The Apple A15 Bionic is a 64-bit ARM-based system on a chip (SoC), designed by Apple Inc. It was announced on September 14, 2021 and released the following September 24th for the 6th-generation iPad mini and the iPhone 13 series.[1]
Design
The A15 contains 6 CPU cores, including 2 high-performance cores and 4 high-efficiency cores which provide a 40% performance improvement over the Apple A12. Its 16 Neural Engine cores can perform up to 15.8 trillion operations per second.[1] It operates at 3.2 GHz on the iPhone 13 series and 2.9 GHz on the 6th-generation iPad mini.[2]
The GPU contains 5 cores in the 6th-generation iPad mini, iPhone 13 Pro and 13 Pro Max. However, the entry-level iPhone 13 and 13 mini are advertised with only 4 GPU cores, indicating the possible use of binning.[1] Benchmarks indicate improved GPU performance over previous generations. However, benchmarks submitted after continued use indicate the presence of throttling.[3]
Products with the Apple A15
- iPad mini (6th generation) — 4 GB on package, 5 GPU cores
- iPhone 13 — 4 GB on package, 4 GPU cores
- iPhone 13 mini — 4 GB on package, 4 GPU cores
- iPhone 13 Pro — 6 GB on package, 5 GPU cores
- iPhone 13 Pro Max — 6 GB on package, 5 GPU cores
References
- ↑ 1.0 1.1 1.2 Apple Event — September 14 by Apple, YouTube. 2021-09-14.
- ↑ Everything We Know About the 2022 iPad Air Coming in March by Juli Clover, MacRumors. 2022-02-07.
- ↑ 'iPhone 13' A15 chip performance continues dominance over Android rivals by Wesley Hilliard, AppleInsider. 2021-09-06.
External links
- Apple A15 specs at GadgetVersus
- Apple A15 Bionic at NotebookCheck
- Apple A15 at Wikipedia
Articles
- Apple A15 Die Shot and Annotation – IP Block Area Analysis by Dylan Patel at SemiAnalysis (2021-10-05)
- Report suggests Apple's A15 Bionic lacks significant CPU upgrades due to chip team brain drain by Mikey Campbell at AppleInsider (2021-09-15)
- Apple CPU Gains Grind To A Halt And The Future Looks Dim As The Impact From The CPU Engineer Exodus To Nuvia And Rivos Starts To Bleed In by Dylan Patel at SemiAnalysis (2021-09-14)
