Fact-checked by Grok 2 weeks ago

Apple A11

The Apple A11 Bionic is a 64-bit ARM-based system on a chip (SoC) designed by Apple Inc. and manufactured by TSMC.^[1] Announced on September 12, 2017, it powers the iPhone 8, iPhone 8 Plus, and iPhone X, marking the first Apple mobile processor to incorporate a dedicated Neural Engine for machine learning and a fully custom-designed GPU.^[2] Built on a 10 nm process with 4.3 billion transistors, the A11 Bionic delivers significant advancements in performance and efficiency over its predecessor, the A10 Fusion, while enabling features like Face ID facial recognition and augmented reality experiences.^[1]^[3] The A11 Bionic's CPU adopts a heterogeneous six-core design, comprising two high-performance Monsoon cores clocked at up to 2.39 GHz and four high-efficiency Mistral cores.^[1] The Monsoon cores provide 25% faster single-threaded performance than the A10's cores, while the Mistral cores offer 70% greater speed at the same power level, with a second-generation performance controller allowing all six cores to operate simultaneously for up to 70% better multi-threaded workloads.^[2] Its three-core GPU, also custom-built by Apple, achieves 30% higher graphics performance than the A10's GPU while consuming half the power, supporting immersive 3D gaming and AR applications.^[2] Complementing these is the dual-core Neural Engine, capable of 600 billion operations per second, which accelerates on-device machine learning tasks such as real-time image processing and neural network inference for Animoji and other AI-driven features.^[2] Overall, the A11 Bionic contributed to up to two hours longer battery life in the iPhone X compared to the iPhone 7, balancing high performance with energy efficiency.^[2]

Background

Announcement

The Apple A11 Bionic system on a chip was publicly announced on September 12, 2017, during Apple's keynote event at the newly opened Steve Jobs Theater in Apple Park, Cupertino, California. The announcement occurred alongside the reveal of the iPhone 8, iPhone 8 Plus, and iPhone X, marking the debut of these devices powered by the new processor. Apple executives, including CEO Tim Cook and Senior Vice President of Worldwide Marketing Phil Schiller, presented the A11 Bionic as a foundational upgrade for the next generation of iPhones, emphasizing its role in enabling advanced mobile experiences.^[4]^[2] The chip was introduced under the name "A11 Bionic," a designation chosen to reflect its inspiration from biological systems, particularly in achieving superior efficiency and performance akin to neural processes in the human brain. Apple highlighted it as the company's first processor with dedicated neural hardware in the form of a dual-core Neural Engine, capable of performing up to 600 billion operations per second for machine learning tasks. Performance claims included a six-core CPU design that delivered up to 70% faster multi-threaded workloads compared to the A10 Fusion chip in the iPhone 7 series, with specific improvements such as 25% faster performance cores and 70% faster efficiency cores.^[2]^[5] In the marketing context, Apple positioned the A11 Bionic as a significant advancement in mobile computing, specifically tailored to support emerging applications in augmented reality (AR) and machine learning. The chip's integration with ARKit was showcased to enable immersive AR experiences through features like world tracking, scene recognition, and real-time graphics rendering at 60 frames per second, while the Neural Engine powered on-device AI capabilities such as facial recognition for Face ID and animated emojis (Animoji). This framing underscored Apple's vision of the A11 as a "bionic" leap forward, bridging hardware efficiency with sophisticated software ecosystems for future innovations in personal computing.^[4]^[2]

