Review of Apple's A15 shows big efficiency gains and adequate speed improvements, which range from 2.5% to 37% over A14 depending on the type of workload
A few weeks ago, we've seen Apple announce their newest iPhone 13 series devices, a set of phones being powered by the newest Apple A15 SoC.
Context & Ripple Effects
The first benchmarks out of the iPhone 13 launch painted A15 as a broad 10-18% CPU and ~15% GPU step over A14. AnandTech's full review sharpens that picture: the headline numbers mask a wide spread, from a near-nothing 2.5% on some workloads to 37% on others, with the biggest wins coming on efficiency rather than peak speed. That fits the silicon itself — Apple's own A15 configuration leans on four efficiency cores alongside two performance cores.
First-order effects
- iPhone 13 buyers get most of their upgrade value in battery life and thermals rather than raw responsiveness, since the review finds speed gains vary from 2.5% to 37% by workload.
- Apple's marketing around A15 now has to survive workload-by-workload scrutiny instead of single-number benchmark comparisons like the early MacRumors figures.
Second-order effects
- The efficiency-first result sets up a hard comparison for the next generation: when the A16 in the iPhone 14 turned out only marginally faster than A15 despite a smaller process node, the plateau AnandTech flagged became the story of Apple's chip cadence.
- Reviewers and buyers shift weight toward performance-per-watt measurements, devaluing the peak-throughput benchmarks that had driven year-over-year phone comparisons.
Third-order effects
- If efficiency keeps outrunning peak-speed gains, smartphone SoC competition reorganizes around sustained workloads — gaming, camera, on-device intelligence — where the efficiency-led design language Apple has pushed since A13 compounds across generations.
- With node shrinks yielding thinner returns (the 5nm-to-4nm A16 delta), architectural efficiency rather than process scaling becomes the primary lever for mobile performance growth.
The trend: Mobile silicon competition is rotating from peak clock-speed gains toward performance-per-watt, as process-node improvements flatten and efficiency-core architectures carry each generation.