We'll now look at the reference (original design) based specs and architecture. The GeForce GTX 670 is based on the new Kepler GPU architecture. It is based on the very same 28nm GK104 GPU which is used on the GeForce GTX 680.
The GeForce GTX 670 boasts 1344 CUDA (shader) cores whereas the GeForce GTX 680 has 1536 CUDA (shader) cores. That's 192 Shader cores less, and that's precisely one CUDA core clusters (SM) less out of the eight available.
The product is obviously PCI-Express 3.0 ready and has a TDP of around 170 Watt (with a typical draw of 150~160W). But let me first show you GK104 die:
NVIDIA GK104 Kepler architecture GPU, you can see the eight SM (CUDA/shader core) clusters, one of these has been deactivated for the GTX 670.
An immediate difference to the GPU core versus the shader processor domain is that both will be clocked at 1:1, meaning both the core and shader domain clock in at 915 MHz. The boost clock for the reference GTX 670 cards is set at 980 MHz though that can vary a bit per card and available power envelope (topping 1 GHz would not surprise me).
As far as the memory specs of the GK104 Kepler GPU are concerned, the boards will feature a 256-bit memory bus connected to 2 GB of GDDR5 video buffer memory. On the memory controller side of things you'll see very significant improvements as the reference memory clock is set at 6 GHz / Gbps. This boils down to to a memory bandwidth of 192 GB/s on that 256-bit memory bus.
With this release, NVIDIA now has the third product in the series 600 cards on its way. The new graphics adapters are of course DirectX 11.1 ready. With Windows 8, 7 and Vista also being DX11 ready all we need are more new games to take advantage of DirectCompute, multi-threading, hardware tessellation and the latest shader 5.0 extensions.
For your reference here's a quick overview of some past generation high-end GeForce cards opposed to the new Kepler based GeForce GTX 680.
GeForce GTX 480
GeForce GTX 580
GeForce GTX 670
ASUS GTX 670 DCUII TOP
GeForce GTX 680
GeForce GTX 690
Stream (Shader) Processors
Core Clock (MHz)
Shader Clock (MHz)
Boost clock (Mhz)
Memory Clock (effective MHz)
For Kepler, NVIDIA kept their memory controllers GDDR5 compatible. Memory wise NVIDIA has nice large memory volumes due to their architecture, we pass 2 GB as standard these days for most of NVIDIA's series 600 graphics cards in the high range spectrum.
The hardware engineers of NVIDIA reworked the memory subs system quite a bit, enabling much higher memory clock frequency speeds compared to previous generation GeForce GPUs. The result is this memory speeds up-to 6 Gbps. Each memory partition utilizes one memory controller on the respective GPU, which will get 256/512 MB of memory tied to it.
The GTX 580 has six memory controllers (6x256MB) = 1536 MB of GDDR5 memory
The GTX 670 has four memory controllers (4x512MB) = 2048 MB of GDDR5 memory
The GTX 680 has four memory controllers (4x512MB) = 2048 MB of GDDR5 memory
As mentioned in the introduction, a 4 GB version would be very possible as well. It all depends on the board partners.
ASUS GeForce GTX 960 Strix 2-way SLI review We review the ASUS Strix edition of the GeForce GTX 960. Actually not one, but two of them as we go SLI all the way as well. This premium version of the GTX 960 comes with an all silent design, good l...
ASUS GeForce GTX 980 Poseidon review We review the ASUS ROG Poseidon GTX 980. Armed with a massive liquid cooling block (DirectCU H2O) , a custom PCB, quality components and a factory overclock this product is bound to perform, whilst stay...
ASUS GeForce GTX 980 Strix review Hot on the heels of the ASUS GTX 970 Strix review, we test the STRIX-GTX980-DC2OC-4GD5 - ASUS GeForce GTX 980 Strix. Armed with that h00t of a cooler, a custom PCB, quality components and a factory overclock this product is...
ASUS GeForce GTX 970 Strix review We review the new ASUS GeForce GTX 970 Strix edition. A product that has been overhauled in terms of PCB and cooling. ASUS also clocks the product a little faster then reference. The new DirectCU II b...