Gainward GeForce RTX 3090 Phantom vs RTX 3090 Phantom “GS”

Since both graphics cards have the same number of ROPs and the GeForce RTX 3090 Phantom “GS” runs at higher rate, its pixel fillrate is higher. Having better pixel fillrate allows the graphics card to draw more pixels on screen and off screen, which is beneficial for some 3D effects in games, or when playing at higher display resolutions.

Another important parameter, that affects overall GPU performance, is texture fillrate. This parameter is calculated as the number of TMUs (Texture Mapping Units), multiplied by graphics frequency. Because two graphics cards have the same number of Texture Mapping Units (TMUs) and the Gainward GeForce RTX 3090 Phantom “GS” card operates at higher rate, its texture fillrate is higher. Having better texture fill rate allows the graphics card to utilize more sophisticated 3D effects and/or map more textures to each textured picture element, which improves games visual appearance.

Maximum Single Precision performance, measured in GFLOPS or billions of Floating Point Operations Per Second, shows how fast the graphics card is at executing programs, that process primarily single-precision floating point data. As a rule, the more CUDA cores or stream processors the GPU has, and the the faster they operate at, the higher Single Precision performance will be. The GeForce RTX 3090 Phantom “GS” has a small lead here. Higher single-precision performance number means the graphics unit will perform better in general computing applications. Since CUDA cores or stream processors are also used as vertex and geometry shaders for 3D image generation, higher performance is also beneficial to games.

Maximum Double Precision performance is similar to the Single Precision performance, except that it applies to double-precision (64-bit) floating point operations. This characteristic is irrelevant to games performance because games do not use double-precision arithmetics. The Gainward GeForce RTX 3090 Phantom “GS” graphics card is faster at processing 64-bit floating-point numbers.

Memory bandwidth of both GPUs is identical.