Apple iPhone 12: The chipset to make smartphones smarter.
When Apple unveils its new iPhones, expect it to make a big deal of the fact they’re the first handsets in the world to be powered by a new type of chip.
This, we’ll likely be told, will let owners do things like edit 4K video, enhance high-resolution photos and play graphically-intensive video games more smoothly than was possible before while using less battery power.
The “five nanometre process” involved refers to the fact that the chip’s transistors have been shrunk down – the tiny on-off switches are now only about 25 atoms wide – allowing billions more to be packed in.
Effectively it means more brain power.
Travel back just four years, and many industry insiders doubted the advance could be delivered so soon.
That it has been, is in large part down to the ingenuity of a relatively obscure Dutch company – ASML.
About the Chip
A blueprint of the chip’s design is encoded into the light, he added.
by passing it through a mask and then shrinking it with lenses.
It then hits a light-sensitive coating on a silicon wafer, causing the chip’s design to be “printed”.
Only two chip manufacturers have put this to commercial use so far :
- Taiwan Semiconductor Manufacturing Company (TSMC) – sole supplier of the A14 to Apple for its latest iPhones, iPads, and Mac computers
- Samsung – which is making a new Qualcomm processor for Android phones, set to be formally unveiled in December.
All this matters because moving to 5nm is key to making our handsets smarter.
As chips advance, more tasks that used to be sent to remote computer servers for processing can be done locally.
We’ve already seen smartphones become capable of transcribing voice notes and recognising people in photos without the need for an internet connection.
Now even more complex “artificial intelligence” jobs become possible, potentially helping smartphones make better sense of the world around them.
“If you have a defined workload – say image processing, audio signal processing, video encoding or cryptography – the maths involved is very well defined,” explained Dr Cutress.
“And you can design the accelerator to do the task as fast as it can or to extend the device’s battery life.”
Apple has already claimed its A14 chip will do machine learning tasks “up to 10 times faster” than the A13.
Expect other smartphone brands to make similar claims as they move to 5nm.