The new Apple M2 was the most exciting announcement made at WWDC 2022 for Mac enthusiasts. It’s the chip that’ll power the M2 MacBook Air and 13-inch MacBook Pro, which will be available next month, and Apple claims it’s up to 18% faster than the M1 while using the same amount of power. It sounds fantastic, but is the M2 truly a generational leap as Apple claims?
The problem is that the pandemic has cut off the supply of chips, which has messed up the schedules for almost every major tech product. Apple doesn’t seem to be any different, and it looks like the company will release a new family of desktop and mobile processors no matter what new tech comes out.
An improvement, not a successor
The new M2 processor from Apple is mostly an update to the M1 instead of a whole new processor. That has mostly to do with how M2 is made. Chips for the M1 and M2 are made by TSMC, and Apple says that the M2 has a “second-generation 5nm” node.
Between CPU generations at TSMC, which is by far the world’s largest semiconductor company, you want a full node improvement. That means making the process of making chips smaller so that more transistors can fit on a chip while making the process more efficient. The problem is that TSMC pushed back its next-gen node, which was supposed to come out in 2021 and seemed like a good fit for Apple’s M2.
The N5 node from TSMC is used to make the M1, and the N5P node will almost certainly be used to make the M2. The real next-generation node is N3, which uses a 3nm process that is up to 15% more powerful and uses 30% less power than N5. N5P, on the other hand, is 7 percent better and uses 15 percent less power.
Those numbers don’t tell us how well the M2 works, but they show how different a node improvement and a generational node jump are. Based on Apple’s numbers and what we know about the M2, it will be faster than the M1, especially when it comes to graphics. But unlike Apple’s first generation of silicon, this won’t be the revolution that Apple is trying to sell. Apple isn’t likely to make anything as unique as the M1 ever again. Read also; Apple “Peek performance”: How to watch and what to expect?
What’s new then?
The M2 from Apple seems to be an improvement on the M1. This starts with the manufacturing process. Even though it’s not a full node jump, Apple can still fit 20 billion transistors on the die. The M1, on the other hand, has 16 billion. That wouldn’t make a huge difference in performance, so Apple has made a few other changes to the M2 to make it faster.
The M2 is still an eight-core processor with four efficient cores and four performance cores, but Apple says the clock speeds are tuned a little faster. Instead of 12MB on the M1, the performance cores now have 16MB of L2 cache, and Apple added 64KB to the instruction cache. The biggest changes are outside of the CPU. There are now two more GPU cores and up to 24GB of unified memory (the M1 topped out at 16GB).
Even though these changes are useful, they usually aren’t as good as a full node jump. By making the M2, Apple has set itself up for a release schedule like Intel’s. Every other generation could be a small improvement, but not a big change. Intel is also a great example of the problems that come up when making different generations of CPUs on the same node (rest in pieces, Intel 14nm).
The question is whether Apple will be able to keep up with performance until the M3 comes out. The M1 kept the business going for two years, and the M2 will probably have to do the same. Now, the problem is that we’re looking at a generation of processors that might only bring small improvements in speed and efficiency, not enough to shake up the whole CPU industry.
Only M2 in name
Apple is sticking with M2 for its new MacBooks. Future MacBook Pros, Mac Studios, and maybe even Mac Pros will likely use different versions of M2. Even so, the name doesn’t tell the whole story. On the architecture side, we’ll have to wait until the M3 to see a real generational jump from Apple silicon. Read more; How to pair Bluetooth Earbuds to the Apple Watch
It’s possible that Apple has a secret ingredient that will make the M2 chip just as important as the M1. We don’t have the processor yet, so all we can do is trust what Apple says. The M2 is called “next-gen,” but the M3 is likely to be “next-gen” in terms of performance. We can only guess right now, so it’s best to wait until Apple tells us more.