Compare any two graphics cards:
GeForce 8800 GTS (G80) 320MB vs GeForce GTS 450
IntroThe GeForce 8800 GTS (G80) 320MB comes with a core clock frequency of 513 MHz and a GDDR3 memory frequency of 792 MHz. It also uses a 320-bit bus, and uses a 90 nm design. It is comprised of 96 SPUs, 48 Texture Address Units, and 20 Raster Operation Units.
Compare that to the GeForce GTS 450, which has a core clock frequency of 783 MHz and a GDDR5 memory frequency of 902 MHz. It also features a 128-bit memory bus, and uses a 40 nm design. It features 192 SPUs, 32 Texture Address Units, and 16 Raster Operation Units.
Power Usage and Theoretical Benchmarks
Power Consumption (Max TDP)
As far as performance goes, the GeForce 8800 GTS (G80) 320MB should theoretically be a bit better than the GeForce GTS 450 overall. (explain)
Texel RateThe GeForce GTS 450 will be a bit (more or less 2%) better at anisotropic filtering than the GeForce 8800 GTS (G80) 320MB. (explain)
Pixel RateThe GeForce GTS 450 should be quite a bit (more or less 22%) faster with regards to anti-aliasing than the GeForce 8800 GTS (G80) 320MB, and able to handle higher screen resolutions more effectively. (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: Memory bandwidth is the largest amount of information (in units of MB per second) that can be moved over the external memory interface within a second. It's worked out by multiplying the bus width by the speed of its memory. If the card has DDR type memory, it must be multiplied by 2 once again. If DDR5, multiply by 4 instead. The better the memory bandwidth, the better the card will be in general. It especially helps with AA, HDR and high resolutions.
Texel Rate: Texel rate is the maximum texture map elements (texels) that are applied per 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 higher this number, the better the video card will be at texture filtering (anisotropic filtering - AF). It is measured in millions of texels applied per second.
Pixel Rate: Pixel rate is the maximum number of pixels the video card can 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 Raster Operations Pipelines by the the core clock speed. ROPs (Raster Operations Pipelines - also called Render Output Units) are responsible for filling the screen with pixels (the image). The actual pixel output rate also depends on many other factors, especially the memory bandwidth of the card - the lower the memory bandwidth is, the lower the ability to reach the maximum fill rate.