Nvidia Corp. is reportedly working on its own x86 central processing unit, a rather unexpected move considering the fact that the company does not have license on x86 instruction set, but also a fairly logical one now that the company has switched its chief scientist.

A news-story by The Inquirer web-site claims that “Nvidia is definitely working on an x86 chip and the firm is heavily recruiting x86 engineers all over Silicon Valley”. Back in 2006 the designer of graphics processing units hired numerous specialists from Stexar company, which also designed central processing units (CPUs) and related technologies. Additionally, back in January ’09 Nvidia replaced its chief scientist David Kirk with William Dally, who is known for his expertise in parallel computing technologies as well as work on Cell/Larrabee-like chips.

Since Nvidia, unlike Advanced Micro Devices or Via Technologies, does not own a license on x86 instruction set originally developed by Intel Corp., it cannot formally launch x86-compatible processors. However, Nvidia may license IBM’s Power architecture or even code-morphing software developed by Transmeta. In that case Nvidia’s microprocessors will hardly be really competitive performance- and support-wise.

Theoretically, Nvidia could launch an x86 CPU without getting a license from Intel Corp., but a legal battle with the world’s largest chipmaker would hardly end in Nvidia’s favour. Moreover, considering the fact that Nvidia will have 0% of the market from the start as well as no infrastructure in place, and that the legal dispute would last for a couple of years only it is hardly possible that by the end of the litigation the company will have substantial enough market share to justify a law suit from Intel.

Nvidia badly needs its own microprocessors in order to promote its core-logic sets as well as platforms. Since forthcoming x86 CPUs from AMD and Intel will have graphics processors integrated, there is hardly a lot of space on the market for Nvidia’s chipsets, hence, creating of system-on-chip products similar to Tegra, but aimed at personal computers, may be an option for Nvidia.

It should be noted that Nvidia has been working hard to adopt its GeForce graphics processing units for general purpose computing, which helps to accelerate processing of special-purpose software, but does not allow GPUs to run Windows operating systems.

Nvidia did not comment on the news-story.

Tags: Nvidia, Geforce, Nforce, x86