Development

The development of the Apple A11 Bionic SoC began approximately three years prior to its 2017 release, aligning with Apple's practice of planning silicon architectures well in advance, during the era of the iPhone 6 and A8 chip.^[6] As the successor to the A10 Fusion, the project involved an expanding team under Johny Srouji, Apple's senior vice president of hardware technologies, focusing on advancing custom designs to enhance performance and efficiency for mobile devices.^[6] This effort built on prior iterations, with design phases emphasizing integration across CPU, GPU, and emerging AI components to support features like augmented reality and facial recognition. A major innovation was the introduction of fully custom CPU cores named Monsoon and Mistral, comprising two high-performance Monsoon cores and four efficiency-oriented Mistral cores, all based on a heavily customized ARMv8-A architecture.^[7] Apple collaborated with ARM Holdings for the foundational instruction set but implemented extensive modifications to optimize for power efficiency and task-specific workloads, moving beyond off-the-shelf designs.^[7] The A11 also marked the debut of a dedicated Neural Engine, a specialized accelerator for on-device machine learning tasks, enabling real-time processing for applications such as Face ID while prioritizing privacy by minimizing cloud reliance.^[8] Engineering challenges centered on transitioning to greater in-house control over silicon components to lessen dependence on third-party intellectual property, including the replacement of Imagination Technologies' PowerVR GPU with Apple's first custom graphics core.^[9] This shift required iterative design adjustments to balance complexity with the 10nm fabrication process, alongside developing a new custom video encoder for efficient 4K encoding.^[7] The team's collaborative approach across hardware and software groups addressed last-minute feature requests, ensuring seamless integration without compromising energy efficiency.^[6]

Architecture

Central processing unit

The central processing unit (CPU) of the Apple A11 Bionic is a hexa-core processor consisting of two high-performance Monsoon cores and four high-efficiency Mistral cores, marking Apple's first implementation of a six-core design in its mobile SoCs.^[2] This configuration leverages a custom implementation of the ARMv8-A instruction set architecture, utilizing a big.LITTLE hybrid approach where the high-performance cores handle demanding tasks and the efficiency cores manage lighter workloads to optimize power consumption.^[1] The Monsoon cores operate at a clock speed of 2.39 GHz, while the Mistral cores run at 1.42 GHz, enabling dynamic scaling based on computational needs.^[10] Apple states that the two Monsoon performance cores deliver 25 percent faster processing compared to the performance cores in the preceding A10 Fusion chip, while the four Mistral efficiency cores provide up to 70 percent faster performance than the A10's efficiency cores.^[2] Overall, the A11's CPU achieves up to 70 percent greater performance in multi-threaded workloads relative to the A10 Fusion, thanks to a second-generation performance controller that enhances thread management.^[2] This design contributes to the chip's industry-leading energy efficiency, allowing devices to sustain longer battery life under mixed workloads compared to prior generations.^[2] The CPU incorporates per-core L1 caches of 64 KB for instructions and 64 KB for data, supplemented by a shared 8 MB L2 cache to reduce latency and improve data access speeds across cores.^[3] These cache structures support the hybrid architecture's goals by minimizing power draw during efficiency-focused operations while maintaining high throughput for performance-intensive applications.^[11]

Graphics processing unit

The Apple A11 Bionic features the company's first fully custom-designed graphics processing unit (GPU), a three-core architecture developed in-house rather than licensed from Imagination Technologies' PowerVR series used in prior generations.^[2]^[12] This shift allowed for tighter integration with the SoC's tile-based deferred rendering (TBDR) pipeline, enabling more efficient memory usage and reduced power draw during rendering. The GPU supports Apple's Metal 2 API, which introduces advanced shading techniques such as tile shading and image blocks to optimize compute workloads directly on the GPU's high-bandwidth tile memory.^[13] In terms of performance, the A11 GPU operates at a clock speed of approximately 1,066 MHz and delivers up to 30 percent faster graphics rendering compared to the A10 Fusion's GPU, with a peak throughput of around 409 gigaflops (GFLOPS) in single-precision floating-point operations.^[2]^[10] This enhancement stems from architectural improvements in geometry processing and tessellation, facilitated by Metal 2's support for efficient patch-based subdivision and raster order groups, which minimize overdraw and enhance scene complexity handling. The design is particularly optimized for augmented reality (AR) applications via ARKit, supporting immersive 3D gaming and AR experiences at 60 frames per second (fps) with reduced latency.^[14]^[2] Power efficiency represents a key advancement, with the A11 GPU achieving the equivalent performance of the A10 GPU while consuming half the power, translating to approximately 30 percent better performance per watt overall.^[7] This is accomplished through the TBDR architecture's ability to defer shading until geometry is resolved per tile, avoiding unnecessary computations and aligning with the demands of mobile graphics-intensive tasks like high-frame-rate video playback and AR rendering.

