In late December, AMD introduced the Radeon HD 7970 which made it overtake nVidia as the world's fastest single-GPU graphics card manufacturer. AMD not only surpassed nVidia in terms of performance, the Radeon HD 7970 was also the world's first video card with a 28 nm GPU, the first with a PCI-Express 3.0 interface, and the first to support DirectX 11.1.
We were eager to see what nVidia's response would be to AMD's advanced new graphics card. That has now arrived in the shape of the GeForce GTX 680, a new graphics card based on nVidia's new 28 nm GK104 GPU.
The new GTX 680 is almost un-nVidian. Whereas nVidia used to focus purely on performance regardless of energy consumption and noise production, its new generation revolves more around efficiency, according to nVidia themselves. If we take a sneak peek at our test results, we see that the GeForce GTX 680 is generally faster than the Radeon HD 7970 while using less power. It should be noted that the new focus on efficiency was not purely a choice, but rather born out of necessity. We will elaborate on this on the next page.
In addition to better performance, the GTX 680 has other improvements compared to its predecessor, the GTX 580. These include an innovative turbo feature, a new anti-aliasing system, and the possibility to connect more than two monitors to one video card.
With an average price of £410, the GeForce GTX 680 is positioned directly opposite the Radeon HD 7970. We will answer the question which of these is the better choice at the end of this article. First, let's examine the hardware.