GeForce GT 640 DDR3 vs GeForce GTX 660

Intro

The GeForce GT 640 DDR3 makes use of a 28 nm design. nVidia has clocked the core frequency at 900 MHz. The DDR3 memory runs at a frequency of 1782 MHz on this model. It features 384 SPUs along with 32 TAUs and 16 ROPs.

Compare those specifications to the GeForce GTX 660, which uses a 28 nm design. nVidia has set the core frequency at 980 MHz. The GDDR5 memory works at a speed of 1502 MHz on this specific model. It features 960 SPUs along with 80 TAUs and 24 ROPs.

Price Comparison

Please note that the price comparisons are based on search keywords, and might not be the exact same card listed on this page. We have no control over the accuracy of their search results.

GeForce GT 640 DDR3 Amazon.com Other US-based stores Amazon.co.uk Amazon.de Amazon.fr

GeForce GTX 660 Amazon.com Other US-based stores Amazon.co.uk Amazon.de Amazon.fr

Specifications

Model	GeForce GT 640 DDR3	GeForce GTX 660
Manufacturer	nVidia	nVidia
Year	June 2012	September 2012
Code Name	GK107	GK106
Fab Process	28 nm	28 nm
Bus	PCIe 3.0 x16	PCIe 3.0 x16
Memory	2048 MB	2048 MB
Core Speed	900 MHz	980 MHz
Shader Speed	900 MHz	980 MHz
Memory Speed	1782 MHz (3564 MHz effective)	1502 MHz (6008 MHz effective)
Unified Shaders	384	960
Texture Mapping Units	32	80
Render Output Units	16	24
Bus Type	DDR3	GDDR5
Bus Width	128-bit	192-bit
DirectX Version	DirectX 11.1	DirectX 11.1
OpenGL Version	OpenGL 4.2	OpenGL 4.3
Power (Max TDP)	65 watts	140 watts
Shader Model	5.0	5.0
Bandwidth	57024 MB/sec	144192 MB/sec
Texel Rate	28800 Mtexels/sec	78400 Mtexels/sec
Pixel Rate	14400 Mpixels/sec	23520 Mpixels/sec

Memory Bandwidth: Bandwidth is the maximum amount of information (in units of MB per second) that can be transferred over the external memory interface in one second. The number is calculated by multiplying the bus width by the speed of its memory. If the card has DDR type RAM, it must be multiplied by 2 once again. If DDR5, multiply by ANOTHER 2x. The higher the card's memory bandwidth, the better the card will be in general. It especially helps with anti-aliasing, HDR and higher screen resolutions.

Texel Rate: Texel rate is the maximum amount of texture map elements (texels) that can be applied in one second. This figure is worked out by multiplying the total number of texture units of the card by the core clock speed of the chip. The better this number, the better the card will be at texture filtering (anisotropic filtering - AF). It is measured in millions of texels processed in one second.

Pixel Rate: Pixel rate is the maximum number of pixels the graphics card can possibly record to the local memory in a second - measured in millions of pixels per second. The number is worked out by multiplying the amount of Render Output Units by the the core speed of the card. ROPs (Raster Operations Pipelines - also called Render Output Units) are responsible for filling the screen with pixels (the image). The actual pixel rate also depends on many other factors, especially the memory bandwidth - the lower the memory bandwidth is, the lower the ability to reach the max fill rate.