Neural processing unit

The Neural Processing Unit in the Apple A11 Bionic, branded as the Neural Engine, marked Apple's first dedicated hardware accelerator for machine learning within a smartphone system-on-chip (SoC).^[2] Introduced in 2017 alongside the iPhone X, this unit enabled on-device AI processing without relying on cloud services, a significant advancement for privacy and performance in mobile devices.^[15] The Neural Engine features a dual-core architecture capable of executing up to 600 billion multiply-accumulate (MAC) operations per second, optimized for real-time neural network inference.^[2] This hardware is tightly integrated into the A11 SoC, allowing seamless data flow and low-latency execution of machine learning tasks directly on the device.^[1] By offloading computations from the CPU and GPU, it enhances energy efficiency for AI workloads compared to general-purpose processors.^[16] Designed specifically for Apple's Core ML framework, the Neural Engine supports essential tasks such as face detection, image recognition, and voice processing, accelerating applications like computational photography and user authentication.^[2] For instance, it powers features including Face ID on the iPhone X, performing secure facial recognition in real time.^[2] This pioneering implementation represented Apple's strategic entry into specialized hardware for machine learning, setting the foundation for subsequent generations of Neural Engines in later SoCs.^[15]

Image signal processor

The Image Signal Processor (ISP) in the Apple A11 Bionic is a custom-designed hardware component optimized for advanced photography and video tasks, supporting dual 12-megapixel camera setups in compatible devices for features like depth-based effects.^[2] It incorporates dedicated hardware for high dynamic range (HDR) processing and noise reduction, enabling superior image quality with vibrant colors, deeper details, and reduced artifacts in challenging lighting.^[4] The ISP performs advanced pixel processing and wide color capture, resulting in faster low-light autofocus and enhanced HDR photo rendering.^[17] Key capabilities include real-time depth mapping using dual-camera data to create natural bokeh effects in Portrait mode, along with facial landmark detection for precise subject isolation.^[4] For video, it supports 4K recording at 60 frames per second with improved optical image stabilization on both wide and telephoto lenses, delivering smoother footage during motion.^[2] The ISP also enables real-time lighting estimation, which aids augmented reality applications by analyzing scene illumination on the fly.^[4] The A11 integrates a dedicated video encoder for High Efficiency Video Coding (HEVC) compression, performing real-time image and motion analysis to optimize quality while minimizing file sizes.^[2] An embedded motion coprocessor complements this by handling sensor fusion for accurate tracking in video and photo modes. In synergy with the Neural Processing Unit, the ISP leverages machine learning for AI-enhanced imaging, such as scene recognition and object identification during capture.^[4] Power management is optimized through efficient hardware acceleration, enabling intensive computational photography without excessive battery drain, as evidenced by sustained all-day usage in imaging-heavy scenarios.^[2]

Manufacturing

Fabrication process

The Apple A11 Bionic system on a chip was exclusively fabricated by Taiwan Semiconductor Manufacturing Company (TSMC).^[18] TSMC produced the A11 using its 10 nm FinFET process node, designated as N10 or 10FF, marking Apple's inaugural adoption of a 10 nm manufacturing technology following the 16 nm node used in the preceding A10 Fusion.^[19]^[20] Production of the A11 ramped up with mass manufacturing commencing in April 2017, enabling a notably smaller die size relative to the 16 nm A10 while accommodating enhanced integration.^[19]^[21] This 10 nm process delivered approximately twice the transistor density per area compared to TSMC's 16 nm FinFET technology, facilitating greater circuit complexity within a compact footprint.^[20]^[22] Additionally, it provided 15% higher performance at equivalent power levels and up to 35% improved power efficiency over the 16 nm node, contributing to better overall energy management in mobile applications.^[20]^[22]

Die and transistor count

The Apple A11 Bionic die measures approximately 87.66 mm².^[1] It contains 4.3 billion transistors, marking a substantial increase from the 3.3 billion transistors in the preceding A10 Fusion.^[1] This transistor density benefits from the 10 nm fabrication process.

