No Intel Discrete Graphics in the Short Term

While this isn’t the first time this headline has found its way over the Internet, some interesting details have been made available as to why a Larrabee card won’t be made, at least in the short term. A blog by Bill Kircos provides some in-site into this decision and future focus of Intel regarding their graphics based technologies.

Many people became excited when Intel unveiled plans to enter the discrete graphics processing market with Larrabee, only to have their hopes diminished back in December when Intel demoted Larrabee from a product to a architecture technology. Bill has now made it perfectly clear by stating “We will not bring a discrete graphics product to market, at least in the short-term.” So what is their focus now, and what will happen to the Larrabee project?

Intel’s graphics development efforts concentrate on three areas. First is processor graphics, that of integrating both CPU and GPU onto a single chip. Second is their System on Chip and Atom processors, as well as HD acceleration with their existing IGP’s. Third is their many-core, programmable architecture, to which we knew as Larrabee.

What’s obvious is Intel’s concentration on heterogeneous computation and harnessing parellel processing technologies. They have their Terascale project, the 48-core Supercomputer on Chip (SCC) and the now former Larrabee project which will probably be used within the High Performance Computing (HPC) sector.

While we may not see a gaming orientated discrete graphics card from Intel anytime soon, there’s a good chance that the HPC sector will see a massively parallel programmable architecture based product in the medium term, further confirming  a potential market that Nvidia has already aggressively persued with CUDA and their Tesla products. All of this now puts the pressure on AMD/ATI to further develop their Firestream technology and/or OpenCL support on their existing graphics cards for the HPC sector.

We will not bring a discrete graphics product to market, at least in the short-term. As we said in December, we missed some key product milestones. Upon further assessment, and as mentioned above, we are focused on processor graphics, and we believe media/HD video and mobile computing are the most important areas to focus on moving forward.