GeForce 8800 GTS (G80) 320MB vs GeForce 8800 GTS (G80) 640MB

Intro

Compare all that to the GeForce 8800 GTS (G80) 640MB, which has GPU core speed of 513 MHz, and 640 MB of GDDR3 memory running at 792 MHz through a 320-bit bus. It also features 96 SPUs, 48 Texture Address Units, and 20 ROPs.

F.E.A.R. 2

Settings:	Maximum Quality
AA:	4x
AF:	8x
Resolution:	1920x1200
Test Machine:	Unknown (Source)

GeForce 8800 GTS (G80) 640MB		30 FPS
GeForce 8800 GTS (G80) 320MB		28 FPS
	Difference: 2 FPS (7%)

Fallout 3

Settings:	Very High Quality
AA:	4x
AF:	8x
Resolution:	1680x1050
Test Machine:	Tom's Hardware Charts Test Rig (Source)

GeForce 8800 GTS (G80) 640MB		38 FPS
GeForce 8800 GTS (G80) 320MB		21 FPS
	Difference: 17 FPS (81%)

Far Cry 2

Settings:	Very High Qualty
AA:	none
AF:	none
Resolution:	1920x1200
Test Machine:	Intel Core i7-920,3 x 2 GB Ram,Windows Vista Ultimate 32 Bit SP1 (Source)

GeForce 8800 GTS (G80) 640MB		31 FPS
GeForce 8800 GTS (G80) 320MB		23 FPS
	Difference: 8 FPS (35%)

Left4Dead

Settings:	Very High Quality
AA:	4x
AF:	8x
Resolution:	1920x1200
Test Machine:	Tom's Hardware Charts Test Rig (Source)

GeForce 8800 GTS (G80) 640MB		38 FPS
GeForce 8800 GTS (G80) 320MB		35 FPS
	Difference: 3 FPS (9%)

Tom Clancy's Endwar

Settings:	High Quality
AA:	4x
AF:	8x
Resolution:	1920x1200
Test Machine:	Tom's Hardware Test Machine (Source)

GeForce 8800 GTS (G80) 640MB		14 FPS
GeForce 8800 GTS (G80) 320MB		8 FPS
	Difference: 6 FPS (75%)

---

GeForce 8800 GTS (G80) 640MB wins

(Based entirely on the benchmarks listed above)

When combining all game benchmark scores on this page together, the GeForce 8800 GTS (G80) 640MB wins overall, by 39 FPS. Please note that we do not have the results of every benchmark ever done for these cards, so the results may differ wildly in different games.

GeForce 8800 GTS (G80) 640MB		189 FPS
GeForce 8800 GTS (G80) 320MB		150 FPS
	Difference: 39 FPS (26%)

---

Power Usage and Theoretical Benchmarks

Both cards have the same power consumption.

Memory Bandwidth

Both cards have exactly the same bandwidth, so theoretically they should have the same performance. (explain)

Texel Rate

Both cards have the exact same texel fill rate, so in theory they should perform equally good at at anisotropic filtering. (explain)

Pixel Rate

Both cards have the exact same pixel fill rate, so theoretically they should be equally good at at anti-aliasing, and be capable of handling the same resolutions. (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.

Price Comparison

Specifications

Memory Bandwidth: Memory bandwidth is the max amount of data (measured in megabytes per second) that can be moved across the external memory interface within a second. It is worked out by multiplying the bus width by its memory clock speed. If it uses DDR type RAM, it must be multiplied by 2 again. If it uses DDR5, multiply by ANOTHER 2x. The higher the memory bandwidth, the better 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 applied in one second. This number is calculated by multiplying the total 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 per second.

Pixel Rate: Pixel rate is the maximum amount of pixels the video card can possibly write to its local memory in one second - measured in millions of pixels per second. The figure is worked out by multiplying the amount of ROPs by the clock speed of the card. ROPs (Raster Operations Pipelines - aka Render Output Units) are responsible for filling the screen with pixels (the image). The actual pixel output rate is also dependant on quite a few other factors, most notably the memory bandwidth - the lower the bandwidth is, the lower the ability to reach the maximum fill rate.