NVIDIA GeForce RTX 3090

The NVIDIA GeForce RTX 3090 is a desktop graphics card that launched in Q3 2020 with a MSRP of $1,499. It is built on the Ampere GPU microarchitecture (codename GA102) and is manufactured on a 8 nm process.

Memory

The RTX 3090 has 24 GB of GDDR6X memory, with a 1,219 MHz memory clock and a 384 bit interface. This gives it a memory bandwidth of 936.2 Gb/s, which affects how fast it can transfer data to and from memory. GPU memory stores temporary data that helps the GPU with complex math and graphics operations. More memory is generally better, as not having enough can cause performance bottlenecks.

Cores and Clock Speeds

The RTX 3090 includes 10,496 CUDA cores, the processing units for handling parallel computing tasks. The GPU operates at a core clock speed of 1,395 MHz and can dynamically boost its clock speed up to 1,695 MHz. Complementing the processing units are 328 texture mapping units (TMUs) for efficient texture filtering and 112 render output units (ROPs) for pixel processing. Additionally, the GPU features 328 tensor cores optimized for AI-accelerated workloads and 82 RT cores dedicated to real-time ray tracing calculations.

Compatibility & Power Consumption

The RTX 3090 is a large graphics card, occupying 3 PCIe 4.0 x16 expansion slots. It supports HDMI 2.1, DisplayPort 1.4a display connections. With a thermal design power (TDP) of 350 W and a large, 3-slot size, it is recommended to use a power supply of 750 W and a case with sufficient space to accommodate the card. Compare your card's model and your case's dimensions to verify compatibility.

Benchmark Performance & Value

The RTX 3090 has the 23rd best 3DMark 11 Performance GPU score among the 567 benchmarked GPUs in our database. It achieves 47.44% of the performance of the best benchmarked GPU, the NVIDIA GeForce RTX 5090. Its 61,569 score and $1,499 launch price (MSRP) gives it a performance per dollar of 41.07. This is the 82nd best in value for the 3DMark 11 Performance GPU benchmark.