At WWDC 2020, Apple revealed perhaps one of the worst kept secrets in tech. They were ditching Intel chips found in the Mac in favour of custom designed Apple Silicon. Of course this isn’t the first time that Apple has made big platform transitions. In the WWDC keynote they pointed out a couple of major transitions, the move from MacOS 9 to MacOS X and the move from PowerPC architecture to Intel x86. Let’s take a look at why Apple made some of these historic transitions and what it tells us about what we can expect moving forward.

RIP MacOS 9 and PowerPC

MacOS 9 was the last version of the so called ‘classic MacOS’, the version of MacOS that shipped with the introduction of the Mac in 1984. It was released as a stop gap in 1999 to support new hardware including a new product called the iMac. Apple was struggling to maintain MacOS and to keep it modern for new software advancements and network connectivity.

Apple tried and failed to build a new future OS framework in the mid 90’s and instead looked elsewhere. The company ended up acquiring a smaller computer firm called NeXT for their forward thinking UNIX based OS called NeXTSTEP. Along with acquiring NeXT, Apple also acquired a particular employee known as Steve Jobs. Steve of course was one of the original co-founders of Apple and was famously kicked out in a corporate coup in the 80’s.

Acquiring NeXT was the single most important thing that happened at Apple in the 90’s. Quite simply Apple would have gone bankrupt without making that acquisition. Apple was 90 days away from liquidation before Steve returned. Steve quickly worked with the teams at Apple to modernise MacOS for urgently needed new products and launched the iMac.

Of course by acquiring NeXT, Apple had the technology and now the leadership talent to build a truly next generation operating system for the Mac. It was built and designed to support different instruction sets by design, meaning it wouldn’t be restricted to PowerPC processors forever. PowerPC offered some great benefits in the 90’s but the development road map began to slow and Apple wanted and needed to be ready. As Steve put it “just in case”.

Steve Jobs famously said to Walt Mossberg and Kara Swisher at All Things D8 “Apple likes to pick technologies in their spring” and by 2000/2001 it was abundantly clear that PowerPC was a declining technology. Fortunately Apple had planned for that future by investing in a new Operating System that could support Intel processors. MacOS X launched in 2001.

MacOS X leads a secret double life….

When Apple announced the transition to Intel processors in 2003, Steve said that the transition would take around two years and would be completed by the end of 2005. The process was gradual and the roll out staggered. Apple had often touted the benefits of PowerPC so why did they plan to make this transition? The main reason is performance per watt. Intel chips offered significantly better power consumption at comparable performance levels. The products Apple wanted to build just weren’t possible with PowerPC architecture, particularly mobile products.

As mentioned, MacOS X was built by design to support Intel chips. Although Apple announced the transition to Intel with the roll out of MacOS X Tiger (the fourth version of Mac OS X), MacOS X secretly always supported Intel instruction sets from day 1. This allowed Apple to begin the transition almost immediately.

The transition wasn’t perfect but it was important for the future of the Mac. Apple knew there would be speed bumps but planned for those and tried to mitigate some of the challenges posed by the transition. It did that with two key pieces of technology, Universal and Rosetta. Universal was in the most basic sense, a requirement that developers ship versions of their Apps that include the installation code for both PowerPC Macs and Intel Macs. Rosetta translates Apps from PowerPC to Intel. Rosetta is Apple’s binary translator designed to make it easier for developers to port apps to Intel chips with less time required to write new code.

Steve remarked that Mac OS X would take the Mac forward for the next 20 years when it was announced at Macworld in 2000. He was right.

2020 and the next big transition.

Starting with the original iPad in 2010, Apple started to develop processors in house. The first chip that shipped in a product was the Apple A4 which launched with the iPad and quickly came to the iPhone a few months later. Apple felt a real need to build the technology themselves so as to deliver the kinds of performance and energy efficiency needed for future iOS devices.

In 2011 Apple shipped their first dual core chips with the A5. The A5 offered particularly remarkable gains in graphics performance, a boost of 9X year over year. In 2013 Apple shocked the entire mobile industry by shipping the first 64 bit processors for a smartphone with the A7. Competitors were shook and tried to downplay the importance of this move. It took 2-3 years for the competition to catch up to this massive development.

Fast forward to 2020 and Apple’s A series chips offer industry leading performance in mobile and in the tablet space. The A12Z processor shipping in the current iPad Pro is faster than some of Apple’s own computers running intel processors already! The A12Z can even transcode three streams of 4K video simultaneously and shows Apple isn’t ignoring graphical performance either. With all of this innovation in the mobile space, it was only a matter of time before Apple started to experiment with what would happen if they put their own chips in the Mac.

At WWDC Apple finally confirmed the long rumoured transition to Intel. The transition feels much live dejavu. Apple is bringing back Universal and Rosetta with version 2.0 of each. These important technologies will help developers and customers make the leap to Apple Silicon Macs while supporting owners of Intel Macs for several years to come. Apple has even stated that they have several Intel Macs still in the pipeline before the transition is complete.

Just because they can….does that mean they should?

Intel processors are widely used in the PC space. Unlike PowerPC, Intel chips enjoy wide adoption rates and support from many PC manufacturers and software companies. So why did Apple feel the need to make this transition?

1. Intel controls their own roadmap. All to often in the last few years, Intel has been slow in shipping new processors and has created delays in Apple’s own product launches. This is something that is counter intuitive to Apple’s philosophy of “Owning the primary technology”. By building their own chips, Apple controls it’s own destiny and can ship products on their own schedule without external disruption.

2. The move to PowerPC was so that Apple could create products that just weren’t possible without Intel chips. History is repeating itself today. The iPad Pro offers better performance than the MacBook Air and even some models of MacBook Pro with better battery life and no fan. Performance per watt of power is much higher in Apple designed chips. It isn’t hard to imagine future Mac form factors that just aren’t possible with the Intel chips available today.

At the original iPhone announcement in 2007, Steve Jobs quoted the technology pioneer Alan Kay, “People who are really serious about software should build their own hardware”. This quote is symbolic for Apple’s belief in owning the core technology and in many ways, the transition to Apple designed silicon for the Mac, was inevitable. In many ways it’s a new beginning and Apple clearly feels that way too. After all MacOS is moving to version 11 later this year too.

I think this is the right move for Apple and for the Mac. The transition won’t be perfect. Some people won’t like it. Critics will be ready to complain when certain apps don’t work quite as well right away. Others will lament the loss of native Windows support via boot camp. In a few years through I predict that those grumbles will fade and consumers won’t care. They’ll be too busy enjoying high performance machines with futuristic performance designed for the 2020’s.

Leave a Reply