Compare any two graphics cards:
GeForce 8800 GT 512MB vs Radeon HD 3870 512MB
IntroThe GeForce 8800 GT 512MB features a core clock speed of 600 MHz and a GDDR3 memory frequency of 900 MHz. It also uses a 256-bit memory bus, and uses a 65 nm design. It features 112 SPUs, 56 Texture Address Units, and 16 ROPs.
Compare all that to the Radeon HD 3870 512MB, which features GPU core speed of 775 MHz, and 512 MB of GDDR3 memory set to run at 900 MHz through a 256-bit bus. It also is made up of 320(64x5) Stream Processors, 16 Texture Address Units, and 16 ROPs.
Power Usage and Theoretical Benchmarks
Power Consumption (Max TDP)
Both cards have the exact same bandwidth, so in theory they should have identical performance. (explain)
Texel RateThe GeForce 8800 GT 512MB will be quite a bit (more or less 171%) better at AF than the Radeon HD 3870 512MB. (explain)
Pixel RateThe Radeon HD 3870 512MB is much (approximately 29%) better at FSAA than the GeForce 8800 GT 512MB, and will be capable of handling higher screen resolutions better. (explain)
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.
Memory Bandwidth: Bandwidth is the largest amount of information (counted in megabytes per second) that can be transferred over the external memory interface in one second. The number is calculated by multiplying the interface width by its memory clock speed. If it uses DDR type RAM, it should be multiplied by 2 once again. If DDR5, multiply by ANOTHER 2x. The better the memory bandwidth, the faster the card will be in general. It especially helps with AA, High Dynamic Range and higher screen resolutions.
Texel Rate: Texel rate is the maximum number of texture map elements (texels) that can be processed per second. This is worked out by multiplying the total number of texture units of the card by the core speed of the chip. The higher the texel rate, the better the graphics card will be at handling texture filtering (anisotropic filtering - AF). It is measured in millions of texels processed per second.
Pixel Rate: Pixel rate is the maximum number of pixels that the graphics card could possibly record to its local memory in a second - measured in millions of pixels per second. Pixel rate is worked out by multiplying the number of Render Output Units by the the card's clock speed. ROPs (Raster Operations Pipelines - also sometimes called Render Output Units) are responsible for filling the screen with pixels (the image). The actual pixel output rate also depends on lots of other factors, most notably the memory bandwidth of the card - the lower the bandwidth is, the lower the ability to reach the maximum fill rate.