Compare any two graphics cards:

GeForce 8800 GTS (G80) 640MB vs GeForce 9800 GTX

Intro

The GeForce 8800 GTS (G80) 640MB uses a 90 nm design. nVidia has clocked the core frequency at 513 MHz. The GDDR3 RAM runs at a frequency of 792 MHz on this specific model. It features 96 SPUs along with 48 Texture Address Units and 20 ROPs.

Compare all of that to the GeForce 9800 GTX, which has a clock frequency of 675 MHz and a GDDR3 memory frequency of 1100 MHz. It also makes use of a 256-bit memory bus, and makes use of a 65 nm design. It is made up of 128 SPUs, 64 Texture Address Units, and 16 Raster Operation Units.

Display Graphs

Hide Graphs

Power Usage and Theoretical Benchmarks

Power Consumption (Max TDP)

GeForce 9800 GTX		140 Watts
GeForce 8800 GTS (G80) 640MB		143 Watts
	Difference: 3 Watts (2%)

Memory Bandwidth

As far as performance goes, the GeForce 9800 GTX should theoretically be a bit better than the GeForce 8800 GTS (G80) 640MB in general. (explain)

GeForce 9800 GTX		70400 MB/sec
GeForce 8800 GTS (G80) 640MB		63360 MB/sec
	Difference: 7040 (11%)

Texel Rate

The GeForce 9800 GTX should be a lot (approximately 75%) better at texture filtering than the GeForce 8800 GTS (G80) 640MB. (explain)

GeForce 9800 GTX		43200 Mtexels/sec
GeForce 8800 GTS (G80) 640MB		24624 Mtexels/sec
	Difference: 18576 (75%)

Pixel Rate

The GeForce 9800 GTX will be a little bit (more or less 5%) faster with regards to full screen anti-aliasing than the GeForce 8800 GTS (G80) 640MB, and also should be capable of handling higher screen resolutions while still performing well. (explain)

GeForce 9800 GTX		10800 Mpixels/sec
GeForce 8800 GTS (G80) 640MB		10260 Mpixels/sec
	Difference: 540 (5%)

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

Display Prices

Hide Prices

GeForce 8800 GTS (G80) 640MB

Amazon.com

Check prices at:

GeForce 9800 GTX

Amazon.com

Check prices at:

Please note that the price comparisons are based on search keywords - sometimes it might show cards with very similar names that are not exactly the same as the one chosen in the comparison. We do try to filter out the wrong results as best we can, though.

Specifications

Display Specifications

Hide Specifications

Model	GeForce 8800 GTS (G80) 640MB	GeForce 9800 GTX
Manufacturer	nVidia	nVidia
Year	Nov 2006 (640)	April 2008
Code Name	G80	G92
Memory	640 MB	512 MB
Core Speed	513 MHz	675 MHz
Memory Speed	1584 MHz	2200 MHz
Power (Max TDP)	143 watts	140 watts
Bandwidth	63360 MB/sec	70400 MB/sec
Texel Rate	24624 Mtexels/sec	43200 Mtexels/sec
Pixel Rate	10260 Mpixels/sec	10800 Mpixels/sec
Unified Shaders	96	128
Texture Mapping Units	48	64
Render Output Units	20	16
Bus Type	GDDR3	GDDR3
Bus Width	320-bit	256-bit
Fab Process	90 nm	65 nm
Transistors	681 million	754 million
Bus	PCIe x16	PCIe x16 2.0
DirectX Version	DirectX 10	DirectX 10
OpenGL Version	OpenGL 3.0	OpenGL 3.0

Memory Bandwidth: Memory bandwidth is the largest amount of information (in units of megabytes per second) that can be transported over the external memory interface in one second. It is worked out by multiplying the bus width by its memory clock speed. If it uses DDR RAM, it must be multiplied by 2 once again. If DDR5, multiply by 4 instead. The better the bandwidth is, the faster the card will be in general. It especially helps with anti-aliasing, High Dynamic Range and higher screen resolutions.

Texel Rate: Texel rate is the maximum texture map elements (texels) that can be processed in one second. This is calculated by multiplying the total number of texture units by the core speed of the chip. The better the texel rate, the better the video 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 maximum number of pixels that the graphics chip 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 Raster Operations Pipelines by the clock speed of the card. ROPs (Raster Operations Pipelines - sometimes also referred to as Render Output Units) are responsible for filling the screen with pixels (the image). The actual pixel output rate is also dependant on many other factors, most notably the memory bandwidth - the lower the memory bandwidth is, the lower the ability to reach the maximum fill rate.

Display Prices

Hide Prices