Compare any two graphics cards:

Radeon HD 6970 vs Radeon HD 7970

Intro

The Radeon HD 6970 has core speeds of 880 MHz on the GPU, and 1375 MHz on the 2048 MB of GDDR5 memory. It features 1536 SPUs along with 96 Texture Address Units and 32 Rasterization Operator Units.

Compare that to the Radeon HD 7970, which features clock speeds of 925 MHz on the GPU, and 1375 MHz on the 3072 MB of GDDR5 RAM. It features 2048 SPUs as well as 128 TAUs and 32 ROPs.

(No game benchmarks for this combination yet.)

Power Usage and Theoretical Benchmarks

Both cards have the same power consumption.

Memory Bandwidth

The Radeon HD 7970, in theory, should be quite a bit faster than the Radeon HD 6970 in general. (explain)

Radeon HD 7970		264000 MB/sec
Radeon HD 6970		176000 MB/sec
	Difference: 88000 (50%)

Texel Rate

The Radeon HD 7970 will be much (about 40%) more effective at anisotropic filtering than the Radeon HD 6970. (explain)

Radeon HD 7970		118400 Mtexels/sec
Radeon HD 6970		84480 Mtexels/sec
	Difference: 33920 (40%)

Pixel Rate

If running with a high resolution is important to you, then the Radeon HD 7970 is a better choice, but only just. (explain)

Radeon HD 7970		29600 Mpixels/sec
Radeon HD 6970		28160 Mpixels/sec
	Difference: 1440 (5%)

Please note that the above 'benchmarks' are all just theoretical - the results were calculated based on the card's specifications, and real-world performance may (and probably will) vary at least a bit.

Price Comparison

Radeon HD 6970

Amazon.com

Radeon HD 7970

Amazon.com

Please note that the price comparisons are based on search keywords - sometimes it might show cards with very similar names that are not exactly the same as the one chosen in the comparison. We do try to filter out the wrong results as best we can, though.

Specifications

Model	Radeon HD 6970	Radeon HD 7970
Manufacturer	AMD	AMD
Year	December 2010	January 2012
Code Name	Cayman XT	Tahiti XT
Fab Process	40 nm	28 nm
Bus	PCIe x16	PCIe 3.0 x16
Memory	2048 MB	3072 MB
Core Speed	880 MHz	925 MHz
Shader Speed	N/A MHz	(N/A) MHz
Memory Speed	1375 MHz (5500 MHz effective)	1375 MHz (5500 MHz effective)
Unified Shaders	1536	2048
Texture Mapping Units	96	128
Render Output Units	32	32
Bus Type	GDDR5	GDDR5
Bus Width	256-bit	384-bit
DirectX Version	DirectX 11	DirectX 11.1
OpenGL Version	OpenGL 4.1	OpenGL 4.2
Power (Max TDP)	250 watts	250 watts
Shader Model	5.0	5.0
Bandwidth	176000 MB/sec	264000 MB/sec
Texel Rate	84480 Mtexels/sec	118400 Mtexels/sec
Pixel Rate	28160 Mpixels/sec	29600 Mpixels/sec

Memory Bandwidth: Bandwidth is the largest amount of information (measured in MB per second) that can be transported across the external memory interface in a second. It's calculated by multiplying the bus width by its memory clock speed. If it uses DDR memory, 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, High Dynamic Range and higher screen resolutions.

Texel Rate: Texel rate is the maximum amount of texture map elements (texels) that are processed in one second. This figure is worked out by multiplying the total amount 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 handling texture filtering (anisotropic filtering - AF). It is measured in millions of texels applied per second.

Pixel Rate: Pixel rate is the most pixels that the graphics card could possibly record to the local memory per second - measured in millions of pixels per second. The number is worked out by multiplying the amount of ROPs by the clock speed of the card. ROPs (Raster Operations Pipelines - aka Render Output Units) are responsible for outputting the pixels (image) to the screen. The actual pixel output rate is also dependant on many other factors, especially the memory bandwidth - the lower the bandwidth is, the lower the potential to get to the maximum fill rate.