Every year, we look at the trajectory of Apple’s A-series (which premiered in the iPhone and is often used in the iPad) to get an idea of how the next chip will perform. By combining what we know about state-of-the-art chip manufacturing and packaging processes and Apple’s past improvements, we can generally understand what’s coming from the next system-on-a-chip (SoC).
This year, as we look ahead to the A16, making that call is harder than ever. The strongest rumors suggest that the A16 will not move forward to a major new node in the manufacturing process, and that it will only appear in the iPhone 14 Pro and Pro Max – with the regular iPhone 14 models using the same improved A15 in the iPhone 13 Pro model. What’s more, we need to consider how Apple uses these designs. Previous A-series chips might get a larger “X” version designed for the iPad Pro, but now that Apple is scaling the design to an entire line of M-series chips designed for Macs, including very powerful computers like the Mac Studio, some of these extra cores may be making their way into the iPhone’s processor.
So let’s dive in and try to predict the A16, perhaps with a slightly larger dose of salt than usual. We’re making our best guesses here, but there are more variables than in previous years, and we’re often surprised by at least a few details when Apple reveals its new iPhone chips.
Pro only chip
Broadly speaking, the latest A-series SoC is powering the entire new iPhone lineup in the fall. Sometimes the Pro models have a bit more RAM, but it’s basically the same chip. With the iPhone 13 line, Apple changed that a bit. The iPhone 13 Pro and Pro Max have more RAM (6GB compared to 4GB in the iPhone 13 and 13 mini), but also have five GPU cores, while the non-pro models have four. It’s almost certainly the same chip design with one of the cores disabled, a common tactic to improve yields with advanced chip manufacturing, but this is the first time Apple has made this differentiation in iPhone models. (Perhaps confusingly, both versions of the chip are still just called “A15” and you have to dive into the tech specs to see the difference.)
The iPhone 13 Pro models have a slightly better A15 chip, but the iPhone 14 Pro may have a completely different chip than the iPhone 14.
IDG
With the A16, most tipsters are suggesting that Apple only plans to include the new chip in the iPhone 14 Pro and Pro Max. The non-pro iPhone 14 models are said to get the A15, albeit a 6GB, 5-core version found in the iPhone 13 Pro models. It’s not yet clear whether Apple will essentially rename this chip “A16”, but it means a small but respectable upgrade for the non-Pro models, even though they don’t get the new SoC.
This is particularly important because it can help signal what A16’s design priorities might be. If it’s only going to appear on significantly more expensive iPhones (and rumors suggest starting prices could go up by $100), perhaps it can afford to be bigger and more expensive, with a different focus. In addition to other features like an always-on display, the iPhone 14 Pro models are said to feature a new 48MP rear wide-angle camera. This can require much more image processing and machine learning power.
New old manufacturing process
The most plausible rumors say that the A16 will be manufactured on a 5nm manufacturing process, just like the A15 and A14 were. It’s not often that Apple works three years in a row on the same process node. Some rumors suggest it will be manufactured with a 4nm process, but TSMC doesn’t actually have one. What it has is N4P, codename for the third generation of the 5nm high-efficiency manufacturing process. N4P offers an 11 percent improvement in performance, a 22 percent improvement in power efficiency, and a 6 percent improvement in density over the original 5nm “N5” manufacturing process.
The A15 was built using the 5n process and the A16 is likely to follow suit.
An apple
It was the A14 that was made with the N5 process, and the A15 used the second-generation N5P process, which offers some minor improvements. In other words, Apple won’t get much performance or power efficiency from the third-generation 5nm manufacturing process. It’s certainly not the leap we’re expecting from TSMC’s 3nm process, which is expected to be ready this year, but it’s likely not in time for Apple to produce tens of millions of them for the iPhone 14 Pro launch this fall. When we move to 3nm, Apple will be able to squeeze about 70 percent more transistors into the same space, with significant power savings.
