Moore’s Law Is Dead. Long Live Moore’s Law.

The year is 1965. The Beatles release four albums. The war rages on in Vietnam. And a research-and-development professional by the name of Gordon Moore makes a bold prediction that will have irrevocable implications on technological development, impacting many sectors of the global economy for decades to come.

Over the last 50 years, Moore’s observation — that the number of transistors on silicon chips and therefore their processing power was doubling approximately every 24 months — has evolved from observation to market demand to Moore’s law.

As we reflect on Moore’s law at its 50th anniversary, we cannot overlook its significance. It is at the heart of the entire technology industry and beyond, driving productivity, the economy and, indeed, human society.

And in fact, Moore’s predictions became a self-fulfilling prophecy. The computing power of chips not only did double every 24 months, they had to double every 24 months or the tech industry — and the economy at large — would suffer dire consequences, stifling innovation and economic advancement.

But not to worry. Silicon technology advanced at the required speed, and Moore’s law held true for many decades. Not only did chips keep shrinking, so did the cost of making them, which allowed constant price reductions for consumers, propelling the chip industry to become a $300 billion titan of the global economy. Moreover, it propelled technological innovation at speeds that had never before been seen. Enter the personal computer, mobile phone, touchscreen devices and many other technologies that have spurred titanic transformations in the way we live, communicate and transact.

“No human invention had ever exhibited the [exponential] rate of improvement,” Michael Malone wrote of the transistor in his book, “The Intel Trinity: How Robert Noyce, Gordon Moore and Andy Grove Built the World’s Most Important Company.” “In later years, writers would search for analogies. One popular comparison in the 1970s drew from the automobile industry, suggesting that if Detroit had kept up with Moore’s law, cars would go 500 mph at 200 mpg … and cost $1.50.”

Looking ahead, the real implications of Moore’s law provide far greater value than the technological advantages alone. Over the last few years, next-generation applications have been dreamed about in labs that will make the past decades of advances appear trivial. Sensors, actuators and overall connectivity between our physical world and networks are booming, with some estimates claiming that today’s 12 billion devices produced annually will grow to more than 50 billion or even a trillion over the next decade. Semiconductors will transform entire industries by helping monitor and manage supply chains with great precision and efficiency.

But here’s the rub. Moore’s law is running out of gas. In 2012, for example, the cost per silicon transistor started to climb due to the investment required for each new, more powerful generation of chip. Further, silicon has reached its performance limitations, failing to fuel innovation at the speed to which we have become accustomed.

What does this mean? It means that without a change in the processor medium fundamentally in use today — namely silicon — the applications that we expect from the near future and beyond, are not likely to see the light of day. These are fun applications, like virtual-reality glasses and advanced gaming features. They are applications designed to add great convenience to our lives, like wireless power and autonomous vehicles. But most significantly, they are life-changing applications, like mobile communications in developing countries and wireless heart pumps that dramatically improve the quality of life for those using them.

And it means that innovation at large will be stifled — the same innovation that fuels economic prosperity and the global economy.

But there is also good news. Today, Gallium nitride (GaN) technology is driving the next wave of innovation. As advances in silicon stall and get more expensive, progress in GaN is giving Moore’s law a second life, with new generations of chips doubling the performance of prior generations while decreasing in cost. In fact, today’s generation of GaN transistors already outperform silicon by a factor of 10.

The computer industry was built on sand, quite literally. The ability to produce smaller, more powerful silicon transistors turned the impossible to the possible and created the world we live in today. But now the sands are shifting by necessity, transforming industries and replacing the very technology that has spawned technological life as we know it today. And as we make the transition, the technology industry continues to bring to reality the wonders never before imagined … and breathing life into Moore’s law once again!

Alex Lidow is co-founder and CEO of Efficient Power Conversion Corporation (EPC). Since 1977, he has been dedicated to making power conversion more efficient upon the belief that this will reduce the harm to our environment from energy production and consumption. In order to pursue this mission, in 1977 he joined International Rectifier as an R&D engineer. In 1978, he co-invented the HEXFET power MOSFET, a power transistor that launched the modern power conversion market and displaced the aging bipolar transistor. Royalties from these patents brought in more than $900 million over the years, and International Rectifier to this day is the largest producer of power MOSFETs in the world. He recently co-authored the first textbook on GaN transistors, “GaN Transistors for Efficient Power Conversion.” In 2004, he was elected to the Engineering Hall of Fame. Reach him @AlexLidow.

This article originally appeared on Recode.net.