Apple and TSMC unveil A20 Pro chip built on 2nm process for iPhone 18 Pro lineup

Apple and TSMC unveiled the A20 Pro on September 9 at Apple's "Surprise and Shine" event, the first iPhone chip built on TSMC's 2 nanometer (N2) process node, which will power the iPhone 18 Pro, iPhone 18 Pro Max, and the new iPhone Duo. Apple claims the A20 Pro delivers twice the sustained performance of the iPhone 16 Pro and 40% more than the iPhone 17 Pro, with dual 16-core Neural Engines totaling 32 cores for 2x the AI compute power of its predecessor. The chip also adds a redesigned 6-core CPU with two "super cores" claimed 20% faster than the A19 Pro's performance cores, a 7-core GPU with up to 40% higher performance than the A19 Pro, a 50% memory bandwidth boost, and wafer-level multi-chip module (WMCM) packaging; Apple did not disclose transistor counts or power consumption figures.

apple logo Apple and TSMC unveil A20 Pro chip built on 2nm process for iPhone 18 Pro lineup The first iPhone processor manufactured on TSMC's 2nm node promises double the AI compute power of its predecessor and 40% more sustained performance. Apple https://cryptobriefing.com/markets/apple/ just gave the semiconductor world its new benchmark. At the company’s “Surprise and Shine” event on September 9, Apple and TSMC revealed the A20 Pro, the first iPhone chip built on TSMC’s 2 nanometer N2 process node. It will power the iPhone 18 Pro, iPhone 18 Pro Max, and the new iPhone Duo. The jump from the previous generation isn’t incremental. Apple claims the A20 Pro delivers twice the sustained performance of the iPhone 16 Pro and 40% more than the iPhone 17 Pro. What’s actually inside the A20 Pro The chip packs a redesigned 6-core CPU. Two of those cores are what Apple calls “super cores,” claimed to be 20% faster than the performance cores in the A19 Pro. The remaining four are efficiency cores, each now paired with dedicated neural accelerators. On the graphics side, a 7-core GPU delivers up to 40% higher performance compared to the A19 Pro. The real headline, though, is the neural engine configuration. Apple doubled down, literally, with dual 16-core Neural Engines for a total of 32 cores. The result is 2x the AI compute power versus its predecessor. Memory bandwidth got a 50% boost thanks to what Apple describes as the widest memory interface ever used in an iPhone. The chip also introduces wafer-level multi-chip module WMCM packaging, a technique that places DRAM directly adjacent to the die. Apple says sustained performance improves by 40% over the iPhone 17 Pro as a direct result. Why the 2nm node matters beyond Apple TSMC’s 2nm process is among the most advanced commercial semiconductor technologies on the planet. Apple has historically been TSMC’s largest and earliest adopter of cutting-edge manufacturing, and the A20 Pro continues that pattern. Advanced packaging has become one of the hottest areas in chip manufacturing, with companies like TSMC, Intel, and AMD https://cryptobriefing.com/markets/amd/ all investing heavily in techniques that go beyond simply shrinking transistors. Apple’s adoption of wafer-level multi-chip modules in a phone, rather than just in data center chips, suggests this technology is ready for high-volume consumer electronics. The AI device race is accelerating Apple’s decision to double its neural engine core count to 32 reflects a strategic bet that the next wave of smartphone differentiation will be driven by on-device AI rather than cloud-dependent processing. Running AI models locally on a phone requires enormous amounts of specialized compute, and the more capable the on-device silicon, the more complex the AI features Apple can offer without routing data through external servers. Apple didn’t disclose transistor counts or specific power consumption figures for the A20 Pro, which means independent benchmarks will be the real test of whether these claimed improvements hold up under real-world conditions. Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our Editorial Policy https://cryptobriefing.com/editorial-policy/ .