/
Navigation
Chronicles
Browse all articles
Explore
Semantic exploration
Research
Entity momentum
Nexus
Correlations & relationships
Story Arc
Topic evolution
Drift Map
Semantic trajectory animation
Posts
Analysis & commentary
Pulse API
Tech news intelligence API
Browse
Entities
Companies, people, products, technologies
Domains
Browse by publication source
Handles
Browse by social media handle
Detection
Concept Search
Semantic similarity search
High Impact Stories
Top coverage by position
Sentiment Analysis
Positive/negative coverage
Anomaly Detection
Unusual coverage patterns
Analysis
Rivalry Report
Compare two entities head-to-head
Semantic Pivots
Narrative discontinuities
Crisis Response
Event recovery patterns
Connected
Search: /
Command: ⌘K
Embeddings: large
TEXXR

Chronicles

The story behind the story

days · browse · Enter similar · o open

Apple's A15 is a 5nm SoC with two performance CPU cores and four efficiency cores, a 16-core Neural Engine capable of 15.8 trillion operations/second, and more

Apple

Context & Ripple Effects

The A15's spec sheet is the latest step in a design lineage Apple has been building since the A10 Fusion introduced the split of high-performance and high-efficiency cores in 2016, and which the seven-core-GPU A12X pushed further at 7nm in 2018. What makes this generation different is what the numbers emphasize: a 16-core Neural Engine rated at 15.8 trillion operations per second sits front and center.

The timing matters because Apple broke its own pattern around this launch — for the first time it reported no headline CPU gains for the A15, which SemiAnalysis linked to an exodus of Apple silicon engineers to Nuvia and Rivos. The subsequent independent review bore that out: efficiency improved substantially, but speed gains over the A14 ranged only from 2.5% to 37% depending on workload.

First-order effects

  • iPhone 13-series buyers get a chip whose gains are concentrated in efficiency and on-device neural throughput (15.8 TOPS) rather than raw CPU speed — a first in Apple's own reporting.
  • Apple's silicon roadmap visibly reweights toward the Neural Engine, making on-device AI throughput the flagship metric where CPU generational claims used to be.

Second-order effects

  • Nuvia and Rivos, the destinations named in the engineer-exodus reporting, gain exactly the Apple-trained heterogeneous-core expertise needed to attack the same performance-per-watt territory, tightening the market for senior SoC talent.
  • With CPU deltas shrinking to workload-dependent percentages, competitors' phone chips are forced to compete on neural-engine TOPS and battery life instead of clock-speed comparisons.

Third-order effects

  • If the pattern holds, mobile silicon differentiation migrates structurally from CPU generations to integrated AI compute — the direction Apple itself later confirmed by shipping the M5 with a Neural Accelerator built into every GPU core.
  • A sustained talent drain from Apple's chip team would make silicon engineering a contested labor market rather than a proprietary moat, spreading big-little-plus-NPU design competence across more vendors.

The trend: Mobile SoC competition is pivoting from annual CPU-speed claims to efficiency and on-device neural throughput, while the engineers who built that playbook become the industry's most contested resource.

Discussion

  • @kevinctofel Kevin C. Tofel on x
    Think I'll be sticking with my iPhone 12. 15 billion transistors in the 13 sounds nice but I can get by with as few at 10 or 11 billion. https://www.apple.com/...
  • @brianroemmele Brian Roemmele on x
    Consider the entire world computing power in 1969 was 66,700 million operations per second. Just a single Apple A15 is 15.8 trillion operations per second. This is how far we have processing power has moved in about half a century. https://twitter.com/... https://twitter.com/...
  • @dylan522p Dylan Patel on x
    Apple's CPU team is struggling and delivered no CPU gains for the 1st time ever! The next major core architecture was likely delayed due to personnel overturn from Nuvia exodus With a new exodus to Rivos and stronger competition, future doesn't look bright https://semianalysis.su…
  • @adamnash Adam Nash on x
    The Apple A15 Bionic chip is just amazing. 15 billion transistors at 5nm. 15.8 trillion operations per second. During my 1st internship (@NASA) we were working on scaling software to take advantage of a future w/ 1 teraflop supercomputers... now you have 15x that in a phone. 😳 ht…
  • @brianroemmele Brian Roemmele on x
    Meet the Apple A15. 15.8 trillion operations per second. The computing power of the entire NASA Apollo 11 Moon Mission in 15 seconds. https://twitter.com/...
  • @benedictevans Benedict Evans on x
    Transistors in the iPhone 13 SoC: 15bn Transistors in all the CPUs in all the Macs sold in 1984: ~25bn.
  • @zhugeex Daniel Ahmad on x
    Apple silicon continues to be the most impressive aspect of these events, not just because of what it can enable in existing products, but because it can help Apple become a leader in new product categories. https://twitter.com/...