ARM announces the Cortex-A77 CPU and Mali-G77 GPU touting performance improvements, and a more energy efficient and powerful machine learning processor
Arm, the company that designs the basic chip architecture for most of the world's smartphones, today announced the launch of its next suite of designs for premium phones.
Context & Ripple Effects
This is the fourth installment of Arm's now-annual spring reveal of premium phone silicon. The arc has been consistent: the Cortex-A73 and Mali-G71 generation pitched mobile VR, the Cortex-A75 cycle pivoted to efficient machine learning, and last year's Cortex-A76 delivered a claimed 35% performance jump. The Cortex-A77 and Mali-G77 continue that cadence while deepening the dedicated machine-learning processor thread.
What matters here is less any single spec than the compounding effect: each yearly refresh resets the performance-and-efficiency floor that every flagship Android SoC licensee builds against, and the successor Cortex-A78 and Mali-G78 designs announced a year later confirm this became a fixed product rhythm rather than a one-off.
First-order effects
- Arm's licensees building 2020-flagship system-on-chips get a new CPU, GPU, and ML processor template to integrate, with energy efficiency as the stated constraint alongside raw performance.
- Premium Android phone makers gain a ready-made architecture path, shortening their own design cycles since the core IP is licensed rather than built.
Second-order effects
- Rival silicon vendors must match the combined CPU-plus-GPU-plus-ML package, not just peak clock speeds, because Arm is now selling machine-learning throughput as part of the base platform.
- The efficiency gains pressure thermal and battery budgets across the premium tier, letting phone makers chase sustained performance instead of headline benchmarks.
Third-order effects
- If the pattern holds, dedicated ML compute stops being an optional accelerator and becomes a default component of the standard mobile architecture, with Arm's annual cadence effectively setting the industry-wide upgrade clock for flagship phones.
- Consolidating AI capability into licensed IP narrows differentiation at the high end, pushing phone makers to compete on system-level integration rather than raw chip specs.
The trend: Mobile chip design is settling into an annual Arm-led rhythm where machine-learning processors ship as standard architecture, not add-ons.