X
Innovation

The inevitable monopoly of Intel

Combine the cost of designing a microprocessor with Moore's Second Law and you get a cost squeeze that is driving everyone else out of the chip business.
Written by Dana Blankenhorn, Inactive

There is a decades-old secret hidden in the news that Intel is resisting a European Community fine of $1.45 billion, arguing its use of rebates was not anti-competitive and that it's not a monopoly in any case.

Whether it is a monopoly or not, Intel will eventually become one. Its fate was sealed practically at its birth.

You will find it in what is known as the Moore's Law paper, actually a magazine article by Intel co-founder Gordon Moore, then working with Robert Noyce at Fairchild.

Here it is in Moore's own words:

The complexity for minimum component costs has increased at a rate of roughly a factor of two per year. Certainly over the short term this rate can be expected to continue, if not to increase.

In other words, while Moore predicted that the density of components on a chip could double every two years, he also predicted that complexity would rise at the same rate.

The first prediction is known as Moore's Law and 44 years later it still works. That's the miracle that has made PCs, the Internet, and your XBox possible.

The second, less well-known prediction, is sometimes called Moore's Second Law. It is holding up as well.

Each time Intel, or any chip-maker, increases circuit density, the costs of the equipment needed to make the chip rises. The investment needed to build a chip-making fabrication plant or "fab" keeps rising, too.

Thus many big chip-makers, like nVidia, are now called "fab-less" because they contract out the actual production and concentrate on design.

Combine the cost of designing a microprocessor with Moore's Second Law and you get a cost squeeze that is driving everyone else out of the chip business.

Even AMD, Intel's longest and best-known rival, is now on the ropes, despite having beaten Intel to the punch on many important innovations this decade, like low-power designs and "multi-core" technology that divides processing into parts.

Even if Intel is not a monopoly today, in time it will be and must become one. Moore's Second Law decrees it.

A lot of ink, both real and virtual, has been spilled in the last decade asking whether Moore's Law can hold up. Even Mr. Moore himself has weighed-in.

The real question, however, is not the fate of Moore's Law, but the fate of Moore's Second Law. On that hangs the fate of Intel, and the whole chip-making industry it gave birth to.

This post was originally published on Smartplanet.com

Editorial standards