Features

Machine learning integration

The A11 Bionic's Neural Engine is tightly integrated with Apple's Core ML framework, which optimizes machine learning models for efficient execution on Apple silicon, including on-device inference for tasks like image classification and natural language processing.^[23] This integration allows developers to deploy pre-trained models directly in iOS apps without relying on cloud services, ensuring low-latency responses and enhanced user privacy by keeping data processing local to the device.^[23] While full on-device training capabilities were expanded in later iOS versions, the A11 enables robust inference for a wide range of neural network architectures.^[24] Key software features powered by this integration include Animoji on the iPhone X, where the Neural Engine processes real-time facial expressions captured by the TrueDepth camera to animate emoji characters in iMessage.^[2] Improved Siri processing benefits from the Neural Engine's acceleration of speech recognition and contextual understanding tasks, enabling faster and more accurate voice interactions.^[8] Additionally, ARKit leverages the hardware for augmented reality applications, supporting features like world tracking and scene understanding through efficient machine learning-based motion and environmental analysis.^[2] This machine learning integration contributes to broader ecosystem benefits, such as reduced latency in AI-driven tasks compared to cloud-dependent alternatives, allowing seamless experiences in apps and system features.^[23] Local processing also enhances privacy by minimizing data transmission to external servers, aligning with Apple's emphasis on on-device computation for sensitive operations like facial recognition.^[2] For developers, Core ML provides straightforward APIs to import, convert, and integrate models from frameworks like TensorFlow or PyTorch into Swift or Objective-C code, with automatic hardware acceleration via the Neural Engine for optimal performance on A11-equipped devices.^[24] Tools like Core ML Tools facilitate model optimization and quantization, enabling custom ML features in apps without deep hardware expertise.^[25]

Security features

The Apple A11 Bionic features a Secure Enclave Processor (SEP), an isolated coprocessor designed to handle sensitive operations such as the generation and storage of encryption keys, processing of biometric data, and secure boot processes.^[26] This SEP runs a customized version of the L4 microkernel known as sepOS and includes a dedicated True Random Number Generator (TRNG) for creating cryptographically secure keys, including 256-bit unique device identifiers (UID) fused during manufacturing that are inaccessible to the main application processor.^[27] The Secure Enclave establishes a hardware root of trust through its Boot ROM, which verifies the integrity of sepOS and subsequent boot stages, ensuring only authenticated code executes during device startup.^[26] Key enhancements in the A11's Secure Enclave include a memory protection engine with AES encryption, CMAC authentication, and anti-replay capabilities for its dedicated DRAM region, providing isolated memory for sensitive operations that prevents exposure to the main CPU.^[27] It supports biometric authentication by securely storing and processing Touch ID and Face ID data—such as mathematical representations of facial features generated via the integrated Secure Neural Engine—without ever transmitting this information off-device.^[28] For Face ID on devices like the iPhone X, the Secure Enclave protects the biometric templates and enables fallback to Touch ID where applicable, tying authentication to user-specific passcode-derived keys.^[26] The A11's Secure Enclave incorporates protections against side-channel attacks, including differential power analysis (DPA) countermeasures in its AES engine and dedicated hardware monitors for power and clock anomalies to detect tampering attempts.^[27] This hardware isolation extends to key management, where file encryption keys for data protection classes (e.g., Class A for complete protection) are generated and wrapped using the UID, ensuring they cannot be exported or accessed by unauthorized software.^[26] Integration with iOS 11 and later versions leverages the Secure Enclave for enhanced data protection features, such as per-file encryption and secure keychain storage, which complement iOS app sandboxing by providing hardware-backed keys for isolating user data and preventing unauthorized access during runtime.^[26] These mechanisms ensure that sensitive operations, like Apple Pay transactions or iCloud keychain syncing, remain confined within the tamper-resistant environment of the Secure Enclave.^[27] In August 2020, security researchers from the Pangu Team disclosed an unpatchable vulnerability in the Secure Enclave of A7 to A11 processors, which could potentially allow the extraction of device encryption keys if an attacker has physical access to the device. Apple has not issued a patch for this hardware-level issue.^[29]

