iPhone 16 Series: TSMC’s N3E Tech Powers AI-Driven Upgrades
Apple is preparing the A18 series chips for this year’s iPhone 16 models, with a significant focus on upgrading the neural network engine. Rumors suggest its computational power might even surpass that of the M4 chip, which is crucial for handling generative AI functions. If the iPhone 16 series can run AI models locally, a more powerful neural network engine will better accommodate various AI tasks.
According to Wccftech, the iPhone 16 series is nearing production, with the entire lineup expected to utilize TSMC’s second-generation 3nm technology, the N3E process. Supply chain insiders reveal that Apple anticipates shipping between 90 million to 100 million units of the iPhone 16 series. They attribute the primary drivers of this demand to upgrades in generative AI and other features, making the new devices more appealing. To better meet market demand, Apple has increased the order volume for the A18 series chips.
Reports indicate that Nvidia has accepted a price hike of about 10% for TSMC’s 4nm orders next year, while Apple’s 3nm orders will remain priced as they are, thanks to their ongoing collaboration on advanced chips. Other clients’ order prices might not be finalized until late Q3 of this year.
It is understood that the A18 series chips might be divided into two variants: the A18 for the iPhone 16 and iPhone 16 Plus, and the A18 Pro for the iPhone 16 Pro and iPhone 16 Pro Max, with the latter offering superior performance. Utilizing TSMC’s enhanced semiconductor process, these chips will boast improved efficiency and performance.
Reports also suggest that the entire iPhone 16 series will come with 8GB of memory. While this hardware configuration may be insufficient for running generative AI functions entirely on-device, the A18 series chips might be larger to accommodate the upgraded neural network engine and increased cache. However, the GPU will maintain the same 6-core configuration as the current A17 Pro, somewhat limiting enhancements in graphics performance.
