Apple’s M2 chip: Anticipating power gains

When the M1-based Macs were released in the fall of 2020, it’s no exaggeration to say that they revolutionized the entire PC market. Apple took the core architecture of the A14 (found in the iPhone 12), raised it to the levels of an ultra-portable laptop, and blew up the competition with what we call A “mind-boggling” improvement over the previous generation. was fast, is very Fast, and provided amazingly good battery life. With the M1 Pro and M1 Max in 2021, Apple expanded the build to high-performance laptops (and supposedly upcoming desktops), and it’s again The The processor that others have been compared to.

The only chips that outperform the M1 Max are Intel’s Alder Lake processorsand then only with a bunch of warnings about the huge discrepancy in power consumption, it only runs at full speed when plugged in, and it relies on a high-powered Nvidia GPU.

So what is Apple doing to show up? With the M1 now nearly a year and a half old, it’s time for its successor, which we’ll assume will be called the M2. While we don’t really know what Apple has until the products are announced (as early as March or at WWDC in June or later this fall), we’ll make educated best guesses about what we can expect from the M2 ( We’ll also give an M2 Max prediction as well).

M2 is to M1 as A15 is A14

The M1 is based on the A14 core architecture, with an increased number of high-performance CPU cores (four instead of two) and a doubling of GPU cores (eight instead of four). It is, for lack of a better explanation, the “A14X” that otherwise would have been in the iPad Pro.

In an effort to predict the features and performance of the M2, we’ll be assuming it will follow a similar pattern, only this time with the A15: the same architecture with twice the high-performance cores and GPU cores. Note that the A15 already owns five GPU cores, with only 5th position enabled on iPhone 13 Pro models. We’ll generate performance expectations by doubling that to 10 cores, which is ironic over rumors about the M2’s configuration.

For M2 Max, we assume the same expansion from M1 to M1 Max. The CPU cores will be swapped for a setup with efficiency of 2 and 8 performance, and the GPU cores will be quadrupled from the M1 core chip (for a total of 40).

Manufacturing process improvement

Before we get to the numbers, we have to mention that the M2 is expected to be manufactured on the TSMC N4P process, an improved form of the 5nm process on which the A14 and M1 are manufactured. TSMC claims it can run 11 percent faster with the same power, or use 22 percent less power with the same performance. It’s not really known if this will translate to faster clock speeds, better battery life, or both. It all depends on Apple’s own configuration, and battery life in particular is subject to a plethora of variables (like how energy efficient the display is, for example).

For our performance predictions here, we do not take into account additional performance from manufacturing process improvements, as there is no way to determine what would be realistic. It’s reasonable to expect the numbers presented here likely to be a few percentage points lower than what Apple might squeeze out of the N4P process, and as we’ve seen a significant improvement in battery life in the iPhone 13 running the A15, Apple may prioritize longevity in its MacBooks that use These chips.

M2 and M2 Max CPU ratings

We started by looking at how faster the A15 is compared to the A14, which features a quad-core, dual-core CPU configuration, but a number of structural improvements. Altogether, the performance of the single core A15 is about 7 percent higher than that of the A14.

The multi-core score is much higher, at 20 percent, despite the same basic configuration. It’s possible that Apple has made architectural optimizations to improve the use of all cores simultaneously, and manufacturing process optimizations may be able to maintain higher clock speeds when all cores are active.

IDG

When we take the M1 and M1 Max scores and add the same performance boost, we get some pretty impressive numbers. Single-core performance will likely outperform the 1800 in the Geekbench 5, which is on par with Intel’s high-performance Core i9 Alder Lake laptop chip.

IDG

It’s the multi-core result that can become very interesting. Rumors say the M2 will have the same quad-efficient, quad-performance design as the M1, while the M2 Max sticks to four efficiency cores and eight performance cores as the M1 Max.

IDG

That could generate scores just under 9000 for the M2 (still not quite as fast as the M1 Max, thanks to this chip with eight high-performance cores). This is a great result for ultra-portable laptops like the MacBook Air. But a score of over 14,000 for the M2 Max would put it well ahead of Intel’s fastest Alder Lake Core i9, which has 14 cores (six performance and eight efficiency). And the M2 Max will likely achieve these performance gains at a fraction of the power.