Devices

iPhone 8 and 8 Plus

The iPhone 8 and iPhone 8 Plus were released on September 22, 2017, marking the first devices to incorporate the Apple A11 Bionic system on a chip.^[4] These models introduced several features enabled by the A11, including wireless charging via Qi standard compatibility, which leverages the chip's power management for efficient inductive charging up to 7.5W.^[30] The A11 also powers the True Tone display technology, which dynamically adjusts screen color temperature based on ambient lighting conditions processed through the device's sensors.^[30] Additionally, camera enhancements, such as improved low-light performance and faster autofocus on the 12MP rear sensors, rely on the A11's integrated image signal processor for real-time adjustments.^[4] Configuration-wise, the iPhone 8 features 2 GB of LPDDR4X RAM paired with the A11 Bionic, while the iPhone 8 Plus includes 3 GB of the same RAM type to support its dual-camera setup.^[31] Both models were available in storage options of 64 GB, 128 GB, and 256 GB using NVMe flash.^[32] These specifications enabled smoother multitasking and app performance compared to prior generations, with the A11's hexa-core CPU contributing to overall responsiveness. As the first smartphones to demonstrate the A11's power efficiency, the iPhone 8 and 8 Plus delivered all-day battery life under typical usage, with the iPhone 8 offering up to 12 hours of internet browsing^[33] and the iPhone 8 Plus up to 13 hours,^[32] thanks to optimized power allocation in the chip. The devices received iOS updates through version 16, ensuring long-term software support for security and feature enhancements.^[34] Positioned as the base models in Apple's 2017 lineup, they emphasized incremental performance upgrades over the iPhone 7, such as the faster A11 processor and new display capabilities, appealing to users seeking reliable evolution without radical design changes.^[35] The iPhone 8 and 8 Plus share the same A11 Bionic architecture as the iPhone X.^[31]

iPhone X

The iPhone X, launched on November 3, 2017, marked the debut of the A11 Bionic chip in a premium smartphone, powering its edge-to-edge 5.8-inch Super Retina OLED display, swipe-based gesture navigation that eliminated the traditional home button, and a rear dual-camera system with wide-angle and telephoto 12-megapixel lenses for enhanced computational photography.^[2]^[36]^[37] Configured with 3 GB of RAM to support fluid multitasking and high-demand applications, the iPhone X pairs the A11 Bionic with a sophisticated TrueDepth camera array on the front, enabling precise depth mapping and secure biometric authentication via Face ID.^[38]^[39] The A11's integrated Neural Engine delivers real-time facial recognition essential for Face ID, powers expressive Animoji that animate based on tracked facial movements, and facilitates advanced augmented reality capabilities through ARKit, leveraging the chip's machine learning acceleration for immersive experiences.^[2]^[5]^[40] Serving as the final flagship iPhone powered by the A11 Bionic before the A12's introduction in later models, the device received full iOS updates up to version 16, was discontinued from Apple's official sales channels on September 12, 2018, and proved pivotal in establishing the bezel-less, notch-forward aesthetic that influenced subsequent smartphone designs.^[41]^[42]^[2]

Performance

Benchmark results

The Apple A11 Bionic demonstrated strong performance in CPU benchmarks, particularly when tested on devices like the iPhone 8 Plus. In Geekbench 5.5, it achieved an average single-core score of 927 and a multi-core score of 2339, reflecting efficient utilization of its hexa-core architecture for both light and demanding tasks.^[3] Graphics benchmarks highlighted the A11's custom Apple-designed three-core GPU, with scores in GFXBench 3.1 Manhattan offscreen averaging 50.6 frames per second and in 3DMark Sling Shot Extreme (ES 3.1) Graphics reaching 3244 points. These results indicated a 30% performance uplift over the A10 Fusion in graphics-intensive tests, enabling smoother rendering in games and augmented reality applications.^[12]^[43] In the AnTuTu v7 benchmark, the A11 Bionic posted overall scores around 226,000, underscoring its integrated neural processing unit's role in holistic system efficiency.^[44] Power efficiency metrics from battery life evaluations showed the A11 enabling 20-25% longer endurance in mixed workloads compared to A10-based devices, attributed to its four efficiency cores delivering up to 70% faster performance at lower power draw.^[45]^[46]

