Perhaps, the launch of the Radeon HD 7970 graphics card powered by Advanced Micro Devices’ latest Southern Islands/GCN (graphics core next) was not the most impressive one as it was a rush announcement two days ahead of Christmas. But AMD does have a next-gen GPU on the market, unlike its major rival Nvidia Corp., a clear advantage. But the latter claims that the model 7970 fails to impress.

“Honestly, we expected more from our competitor’s new architecture,” an unnamed official Nvidia initiative told NordicHardware web-site.

Nvidia itself has been astonishingly quiet about its next-generation The company has learnt its lesson when it unveiled code-named Fermi architecture about half a year ahead of actual release and would like to ensure instant availability of next-generation graphics cards after the formal launch, according to unofficial information. At present Nvidia considers scheduling the launch of code-named Kepler products on March or April. Officially, the company claims that its Kepler will be released in the first half of 2012.

While the launch of the AMD Radeon HD 7970 was rather weak in terms of publicity and in many other terms, it is clear that today, around a month after the formal product announcement, AMD has actual new-generation graphics cards on the market. Meanwhile, it will take another two or more months for Nvidia to start selling its next-gen GeForce GTX 780-series. Naturally, Nvidia is interested in slowing down sales of AMD’s code-named Tahiti graphics solution so that to satisfy delayed demand for high-end graphics boards later this year.

Kepler is Nvidia's next-generation graphics processor architecture that is projected to bring considerable performance improvements and will likely make the GPU more flexible in terms of programmability, which will speed up development of applications that take advantage of GPGPU (general purpose processing on GPU) technologies. Some of the technologies that Nvidia promised to introduce in Kepler and Maxwell (the architecture that will succeed Kepler) include virtual memory space (which will allow CPUs and GPUs to use the "unified" virtual memory), pre-emption, enhance the ability of GPU to autonomously process the data without the help of CPU and so on. Entry-level chips may not get all the features that Kepler architecture will have to often. Production of Kepler chips was supposed to start in Q4 2011.

Tags: Nvidia, Kepler, Geforce, Quadro, Tesla, Tahiti, Southern Islands, GCN, 28nm