NVIDIA GeForce GTX 690 Launched

NVIDIA today announced the GeForce GTX 690, the world's fastest consumer graphics card -- with a bold industrial design to match. Powered by dual Kepler architecture-based GeForce GPUs, the GTX 690 is meticulously designed -- inside and out -- to deliver the most refined, elegant and smooth PC gaming experience possible.

The surprise announcement was made by NVIDIA CEO and co-founder Jen-Hsun Huang during his keynote address at the NVIDIA Game Festival in Shanghai, which is being attended by more than 6,000 gamers from across China.

Nvidia GeForce GTX 690 Specifications:

Graphics Card: GeForce GTX 690 4GB

Graphics Processing Clusters: 8

Streaming Multiprocessors: 16

CUDA Cores: 3072

Texture Units: 256

ROP Units: 64

Graphics Clock: 915 MHz (1019 Boost)

Memory Clock (Data rate): 6008 MHz

Total Video Memory: 4096MB GDDR5

Memory Interface: 512‐bit

Total Memory Bandwidth: 192 GB/s per GPU

Fabrication Process: 28 nm

Transistor Count: 7 Billion total

Connectors: 3x Dual‐Link DVI, 1x Mini-DisplayPort 1.2

Form Factor: Dual Slot

Power Connectors: 2x 8‐pin PCI-E Power

Recommended Power Supply: 600 Watts

Thermal Design Power (TDP): 300 Watts

source

NVIDIA GeForce GTX 670

Here is the first picture of an NVIDIA reference design GeForce GTX 670. For the most part, its design will resemble that of reference design GTX 680. The card pictured below is said to have had Leadtek branding, which later got removed. GeForce GTX 670 is carved out of 28 nm GK104 silicon by disabling one of the card's eight SMX units, resulting in a CUDA core count of 1344. Its clock speeds are lower than those of the GTX 680, with around 950 MHz core base clock, and core boost clock a little over 1 GHz. It features 2 GB of memory across a 256-bit wide GDDR5 memory interface. The latest price speculation puts it at US $399-$429.

Based on the Nvidia GK104 (3.54 billion transistors on 294mm ² 28mn chip area)

DirectX 11.1, Kepler architecture

4 Raster Engines (GPC)

7 shader clusters (SMX)

1344 (1D) shader units

112 texture units (TMUs)

32 raster operation units (ROPs)

256-bit DDR memory interface

2 GB of GDDR5 memory configuration

Chip clock: 900 MHz or something more

Memory clock: unknown, assumed ~ 2500 MHz

PCI Express 3.0 (backwards compatible with PCI Express 1.x and 2.0)

Estimated TDP ~ 160 watts

source