NVIDIA GeForce GTX 680 vs AMD Radeon Pro Vega 20

The NVIDIA GeForce GTX 680 is an end-of-life desktop graphics card that released in Q1 2012 with a MSRP of $499. It is built on the Kepler GPU microarchitecture (codename GK104) and is manufactured on a 28 nm process.

Memory

The GTX 680 has 2 GB of GDDR5 memory, with a 1,502 MHz memory clock and a 256 bit interface. This gives it a memory bandwidth of 192.3 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 GTX 680 includes 1,536 CUDA cores, the processing units for handling parallel computing tasks. The GPU operates at a core clock speed of 1,006 MHz and can dynamically boost its clock speed up to 1,058 MHz. Complementing the processing units are 128 texture mapping units (TMUs) for efficient texture filtering and 32 render output units (ROPs) for pixel processing.

Compatibility & Power Consumption

The GTX 680 occupies 2 PCIe expansion slots. It supports DVI, HDMI 1.4a, DisplayPort 1.2 display connections. NVIDIA recommends a power supply of at least 450 W to handle the GPU's thermal design power (TDP) of 195 W.

Benchmark Performance & Value

The GTX 680 has the 178th best 3DMark 11 Performance GPU score among the 543 benchmarked GPUs in our database. It achieves 9.77% of the performance of the best benchmarked GPU, the NVIDIA GeForce RTX 4090. Its 10,216.5 score and $499 launch price (MSRP) gives it a performance per dollar of 20.47. This is the 85th best in value for the 3DMark 11 Performance GPU benchmark.