Parallel Universes

There are no rules for knowing when to step out of the box. Good timing is everything, and that may have been one of the greatest talents of the late Steve Jobs. Knowing when, in Apple’s terminology, to “Think Different,” is every bit as important as the act of thinking differently—particularly when you realize that most of Apple’s big wins since the iPod stormed onto the consumer electronics market 11 years ago are not new ideas and don’t use the latest technology.

Unfortunately, good timing isn’t easy to come by. It isn’t clear yet whether Jobs’ successors inside Apple truly have it, but it’s abundantly clear that Apple’s rivals do not. The unexpected departure of Intel CEO Paul Otellini next May and the ongoing housecleaning and struggles at Microsoft are cases in point. Intel failed to grasp the handheld revolution just as Microsoft failed to understand the value of networking, gaming, Internet search, and handheld mobility until well after the markets had reached the explosive growth stages.

It’s hard to count Intel or Microsoft out, of course. They have deep war chests of cash, and people still need PCs for content creation. But the bulk of the action is in devices that are geared toward content consumption and only limited creation. That’s where the volume is, and it’s where the big growth opportunities are. And to gain a foothold in both of those markets, both Microsoft and Intel will have to think differently.

Intel could turn its expertise in both processor and process technology into generating low-cost, low-power 7nm or 10nm platforms for stacked die or even planar SoCs. But that requires a delicately timed switch from dictating what the market needs to listening to what the market wants, and Intel has always functioned in an insular manner, laser-focused on Moore’s Law and the next processor release. That’s why a search is under way at Intel for a new CEO, who very well could be brought in from outside the company. The word “vision” has cropped up in numerous analyst reports. There has yet to be any mention of the word “timing” to go with that.

Microsoft, meanwhile, has the opportunity to either go full-bore into hardware development, or to scale back its OS into a thinner, less-power-hungry and lower-cost software layer that improves performance of a variety of devices without bogging it down in Office-like functionality and performance—or both. But Microsoft has been late on almost every product and behind on every market push throughout its history except the introduction of DOS, even though it has trounced the competition over the long haul with its business acumen.

Nor are these challenges unique to Intel and Microsoft. The same challenges they face are the ones now facing big EDA companies. How much future business will be driven by companies such as Intel, when moving to the next process node requires huge volumes to reach profitability? Microsoft already has hedged its bets with RT. This also explains why all of the big EDA companies are diversifying. But in the core EDA markets they are moving along the same path as in the past—tools that work at 20nm and 14nm for a shrinking number of customers and a shrinking number of increasingly complex designs. Going in different directions requires leaps of faith coupled with some very good market trend information. And it requires some very good timing.

Knowing when to make a change is every bit as important as the change itself, which in EDA’s case requires some very costly investment. It’s also what keeps some very smart people awake at night worrying about exactly when is the right time to make a move in a different direction and what that direction should be.

—Ed Sperling