GeForce RTX 3070 vs Radeon RX 6900 XT

Intro

Compare those specifications to the Radeon RX 6900 XT, which has a clock speed of 1825 MHz and a GDDR6 memory speed of 2000 MHz. It also uses a 256-bit bus, and uses a 7 nm design. It features 5120 SPUs, 320 TAUs, and 128 Raster Operation Units.

Display Graphs

Hide Graphs

Power Usage and Theoretical Benchmarks

Power Consumption (Max TDP)

GeForce RTX 3070		220 Watts
Radeon RX 6900 XT		300 Watts
	Difference: 80 Watts (36%)

Memory Bandwidth

Theoretically speaking, the Radeon RX 6900 XT should be -100% quicker than the GeForce RTX 3070 overall, because of its higher bandwidth. (explain)

Radeon RX 6900 XT		524288 MB/sec
GeForce RTX 3070		(Unknown) MB/sec
	Difference: 524288 (-100%)

Texel Rate

The Radeon RX 6900 XT is quite a bit (more or less 112%) more effective at anisotropic filtering than the GeForce RTX 3070. (explain)

Radeon RX 6900 XT		584000 Mtexels/sec
GeForce RTX 3070		276000 Mtexels/sec
	Difference: 308000 (112%)

Pixel Rate

If using a high screen resolution is important to you, then the Radeon RX 6900 XT is the winner, by a large margin. (explain)

Radeon RX 6900 XT		233600 Mpixels/sec
GeForce RTX 3070		144000 Mpixels/sec
	Difference: 89600 (62%)

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 max amount of data (in units of megabytes per second) that can be transferred past the external memory interface within a second. The number is calculated by multiplying the card's interface width by the speed of its memory. If it uses DDR type memory, it should be multiplied by 2 once again. If DDR5, multiply by ANOTHER 2x. The higher 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 processed in one second. This figure 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 processed in a second.

Pixel Rate: Pixel rate is the most pixels the graphics card can possibly write to the local memory per second - measured in millions of pixels per second. Pixel rate is worked out by multiplying the amount of Render Output Units by the the core speed of the card. ROPs (Raster Operations Pipelines - sometimes also referred to as Render Output Units) are responsible for outputting the pixels (image) to the screen. The actual pixel output rate is also dependant on many other factors, especially the memory bandwidth of the card - the lower the bandwidth is, the lower the ability to get to the maximum fill rate.