Compare any two graphics cards:
GeForce 8600 GT 512MB GDDR3 vs GeForce 9500 GT 1GB GDDR3
IntroThe GeForce 8600 GT 512MB GDDR3 comes with core speeds of 540 MHz on the GPU, and 700 MHz on the 512 MB of GDDR3 memory. It features 32 SPUs as well as 16 TAUs and 8 Rasterization Operator Units.
Compare all that to the GeForce 9500 GT 1GB GDDR3, which uses a 55 nm design. nVidia has set the core speed at 550 MHz. The GDDR3 RAM runs at a frequency of 800 MHz on this specific card. It features 32 SPUs along with 16 TAUs and 8 ROPs.
Power Usage and Theoretical Benchmarks
Power Consumption (Max TDP)
In theory, the GeForce 9500 GT 1GB GDDR3 should be 14% faster than the GeForce 8600 GT 512MB GDDR3 overall, due to its higher data rate. (explain)
Texel RateThe GeForce 9500 GT 1GB GDDR3 will be just a bit (approximately 2%) more effective at texture filtering than the GeForce 8600 GT 512MB GDDR3. (explain)
Pixel RateThe GeForce 9500 GT 1GB GDDR3 should be a small bit (more or less 2%) better at FSAA than the GeForce 8600 GT 512MB GDDR3, and should be able to handle higher resolutions without slowing down too much. (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 (measured in megabytes per second) that can be moved across the external memory interface within a second. The number is calculated by multiplying the interface width by the speed of its memory. If it uses DDR type RAM, it must be multiplied by 2 again. If DDR5, multiply by ANOTHER 2x. The better the card's 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 can be processed per second. This number is calculated by multiplying the total number of texture units of the card by the core clock speed of the chip. The higher this number, the better the graphics card will be at texture filtering (anisotropic filtering - AF). It is measured in millions of texels processed in a second.
Pixel Rate: Pixel rate is the most pixels that the graphics card can possibly record to the local memory in one second - measured in millions of pixels per second. The number is calculated by multiplying the number of Raster Operations Pipelines by the the core speed of the card. ROPs (Raster Operations Pipelines - aka Render Output Units) are responsible for drawing the pixels (image) on the screen. The actual pixel output rate is also dependant on lots of other factors, especially the memory bandwidth - the lower the bandwidth is, the lower the ability to reach the maximum fill rate.