M2 and M2 Max graphics ratings

As with our CPU predictions, we estimate the M2’s graphics performance by simply looking at bumping up the performance from the A14 to the A15 (on iPhone Pro models) and assuming the same height from the M1 to the M2. The A15 has 25 percent more GPU cores than the A14 (five instead of four), but the extra core is only enabled on the iPhone 13 Pro phones. Rumor has it that the M2 will go from a maximum of eight GPU cores to 10, and the M2 Max from 32 to 40, all while keeping that ratio intact.

IDG

3DMark’s Sling Shot benchmark is a bit outdated, but it’s still a reasonably useful measure of how well a GPU will perform when running simpler, older graphics technologies. The A15 is 24 percent faster than the A14. You can chalk that up to get 25 percent more cores along with some architectural improvements.

IDG

Unfortunately, the Sling Shot test is only available for mobile platforms. This means that we cannot test it on the M1 Max, as this chip is only available in MacBooks. But since the M1 can be found in the iPad Pro, we can run this useful but somewhat outdated benchmark on the M1 and assume we might see a score of over 10,000 with the M2, which would easily put any other modern tablet to shame.

A more recent cross-platform test is the new 3DMark’s Wild Life. In the more extreme version of this test, the A15 is 30 percent faster than the A14.

IDG

Upgrading to the M2 and M2 Max, we can expect more than 6000 for the lower-performance chip and more than 26000 for the more powerful Apple processor. This last result is noteworthy, as it is very close to the portable variant of the GeForce GTX 3070.

IDG

While such a performance boost would put the M2 Max in line with a fairly high-end discrete GPU, it’s still even half the performance of the best desktop GPU. More importantly, serious high-end GPUs from AMD and Nvidia (and Intel’s upcoming Arc series) feature hardware-accelerated ray tracing, something Apple’s own GPUs have so far absent. While I think support for this feature has to happen sooner or later, it probably won’t reach the M2 line. Apple is already years behind the competition there, and it’s not likely to get any better this year.

Other considerations: neural engine, video encoder, memory

While the CPU and GPU are easy to identify, they aren’t the only important parts of Apple’s system on chips. RAM is very important, of course, and we expect high-speed LPDDR5 to feature prominently again in these chips, as it provides the right mix of performance and power efficiency. Apple has done a great job of making sure their chips have enough memory bandwidth to take full advantage of the CPU cores and GPU cores, and we see no reason why the M2 (or M2 Max) should be the first to take seriously the bandwidth limitation.

The Neural Engine is a dedicated processor block designed to accelerate machine learning and AI tasks, and while the A15 has 16 Neural Engine cores just like the A14, it’s been reengineered for better efficiency and can perform more than 40 percent of operations per second. We expect the same neural engine to make the jump to M2.

Finally, the A15 features an improved video processing engine, which speeds up video decoding and encoding. The M1 and M1 Pro have one, while the M1 Max has two. We wouldn’t be surprised at all if we find this continues with the M2 – one video engine optimized in the M2 and M2 Pro, and two of them in the M2 Max. It would be nice to see more improvements here, since Macs are often used as video production machines. In particular, Apple’s SoCs don’t yet have hardware support for AV1 code (neither encoding nor decoding), and it would be great to see that added in the M2 version of the video processing engine.

general expectations

While it’s exciting to think that the M2 may be some major architectural upset, it’s likely that it will continue the A15’s technology in the same way that the M1 was derived from the A14. Apple may be moving away from its Mac silicon away from the A-series architecture further in the future, but in the second-generation chip, the most likely scenario is “scaling the A15 in exactly the same way”.

Performance constraints will vary greatly depending on the task or standard at hand, and factors such as SSD performance and cooling always play a role. But overall, we think it’s reasonable to expect about 20 percent better CPU performance (and less than 10 percent on a single core), while adding additional cores will raise GPU performance by about 30 percent. These improvements will be accompanied by improved machine learning and video encoding performance as well.

All of this has to happen roughly in the same energy envelope, too. Apple rarely rolls back on battery life, so you can expect similar laptops equipped with the M2 chipset to have the same or better battery life while making these impressive performance gains. And since Apple prioritized battery life with the A15 chip, it could get a lot better.