Compare any two graphics cards:
GeForce 8800 GTS (G80) 320MB vs Radeon HD 4730
IntroThe GeForce 8800 GTS (G80) 320MB makes use of a 90 nm design. nVidia has clocked the core speed at 513 MHz. The GDDR3 memory runs at a frequency of 792 MHz on this particular card. It features 96 SPUs as well as 48 Texture Address Units and 20 ROPs.
Compare those specifications to the Radeon HD 4730, which features a core clock frequency of 700 MHz and a GDDR5 memory speed of 900 MHz. It also uses a 128-bit bus, and uses a 55 nm design. It features 640(128x5) SPUs, 32 TAUs, and 8 Raster Operation Units.
Power Usage and Theoretical Benchmarks
Power Consumption (Max TDP)
In theory, the GeForce 8800 GTS (G80) 320MB should perform a small bit faster than the Radeon HD 4730 in general. (explain)
Texel RateThe GeForce 8800 GTS (G80) 320MB will be a little bit (more or less 10%) better at AF than the Radeon HD 4730. (explain)
Pixel RateIf running with high levels of AA is important to you, then the GeForce 8800 GTS (G80) 320MB is superior to the Radeon HD 4730, by far. (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 maximum amount of data (in units of megabytes per second) that can be transported past the external memory interface in one second. It is worked out by multiplying the bus width by the speed of its memory. In the case of DDR type memory, it should be multiplied by 2 once again. If DDR5, multiply by 4 instead. 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 high resolutions.
Texel Rate: Texel rate is the maximum number of texture map elements (texels) that are applied per second. This number is calculated by multiplying the total amount of texture units by the core speed of the chip. The better the texel rate, the better the graphics card will be at texture filtering (anisotropic filtering - AF). It is measured in millions of texels in one second.
Pixel Rate: Pixel rate is the most pixels that the graphics card can possibly write to its local memory in one second - measured in millions of pixels per second. The number is calculated by multiplying the number of Render Output Units 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 also depends on quite a few other factors, most notably the memory bandwidth of the card - the lower the bandwidth is, the lower the potential to reach the maximum fill rate.