Comparisons to predecessors

The Apple A11 Bionic represented a significant advancement over its predecessor, the A10 Fusion, particularly in CPU architecture. The A11 features a hexa-core design with two high-performance cores that deliver 25% faster performance compared to the A10's high-performance cores, and four high-efficiency cores that are up to 70% faster than the A10's efficiency cores.^[4] This configuration, supported by a second-generation performance controller, enables up to 70% greater multi-threaded performance while preserving battery life.^[2] In graphics processing, the A11 introduced Apple's first fully custom-designed GPU with three cores, achieving 30% better performance than the A10's PowerVR-based GPU.^[4] This design also allows the A11 GPU to match the A10's performance while consuming half the power, contributing to improved energy efficiency in graphics-intensive tasks.^[47] A key new capability in the A11 is the dedicated Neural Engine, a dual-core unit capable of 600 billion operations per second, which was absent in the A10 and enables advanced machine learning features like real-time Face ID and augmented reality applications.^[2] The A11's overall design incorporates approximately 30% more transistors (4.3 billion versus the A10's 3.3 billion), fabricated on a 10 nm process compared to the A10's 16 nm, allowing for these expanded features and a smaller die size that aids broader system integration.^[48]^[49] Efficiency improvements extend to power management, with the A11 providing up to two hours longer battery life in devices compared to the iPhone 7's A10, alongside better thermal performance due to the refined core utilization and process node advantages.^[2]

