Oops .. i am starting to think like everyone else here.


Graphics Processing Cluster

And how does that differ from an SMX?

The CUDA architecture is built around a scalable array of multithreaded Streaming Multiprocessors (SMs). When a CUDA program on the host CPU invokes a kernel grid, the blocks of the grid are enumerated and distributed to multiprocessors with available execution capacity. The threads of a thread block execute concurrently on one multiprocessor, and multiple thread blocks can execute concurrently on one multiprocessor. As thread blocks terminate, new blocks are launched on the vacated multiprocessors.
A multiprocessor is designed to execute hundreds of threads concurrently. To manage such a large amount of threads, it employs a unique architecture called SIMT (Single-Instruction, Multiple-Thread) that is described in Section 4.1. The instructions are pipelined to leverage instruction-level parallelism within a single thread, as well as thread-level parallelism extensively through simultaneous hardware multithreading as detailed in Section 4.2. Unlike CPU cores they are issued in order however and there is no branch prediction and no speculative execution. . . .

SIMT Architecture

The multiprocessor creates, manages, schedules, and executes threads in groups of 32 parallel threads called warps. Individual threads composing a warp start together at the same program address, but they have their own instruction address counter and register state and are therefore free to branch and execute independently. The term warp originates from weaving, the first parallel thread technology. A half-warp is either the first or second half of a warp. A quarter-warp is either the first, second, third, or fourth quarter of a warp.
When a multiprocessor is given one or more thread blocks to execute, it partitions them into warps and each warp gets scheduled by a warp scheduler for execution. The way a block is partitioned into warps is always the same; each warp contains threads of consecutive, increasing thread IDs with the first warp containing thread 0. Section 2.2 describes how thread IDs relate to thread indices in the block.

Kepler GK110 was built first and foremost for Tesla, and its goal was to be the highest performing parallel computing microprocessor in the world.

i would suggest that you wait for the Whitepaper. It will be available in about 17 hours at 10:30AM Pacific Time.

i have to agree that in a gaming GPU, the GTX 690 will probably be faster than the GTX 780 (if that is what is ends up being called). However, Nvidia made sweeping changes to Kepler to make it more efficient for programming.

They stopped with Fermi because of its inefficiency compared to Kepler. Clock speeds are also lower. It's going to be interesting. Frankly, i am very happy with my GTX 690 for gaming
.. and i want to go home and GAME .. this programming is reaching a near-overload for me. But i did learn a lot.

Who wouldn't be happy with a GTX 690?!? Especially the guy that got it without really paying for it!