After months of speculation, NVIDIA's Kepler GK110 GPU-based GeForce graphics card is finally official. Sharing similar specs to the Tesla K20 enterprise part, the GTX Titan boasts 2688 CUDA Cores and 7.1 billion transistors, making it by far the most powerful single GPU ever created for consumers. It also utilizes improved NVIDIA technologies like GPU Boost 2.0. We take a closer look.

NVIDIA's GeForce GTX Titan boasts 2688 CUDA Cores and 7.1 billion transistors, making it by far the most powerful single GPU card ever created.

The details of the Xbox 720 GPU have been leaked online. According to the site VGleaks, the new GPU has an 800MHz clockspeed and features twelve shader cores. The site also claims that the DirectX 11 compliant graphics chip is also capable of supporting 2x, 4x and 8x MSAA anti-aliasing effects.

As the year comes to a close it's time to take a look again at AMD's Radeon HD 7970 GHz Edition and NVIDIA's GeForce GTX 680 to see who can end the year as the reigning king of GPUs. We'll also take a look at which is the better buy if you're planning on building a new system early in 2013.

According to an article from SemiAccurate, AMD is ending its ‘Vision’ branding for its APU platforms and PC systems using these platforms.

Details of the new mobile GPU series from AMD have been revealed. The AMD Radeon HD 8000M series mobile GPUs are touted to deliver better performance than the current Radeon HD 7000M series, and with improved energy efficiency as well.

NVIDIA is rumored to be launching its GeForce GTX 650 Ti next month in October. Its specifications have also been leaked; read on to discover how it stacks up against the GeForce GTX 660 and 650 GPUs.

At a recent Q & A session to an audience at the Aalto Center for Entrepreneurship in Otaniemi, Finland, Linus Torvalds who is the brain child behind the Linux kernel project, complained harshly about the lack of Linux support from NVIDIA. NVIDIA has issued their diplomatic response to his outburst.

With Intel already dominating the CPU arena, what else can first-runner-up AMD do to catch up or worse still, stay relevant? With the introduction of the Heterogeneous Systems Architecture consortium, it seems as though AMD seeks safety in numbers.

NVIDIA announces that open-source LLVM compiler now supports NVIDIA GPUs as a result of the company's contribution of its CUDA complier to the LLVM Project.