References

[1]
A11 Bionic - Apple - WikiChip
A11 Bionic is a 64-bit hexa-core ARM performance mobile system on a chip introduced by Apple in late 2017 as a successor to the A10.
[2]
The future is here: iPhone X - Apple
Sep 12, 2017 · A11 Bionic, the most powerful and smartest chip ever in a smartphone, features a six-core CPU design with two performance cores that are 25 ...A11 Bionic Chip · Apple (AU) · Apple (CA) · Apple (IL)
[3]
Apple A11 Bionic SoC - Benchmarks and Specs - Notebookcheck
With 4.3 Billion transistors, the A11 Bionic is a rather big chip especially compared to the Snapdragon 835 (3 Billion) or a Skylake desktop quad-core Soc (1.75 ...
[4]
iPhone 8 and iPhone 8 Plus: A new generation of iPhone - Apple
Sep 12, 2017 · Apple announces iPhone 8 and iPhone 8 Plus, featuring a new glass and aluminum design, Retina HD displays, A11 Bionic chip and wireless ...Apple (CA) · Apple (SG) · Apple (HK) · iPhone 8 and iPhone 8 Plus
[5]
Apple unveils A11 bionic neural engine AI chip in iPhone X - CNBC
Sep 12, 2017 · The dual-core "A11 bionic neural engine" chip can perform 600 billion operations per second, Apple executive Phil Schiller said at the inaugural ...Missing: keynote | Show results with:keynote
[6]
The inside story of the iPhone X 'brain,' the A11 Bionic chip - Mashable
Sep 14, 2017 · Srouji told me that when Apple architects silicon, they start by looking three years out, which means the A11 Bionic was under development when ...Missing: timeline | Show results with:timeline
[7]
Inside iPhone 8: Apple's A11 Bionic introduces 5 new custom silicon ...
Sep 23, 2017 · Here's a look at the new Apple GPU, Neural Engine, its 6 core CPU, NVMe SSD controller and new custom video encoder inside the package.
[8]
Apple's 'Neural Engine' Infuses the iPhone With AI Smarts - WIRED
Sep 13, 2017 · Apple fires the first shot in a war over mobile-phone chips with a 'neural engine' designed to speed speech, image processing.
[9]
Apple Replacing PowerVR With In-House GPU - PCMag
Apr 3, 2017 · Apparently Apple is developing its own GPU, which will replace PowerVR and any other Imagination-owned technology or IP in future devices.
[10]
Apple A11 - Full Information and Specifications - iGotOffer
Oct 3, 2017 · It features the 2 x 4 cores structure: two high-performance cores called Monsoon and four high-efficiency cores called Mistral.
[11]
Apple A11 Bionic GPU - Benchmarks and Specs - Notebookcheck
It can be found in the Apple iPhone 8, 8 Plus and iPhone 10 and supports Metal 2. The architecture should be similar to the older PowerVR graphics cards like ...
[12]
Metal 2 on A11 - Overview - Tech Talks - Videos - Apple Developer
Sep 20, 2017 · The seamless integration of Metal 2 with the A11 Bionic chip lets your apps and games realize entirely new levels of performance and...
[13]
Apple A11 Bionic: specs and benchmarks - NanoReview
Apple A11 Bionic – an 6-core chipset that was announced on September 12, 2017, and is manufactured using a 10-nanometer process technology.Apple iPhone X · Apple iPhone 8 · Apple iPhone 8 Plus · Snapdragon 855 vs A11...
[14]
Tailor your apps for Apple GPUs and tile-based deferred rendering
The features include imageblocks, tile shading, raster order groups, and imageblock sample coverage control. A11 and later Apple GPUs also improve fragment ...Missing: ARKit | Show results with:ARKit
[15]
Deploying Transformers on the Apple Neural Engine
Jun 6, 2022 · It will help developers minimize the impact of their ML inference workloads on app memory, app responsiveness, and device battery life.Missing: challenges | Show results with:challenges
[16]
Apple Neural Processor - Mirabilis Design
Apr 19, 2021 · The neural engine allows Apple to implement neural network and machine learning in a more energy-efficient manner than using either the main CPU ...
[17]
Inside Apple's new A11 Bionic processor - ZDNET
Sep 12, 2017 · It features 4.3 billion transistors and its two performance cores are 25 percent faster than the A10 chip, and the four high-efficiency cores ...Missing: node | Show results with:node
[18]
TSMC to Be Sole Supplier of A11 Chip for Apple's 'iPhone 8'
Jul 18, 2016 · TSMC is said to have begun taping out the design for Apple's A11 processor in May. Xilinx, MediaTek, and HiSilicon are said to be the other ...
[19]
TSMC to start mass production of Apple A11 chip in April ahead of ...
Mar 27, 2017 · The report says the A11 chip is built using a FinFET process, packaged with a 'wafer-level integrated fan-out' technology … it sounds advanced ...
[20]
10 nm lithography process - WikiChip
Jun 27, 2025 · This allows for a full node shrink, enabling a 2X increase in logic density compared to their 16nm process. The 10FF process will have 15% ...
[21]
Apple's A11 Bionic Chip Decreases CPU1 Surface Area By 30%
Sep 27, 2017 · Apple's Latest A11 Bionic Processor Shrinks Die 30% Against A10 Fusion; GPU Design Doesn't Change Much Over Predecessor. The big question on ...
[22]
TSMC's 10nm FinFET Process Is On-Track - Ready To Begin Ramp ...
Nov 10, 2015 · TSMC's 10nm FinFET (CLN10FF) will have 110 – 120 per cent higher transistor density compared to its 16nm FinFET+ (CLN16FF+) process tech, 15 ...
[23]
Core ML - Machine Learning - Apple Developer
Core ML is optimized for on-device performance of a broad variety of model types by leveraging Apple silicon and minimizing memory footprint and power ...Core Ml · What's New · Xcode IntegrationMissing: A11 Bionic
