Geforce GTX 670 vs Radeon HD 4870 1GB

Intro

Compare those specs to the Radeon HD 4870 1GB, which comes with core speeds of 750 MHz on the GPU, and 900 MHz on the 1024 MB of GDDR5 RAM. It features 800(160x5) SPUs along with 40 TAUs and 16 Rasterization Operator Units.

Display Graphs

Hide Graphs

Power Usage and Theoretical Benchmarks

Power Consumption (Max TDP)

Radeon HD 4870 1GB		150 Watts
Geforce GTX 670		170 Watts
	Difference: 20 Watts (13%)

Memory Bandwidth

As far as performance goes, the Geforce GTX 670 should in theory be quite a bit better than the Radeon HD 4870 1GB overall. (explain)

Geforce GTX 670		192000 MB/sec
Radeon HD 4870 1GB		115200 MB/sec
	Difference: 76800 (67%)

Texel Rate

The Geforce GTX 670 is a lot (about 242%) better at AF than the Radeon HD 4870 1GB. (explain)

Geforce GTX 670		102480 Mtexels/sec
Radeon HD 4870 1GB		30000 Mtexels/sec
	Difference: 72480 (242%)

Pixel Rate

The Geforce GTX 670 should be a lot (more or less 144%) faster with regards to AA than the Radeon HD 4870 1GB, and also should be capable of handling higher resolutions while still performing well. (explain)

Geforce GTX 670		29280 Mpixels/sec
Radeon HD 4870 1GB		12000 Mpixels/sec
	Difference: 17280 (144%)

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 (measured in MB per second) that can be moved past the external memory interface within a second. It is calculated by multiplying the interface width by its memory clock speed. If it uses DDR type memory, it must be multiplied by 2 again. If DDR5, multiply by ANOTHER 2x. The higher the bandwidth is, the faster the card will be in general. It especially helps with anti-aliasing, HDR and higher screen resolutions.

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

Pixel Rate: Pixel rate is the most pixels the video card could possibly record to its local memory in one second - measured in millions of pixels per second. The number is calculated by multiplying the amount of Raster Operations Pipelines by the clock speed of the card. ROPs (Raster Operations Pipelines - also sometimes called Render Output Units) are responsible for filling the screen with pixels (the image). The actual pixel output rate also depends on quite a few other factors, most notably the memory bandwidth - the lower the memory bandwidth is, the lower the ability to get to the max fill rate.