Despite the fact that this third-generation 5nm process isn’t dramatically better than the second-generation one Apple used last year, we think the A16 will have to get significantly bigger. We’re probably looking at a whopping 18 billion to 20 billion transistors, compared to the A15’s 15 billion. Most of that will likely go to the Neural Engine for machine learning, the image signal processor and video encoder and decoders, and some general CPU performance improvements.
CPU performance
It doesn’t make sense for Apple to expand beyond the current basic iPhone configuration. Two high-performance cores and four high-performance cores are more than enough for a platform that almost always runs one app at a time, full-screen, with some light background processes and the like from other apps. At least for now.
Apple is probably more interested in improving energy efficiency and faster processing cores. High-performance cores, in particular, are likely to be an area of focus, as the company is counting on them to really boost home Mac performance when the CPU core design ends up in M-series chips. The M3 line will likely have the same CPU core designs found in the A16, just scaled to a higher core count and perhaps with larger caches.
Don’t be surprised if the A16’s single-core processor performance generates a Geekbench 5 score around 2,000.
IDG
A rumor from an unverified source on Twitter recently reported that the A16 tested 42 percent faster than the A15. This is almost certainly an unreasonable expectation. I think Apple will switch to LPDDR5 this year, which should improve memory bandwidth, and some other improvements combined with higher peak frequencies could lead to a 15% performance improvement for Apple.
If Apple manages to continue recent trends, the multi-core Geekbench 5 score could be around 5700.
IDG
There is one exception worth mentioning. When ARM finalized the ARMv9 instruction set last year, there was some speculation as to whether Apple would be the first to incorporate it into the design. I don’t think it was done quite in time, but Apple likes to be on the bleeding edge here. The ARMv9 instruction set is now available in several core ARM-licensed designs (Cortex-A710, Coretex-A510). They will be used in chips such as the Qualcomm Snapdragon 7 Gen 1, which will appear in Android phones later this year.
Apple does not license core designs from ARM; it makes its own designs that are compatible with the ARM instruction set. The A15 supports the ARMv8.5-A instruction set, and moving to ARMv9 could have a positive effect on some very specific performance areas. In particular, wide SIMD tip operations (which are usually more of a high-end desktop thing) should be much faster and more flexible.
The A16 may support the same ARMv9 compatible design as the Snapdragon 7 Gen 1.
Qualcomm
ARM talked about performance improvements of 30 percent at the same time as it promoted the ARMv9 instructions, but the company was specifically talking about its own licensed processor core designs there. There doesn’t necessarily have to be anything in the instruction set to increase performance in this way, and Apple already uses its own designs with its own customizations and extensions.
In short, the A16 may be Apple’s first ARMv9-compatible design, and one of the first on the market at all, and it may make some very specific types of CPU operations faster, but overall performance is unlikely to skyrocket.
GPU performance
Apple continues to invest heavily in GPU performance, despite the fact that iPhones are usually miles ahead of other premium smartphones in this area. Now that the company is scaling its chip designs to laptops and desktops, with powerful integrated graphics replacing AMD’s discrete GPUs, that’s even more important.
Without a huge boost in transistor density from moving to the next major technology node, Apple probably can’t afford to increase the number of GPU cores much. We think the A16 will have either five GPU cores (like the full A15 configuration) or six, but probably not more. However, architectural improvements and the move to LPDDR5 RAM for increased memory bandwidth will help boost graphics performance.
3DMark’s Sling Shot test is a bit out of date, and performance improvements are starting to trickle down.
IDG
We think it’s reasonable to expect a 25 percent to 30 percent improvement in GPU performance, roughly in line with the last few A-series processors. You’ll see this especially in benchmarks and tests that are currently limited by memory bandwidth.
The more advanced 3DMark Wild Life test better represents the improvements in the graphics architecture. Expect the A16 to achieve between 85 and 90 fps.
IDG
Apple will likely need to include hardware to accelerate ray tracing in a future GPU. It is quickly becoming the standard for laptops and desktops, and the M-series chips derive their architectures from Apple’s A-series. (Or more accurately, they are all designed as a family of products sharing basic architectural elements.) Even the main mobile chip designs from Qualcomm and ARM have ray tracing on their…
Add Comment