[24]
Core ML | Apple Developer Documentation
Core ML provides a unified representation for all models. Your app uses Core ML APIs and user data to make predictions, and to train or fine-tune models, all on ...Getting a Core ML Model · Integrating a Core ML Model... · MLModel
[25]
https://coremltools.readme.io/
[26]
[PDF] Apple Platform Security
devices require the A11 Bionic chip or later, and only Mac computers with Apple silicon are supported. Apple Vision Pro requires visionOS 1.1 or later.<|control11|><|separator|>
[27]
Secure Enclave - Apple Support
Dec 19, 2024 · The Secure Enclave Processor, starting with the A11 and S4, includes a memory-protected engine and encrypted memory with anti-replay ...
[28]
https://www.apple.com/business/docs/FaceID_Security_Guide.pdf
[29]
iPhone 8 and 8 Plus announced with wireless charging, True Tone ...
Sep 12, 2017 · iPhone 8 and 8 Plus announced with wireless charging, True Tone display, A11 Bionic processor ... True Tone display, and upgraded cameras ...
[30]
iPhone 8 and 8 Plus review: The curious case of the time-traveling ...
Sep 28, 2017 · Apple has also added a dual-core Neural Engine to the A11's image signal processor (ISP), which reportedly is the source of the Bionic name.<|control11|><|separator|>
[31]
iPhone 8 Plus - Technical Specifications - Apple Support
A standard configuration uses approximately 11GB to 14GB of space (including iOS and preinstalled apps) depending on the model and settings. Preinstalled apps ...
[32]
iOS compatibility: Find out the latest version your iPhone can run
Sep 16, 2025 · Here's a quick guide so you can find out if your iPhone is compatible with the most recent iterations of the iPhone operating system.
[33]
I Tried The iPhone 8 — And It's Better, But Mostly The Same, As The 7
Sep 19, 2017 · The iPhone 8 will bring you up to speed. Compared to the older models, it has a much better camera, faster processor, second-generation Touch ID, 3D touch ...Missing: positioning | Show results with:positioning
[34]
Apple iPhone X Full Specifications - PhoneArena
Oct 28, 2025 · Apple iPhone X · Released · Nov 03, 2017 · 5.8-inch · 2436x1125px · Dual camera · 2716 mAh · 3GB · 64GB · Apple A11 Bionic · iOS 15.x.
[35]
iPhone X - Technical Specifications - Apple Support
Chip. A11 Bionic chip with 64-bit architecture. Neural engine. Embedded M11 motion coprocessor. Camera. 12MP wide-angle and telephoto cameras. Wide-angle: ƒ/1.8 ...Camera · Built-In Apps · LanguagesMissing: processor | Show results with:processor
[36]
iPhone X (2017) Review: Revolutionary Evolutionary Phone
Rating 4.0 · Review by Karen HaslamAug 15, 2019 · In terms of processor, RAM, storage and battery the iPhone X offers the following: A11 Bionic chip; 3GB RAM; Six-core CPU; An Apple-designed GPU ...Design And Build Quality · Face Id · Cameras & PhotographyMissing: support | Show results with:support<|control11|><|separator|>
[37]
The future is here: iPhone X - Apple (FI)
Apple announces iPhone X, the future of the smartphone, in an all-glass design with a Super Retina display, A11 Bionic chip and wireless charging.Reinvented Front And Back... · Introducing A11 Bionic · Designed For A Wireless...
[38]
iOS 11 ARKit Compatibility For Apps: Check If Your Device Is ...
Sep 19, 2017 · According to Apple, in order to use ARKit, iOS devices need to have an A9, A10 or A11 chip beating at their heart. While certainly not all ...
[39]
iOS 16 and iPadOS 16 are compatible with these devices
Nov 5, 2023 · iPhone · iPhone 14 · iPhone 14 Plus · iPhone 14 Pro · iPhone 14 Pro Max · iPhone 13 · iPhone 13 mini · iPhone 13 Pro · iPhone 13 Pro Max.
[40]
Apple discontinues iPhone X and the last iPhones with headphone ...
Sep 12, 2018 · Less than one year since it was released, Apple has discontinued the iPhone X with the introduction of its three new iPhones today.
[41]
Geekbench: A11 Bionic smokes past Android rivals, beats iPad Pro ...
Jan 18, 2018 · Geekbench's John Poole says the benchmarks are real, saying the A11's two high performance cores are probably clocked at 2.5GHz, up from the ...Missing: announcement | Show results with:announcement
[42]
Apple A11 scores 25 percent higher on Antutu than highest ...
Sep 16, 2017 · Now there are the results of an Antutu benchmark online which apparently show the A11 with a combined score of 226,058, which is slightly over ...
[43]
New Test Finds iPhone 8 Plus Has The Longest-Lasting Battery Of ...
Oct 2, 2017 · The 8 Plus topped the iPhone 8's battery life by a whopping 25%. In fact, it lasted 10 hours and 35 minutes, outperforming every other high-end flagship phone.
[44]
3 Things You Need to Know About Apple's A11 Bionic Chip
Sep 13, 2017 · The A11 Bionic has six processor cores -- two high-performance cores (25% faster than the ones in the A10 Fusion) and four high-efficiency cores ...
[45]
Apple A11 Bionic: The fastest six-core mobile processor around
Sep 12, 2017 · In terms of graphics performance, the GPU is said to be 30% faster ... half the power when working at the same rate as the A10. That's ...
[46]
Apple A10 Fusion SoC - NotebookCheck.net Tech
All in all the chip includes 3.3 billion transistors, which is more than a current AMD Bristol Ridge (3.1) and Skylake Quad-Core (1.75) X86 SoCs.
[47]
Apple A11 Bionic vs Apple A10 Fusion - Notebookcheck
Two high performance cores are clocked at up to 2.34 GHz and should be around 40% faster than the ...