Apple launched a new line of MacBook Pro laptops this week. Apple's engineers and executives clarified that these new MVPs are the M1 Pro or M1 Max chips. They are responsible for the super-high resolution displays and extended battery life of the Mac devices. The laptops are the culmination of a 14-year-old strategy that transformed Apple literally under the hoods of its products. It took a huge effort to create and manufacture its own chips. Apple is replacing the microprocessors that it purchases from vendors such as Intel and Samsung by its own. These chips are optimized to meet the needs of Apple users. This has been a remarkable success. Apple was once a company that was defined by its design. Apple is still a design company, but I consider it to be a silicon company.
Two days later, I had an unusual on-the-record conversation with John Ternus (aka Joz), senior marketing VP Greg Joswiak (aka Joz), and John Ternus (senior hardware engineering VP). Srouji was someone I'd been asking Apple for many years. His title is a hint at his position as Apple's chip czar. Although he has been seen on camera at Apple events recently, he avoids the spotlight. Srouji, an Israeli engineer, joined Apple in 2008 to fulfill a mandate by Steve Jobs. He felt that the chips inside the original iPhone were not capable of meeting his needs. Srouji's mission was to help Apple make its own silicon. It has been a great effort. I think Srouji secretly succeeds Jony Ive as the creative genius behind Apple's secret sauce.
Srouji won't admit to it. Apple executives are trained to focus their hyperbole on Macs and iPhones and iPads, rather than themselves. He claims that Apple makes the best silicon in the entire world. Apple is a product company, which I keep in mind. This is heaven if you are a chip designer because you build silicon for a company which builds products.
Advertisement
Srouji is very clear about the benefits of making your own chips as opposed to purchasing from an Intel vendor. This week, the Intel MacBook Pros were withdrawn from the market in favor of M's. He says that merchant vendors, which are companies that sell off-the-shelf silicon or components to customers, must determine the lowest common denominator. What is it that everyone uses over many years? The silicon, hardware, software and industrial design all form one team. This allows us to achieve a particular vision. This gives us an incredible opportunity and freedom to design something unique and optimized for silicon. All of it would be made from silicon. As we sit down together, we ask each other: Is it governed by physics? We sit down together and decide if it is gated by physics. Or can we get beyond it?
Consider that the only restriction Apple chipmakers accept is the physical limit of whats possible.
Srouji explained that his journey with Apple was one of conscious iteration and building upon a solid foundation. The company's strategy has focused on integrating functions that were previously distributed across many chips into one entity known as SOC (system-on-a chip). He says that the key to building the best chip is having the right architecture. We started with architecture that would scale. Scaling refers to the ability to scale performance, features, and the power envelope of a watch, iPad, or iMac. Then we began to figure out the technology within the chip. We wanted to own them all one at a time. The CPU was our first priority. Then we moved on to the graphics. Next, we moved into signal processing and display engine. We grew our engineering skills and knowledge year after year. And then, a few more years later, you have really good architecture and IP and a team that can repeat that recipe.
Advertisement
Ternus explains: In the past, you had one chip design team in one company. They have their priorities and optimizations. Then, the product team and another company have to take that chip and make sure it works in their designs. We started with these MacBook Pros right at the beginning. The chip was being developed as the system was being planned. These high-performance components make power delivery difficult and important. The team came up with a solution by working together early on. The system team actually managed to change the SOC's shape, aspect ratio and orientation so it could nest in the rest of system components. This may have helped Apple restore the missing ports many had longed to see in the old MacBook.
These executives clearly believe that the new Macs are a significant step in Apple's strategy. It is not the last. A future milestone could be silicon that can be customized to enable augmented reality systems. This would produce the graphics intensity, precision geography, and low power consumption required for AR spectacles. The VPs didn't comment on this, as expected.
Before we end the conversation, Joswiak asks me about the discontinued Touch Bar. This dynamic function-key feature was launched by Apple five years ago with much fanfare, but never gained traction. It is no surprise that Joswiak views it as a wonderful gift for new users. It's clear that Pro customers love the full-size tactile feel of these function keys. So that was our decision. He says that we are proud of this decision. Apple still sells the obsolete 13-inch MacBook Pro with soft keys intact for those who love the Touch Bar, however they may feel.
The Touch Bar's story reminds us that even the most powerful silicon cannot guarantee designers will make good decisions. Srouji points out that when it is done correctly, it can lead to endless possibilities for innovation that would not have been possible otherwise. The launch of the MacBook Pro was not the best indicator of Apple's silicon success, but Google's unveiling the Pixel 6 smartphone. Google claimed that the phone's key features were the result of a decision by Google to follow the 14-year-old path Srouji and Apple took in building their own chip, the Tensor processor.
Is this an example of Imitation is The Sincerest Form of Flattery? I ask the Apple team.
You took my line! says Joswiak. They clearly believe they are doing something right.
What would you give Google or any other company to help them on their journey with silicon?
Joz says, Oh, I don't know. Buy a Mac
This story first appeared on wired.com
