Virage absorbs a key piece of NXP: signs for the future of IP?

In a startling move announced this morning, growing semiconductor intellectual property vendor Virage Logic announced that, through an unusual and complex agreement, it has in effect acquired much of the technological heart of NXP Semiconductors. The agreement, billed by the two companies as a strategic alliance, cannot but bring profound change to both organizations.

Under the agreement NXP will transfer about 160 employees, with their associated assets, IP, roughly 25 families of patents, and four years’ of cash payments totaling $60 million, to Virage. In exchange, Virage will give NXP 2.5 million shares of common stock, valued at today’s closing at about $15M, will continue to make the acquired IP and its support available to NXP for 3.5 years, and will license Virage’s existing standard-product IP to NXP for the same 3.5-year period. The companies expect to finalize the agreement in the fourth quarter of this year.

The NXP IP involved, according to Virage CEO Alex Shubat, includes I/O technology, analog and mixed-signal blocks used in media SoCs, and what Shubat described as SoC infrastructure IP. This last category includes system controller blocks, bus interfaces, proprietary processor-core hardening technology, and NXP’s IP delivery and automated SoC assembly technology. Fundamentally, it appears that NXP has divested all the central technology services that allowed them to produce differentiated SoC designs.

Unusually, the team Virage is acquiring will remain in Eindhoven, Netherlands, their present location, but under new Virage management. The agreement even includes some of the team’s equipment. So it appears the two companies are making every effort to keep the group together and functioning.

For Virage the agreement brings a treasure trove of IP—particularly in the analog signal-path area—and technology for assembling IP into an SoC. "Eindhoven is a world center of excellence for analog design," Shubat observed. "And of the 160 people who will join us, most are technical, not administrative or marketing."

The agreement will increase the size of Virage’s technical staff—as counted including the company’s intended acquisition of ARC Cores—by nearly a third. And it may fundamentally alter the nature of the company.

Taken with Virage’s other recent acquisitions—especially the CPU IP vendor ARC—the picture that emerges is not that of a conventional IP vendor anchored in industry-standard interfaces and logic libraries. Rather, Virage begins to look more like a one-stop IP, methodology, and services organization for all aspects of creating a leading-edge SoC. Their portfolio will include libraries and memory technology, standards-based interface IP, analog functional blocks, headline IP such as CPUs and accelerators, infrastructure IP for creating the increasingly complex networks-on-chip and system controllers necessary to make SoCs function reliably, strong foundry relationships, and even a proven push-button SoC development methodology based in the technology behind IP-XACT. Such a company is only a small step away from becoming a powerful fabless ASIC player in the media IC market.

There are risks for Virage as well. Even though the company has been reorganizing for over a year to support rapid expansion, no executive adds a third to their engineering department lightly. This agreement will impact every business unit at Virage, and will create some new ones, Shubat said.

The picture for NXP is entirely a different one. The joint press release suggests that the agreement will allow NXP to accelerate its shift away from expensive-to-develop and competition-prone media SoCs, and toward highly-differentiated analog ICs. Increasing focus is often a good idea. And NXP will retain access to its team’s IP for another 42 months, even though the company may lose the ability to influence the team’s project priorities.

But there is a darker interpretation of the agreement too. This view suggests a heavily-indebted, private-equity-owned company, shut off from short-term credit markets and struggling to service its debts. Unable to rapidly increase revenue, the company would have to turn to the expense side of the income statement to liberate more cash flow for debt service.

That scenario would certainly explain NXP’s willingness to make $60 million in future cash payments in order to get a superb engineering team—and their perhaps $30 million/year in direct expenses—off the books. Saving expenses in an emergency can work. But like a sailing ship fleeing pirates, a company has to be certain that the stuff they pitch overboard to lighten ship won’t turn out to be necessary later. You only roll the guns overboard if you are certain to lose should you keep them.