GeForce RTX 3090 vs Radeon Pro Duo

Intro

Compare all that to the Radeon Pro Duo, which makes use of a 28 nm design. AMD has clocked the core speed at 1000 MHz. The HBM RAM is set to run at a frequency of 500 MHz on this particular model. It features 4096 SPUs as well as 256 TAUs and 64 ROPs.

Display Graphs

Hide Graphs

Power Usage and Theoretical Benchmarks

Both cards have the same power consumption.

Memory Bandwidth

Theoretically speaking, the Radeon Pro Duo is 7% quicker than the GeForce RTX 3090 in general, due to its greater bandwidth. (explain)

Radeon Pro Duo		1024000 MB/sec
GeForce RTX 3090		958668 MB/sec
	Difference: 65332 (7%)

Texel Rate

The Radeon Pro Duo should be a small bit (about 12%) more effective at AF than the GeForce RTX 3090. (explain)

Radeon Pro Duo		512000 Mtexels/sec
GeForce RTX 3090		457560 Mtexels/sec
	Difference: 54440 (12%)

Pixel Rate

The GeForce RTX 3090 is a lot (more or less 22%) better at AA than the Radeon Pro Duo, and will be able to handle higher screen resolutions better. (explain)

GeForce RTX 3090		156240 Mpixels/sec
Radeon Pro Duo		128000 Mpixels/sec
	Difference: 28240 (22%)

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.

One or more cards in this comparison are multi-core. This means that their bandwidth, texel and pixel rates are theoretically doubled - this does not mean the card will actually perform twice as fast, but only that it should in theory be able to. Actual game benchmarks will give a more accurate idea of what it's capable of.

Price Comparison

Specifications

Memory Bandwidth: Bandwidth is the max amount of data (measured in MB per second) that can be transferred over the external memory interface in a second. It's worked out by multiplying the bus width by its memory speed. If it uses DDR type memory, it must be multiplied by 2 again. If DDR5, multiply by ANOTHER 2x. The higher the card's memory bandwidth, the faster 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 per second. This figure is calculated by multiplying the total texture units of the card by the core clock speed of the chip. The better the texel rate, the better the card will be at handling texture filtering (anisotropic filtering - AF). It is measured in millions of texels processed in one second.

Pixel Rate: Pixel rate is the maximum number of pixels that the graphics chip could possibly write to the local memory per second - measured in millions of pixels per second. The figure is worked out by multiplying the amount of Raster Operations Pipelines by the the core clock speed. ROPs (Raster Operations Pipelines - also called Render Output Units) are responsible for filling the screen with pixels (the image). The actual pixel rate is also dependant on many other factors, most notably the memory bandwidth - the lower the memory bandwidth is, the lower the potential to reach the max fill rate.