GeForce GTX 660 vs Radeon HD 4890 1GB

Intro

Compare those specifications to the Radeon HD 4890 1GB, which makes use of a 55 nm design. AMD has clocked the core frequency at 1000 MHz. The GDDR5 memory works at a frequency of 975 MHz on this particular card. It features 800(160x5) SPUs as well as 40 Texture Address Units and 16 ROPs.

Display Graphs

Hide Graphs

Power Usage and Theoretical Benchmarks

Power Consumption (Max TDP)

GeForce GTX 660		140 Watts
Radeon HD 4890 1GB		190 Watts
	Difference: 50 Watts (36%)

Memory Bandwidth

The GeForce GTX 660 should theoretically perform a little bit faster than the Radeon HD 4890 1GB overall. (explain)

GeForce GTX 660		144192 MB/sec
Radeon HD 4890 1GB		124800 MB/sec
	Difference: 19392 (16%)

Texel Rate

The GeForce GTX 660 is much (about 96%) better at anisotropic filtering than the Radeon HD 4890 1GB. (explain)

GeForce GTX 660		78400 Mtexels/sec
Radeon HD 4890 1GB		40000 Mtexels/sec
	Difference: 38400 (96%)

Pixel Rate

The GeForce GTX 660 is much (approximately 47%) faster with regards to full screen anti-aliasing than the Radeon HD 4890 1GB, and also will be capable of handling higher resolutions better. (explain)

GeForce GTX 660		23520 Mpixels/sec
Radeon HD 4890 1GB		16000 Mpixels/sec
	Difference: 7520 (47%)

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 largest amount of data (counted in MB per second) that can be transported across the external memory interface in a second. It's calculated by multiplying the card's interface width by the speed of its memory. If it uses DDR type memory, it must be multiplied by 2 once again. If DDR5, multiply by ANOTHER 2x. The better the bandwidth is, the better the card will be in general. It especially helps with anti-aliasing, HDR and high resolutions.

Texel Rate: Texel rate is the maximum texture map elements (texels) that can be applied per second. This figure is worked out by multiplying the total texture units of the card by the core clock speed of the chip. The better this number, the better the video card will be at handling texture filtering (anisotropic filtering - AF). It is measured in millions of texels applied in one second.

Pixel Rate: Pixel rate is the maximum number of pixels that the graphics chip could 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 colour ROPs by the the core 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 rate is also dependant on quite a few other factors, especially the memory bandwidth - the lower the bandwidth is, the lower the potential to get to the maximum fill rate.