GeForce 8800 GTS (G80) 640MB vs Radeon HD 3850 512MB

Intro

Compare those specifications to the Radeon HD 3850 512MB, which comes with GPU clock speed of 668 MHz, and 512 MB of GDDR3 memory set to run at 828 MHz through a 256-bit bus. It also features 320(64x5) Stream Processors, 16 TAUs, and 16 ROPs.

(No game benchmarks for this combination yet.)

Power Usage and Theoretical Benchmarks

Power Consumption (Max TDP)

Radeon HD 3850 512MB		75 Watts
GeForce 8800 GTS (G80) 640MB		143 Watts
	Difference: 68 Watts (91%)

Memory Bandwidth

Theoretically speaking, the GeForce 8800 GTS (G80) 640MB is 20% faster than the Radeon HD 3850 512MB overall, due to its greater bandwidth. (explain)

GeForce 8800 GTS (G80) 640MB		63360 MB/sec
Radeon HD 3850 512MB		52992 MB/sec
	Difference: 10368 (20%)

Texel Rate

The GeForce 8800 GTS (G80) 640MB is quite a bit (more or less 130%) better at anisotropic filtering than the Radeon HD 3850 512MB. (explain)

GeForce 8800 GTS (G80) 640MB		24624 Mtexels/sec
Radeon HD 3850 512MB		10688 Mtexels/sec
	Difference: 13936 (130%)

Pixel Rate

If using high levels of AA is important to you, then the Radeon HD 3850 512MB is a better choice, but only just. (explain)

Radeon HD 3850 512MB		10688 Mpixels/sec
GeForce 8800 GTS (G80) 640MB		10260 Mpixels/sec
	Difference: 428 (4%)

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

Specifications

Memory Bandwidth: Bandwidth is the maximum amount of information (in units of MB per second) that can be transported over the external memory interface within a second. It's calculated by multiplying the card's interface width by its memory clock speed. In the case of DDR type RAM, the result should be multiplied by 2 again. If it uses DDR5, multiply by 4 instead. The higher 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 worked out by multiplying the total number of texture units of the card by the core speed of the chip. The higher this number, the better the graphics card will be at handling texture filtering (anisotropic filtering - AF). It is measured in millions of texels in a second.

Pixel Rate: Pixel rate is the most pixels that the graphics chip could possibly write to its local memory in one second - measured in millions of pixels per second. The figure is calculated by multiplying the amount of ROPs by the the core speed of the card. ROPs (Raster Operations Pipelines - also sometimes called Render Output Units) are responsible for outputting the pixels (image) to the screen. The actual pixel output rate also depends on many other factors, most notably the memory bandwidth of the card - the lower the bandwidth is, the lower the ability to get to the max fill rate.