264 Base, Main, and High Profile Level 4.1. It should not to be confused with CUDA, which is software based encoding accelerated by GPU cores, NVENC is actual hardware based video encoding.

Actually, NVENC a dedicated graphics encoder engine that incorporates a new hardware based H.264 video encoder inside the GPU. In the past this functionality was managed over the shader processor cores yet with the introduction of Kepler some extra core logic has been added dedicated to this function.

AMD Radeon HD 7970 vs. Nvidia GeForce GTX 680.

Today Nvidia is pulling the wraps off the GK110-based GeForce GTX Titan, a single-GPU card that is expected to easily capture the title of Baddest Ass GPU in the world when benchmarks are released this Thursday, February 21st. The Titan is Nvidia's "Big Kepler" GPU, and has double the transistors and almost double the CUDA cores of the mid-range GK104 chip found in its flagship GeForce GTX 680 GPU. Though it runs at a lower clock speed in stock trim, it should still offer a sizable performance improvement over the already capable GTX 680.

The GK110 has been in use for over a year in the super-computing world, most notably in the Tesla K20X GPUs used by the Oak Ridge National Laboratory's Titan supercomputer, which is currently the fastest super computer in the world. Though benchmarks for the Titan haven't been posted just yet, Nvidia tells us it should be neck-and-neck with the dual-GPU GeForce GTX 690, which is currently the fastest single card GPU in existence. In other words, the Titan should be almost fast as dual GTX 680s in SLI, but from a single GPU. Oh, and guess what? It's actually quieter than both of them, even under load (yes, we've heard it).

For more, visit Nvidia GeForce GTX Titan: A Massive GPU That Might Be Unbeatable - from GigaOM, Feb 19, 2013.

NVEnc is currently exposed through a proprietary API, though Nvidia does have plans to provide access to NVEnc through CUDA.

CUDA downloads, http://developer.nvidia.com/cuda-downloads.