Success in Silicon

Discover Positive Business Change Achieved Through Successful System-on-Chip Projects Contents

3 Evolving Intelligent Devices for Immersive Experience

6 Application-specific Integrated Circuit (ASIC) Design for Extreme Environments

8 Faster Take Off for Secure IC Company

10 Bringing Intelligence to Endpoint Devices

13 Hailo Empowering Intelligence

16 Connecting a New Generation of IoT Devices

2 Evolving intelligent devices for immersive experiences

January, 2019 Case Study

Intuitive Goal Company snapshot Develop a highly capable computer vision processor with the smallest possible footprint Name: Inuitive for use in AR/VR, drones and robotics applications. Vertical: Mobile, VR/AR, Robotic, ‘AIoT’ (AI of Things) Arm Solution Size: 80 employees Inuitive combined several types of processor into a single, low-power system-on-chip (SoC), HQ: Raanana, Israel with the Linux-capable Arm Cortex-A5 processor at its core. Founded: 2012

Benefits to Intuitive Cortex-A5 achieves balance between processing capability and power efficiency Linux-based chip with extensive ecosystem that enables rapid software development

A power-efficient, intelligent device for immersive experiences Inuitive develops intelligent vision technologies for deployment in applications such as augmented and virtual reality, drones, robotics and autonomous vehicles. Founded in 2012 by Israeli high-tech veterans Shlomo Gadot and Dor Zepeniuk, the Israel-based fabless semiconductor company is now made up of over 80 R&D professionals.

Inuitive released its first product, the NU3000 image and computer vision processor, in 2014. This 40nm SoC was used in Google Project Tango ecosystem for real-time depth generation, mapping, localization, navigation and other complex signal processing algorithms.

The new NU4000 builds on the success of the NU3000, offering improved 3D depth perception, SLAM-on-chip acceleration, deep learning and computer vision processing. In order to maximize the chip’s appeal, Inuitive integrated several types of processor “building blocks” to address different tasks within the SoC’s target uses – most prominently the growing AR, VR and XR market. Now at 12nm and almost half the die size of its predecessor, the NU4000 SoC combines the Inuitive proprietary 3D imaging and vision hardware accelerators engines with a suite of third-party IP: CEVA-XM4, Synopsys EV62 intelligent vision and deep learning processors, and the power-efficient Arm Cortex-A5 processor, running at 1Ghz.

3 “Hitting the balance between processing capability and power efficiency was our top priority in the NU4000” explains co-founder, CTO and VP of R&D Dor Zepeniuk. “Raw processing power was never the main value proposition for this solution; this is a chip designed for truly mobile markets such as drones, so we needed a physically-lightweight and power-efficient design capable of performing required functions without consuming too much power or silicon.”

Unlocking software innovation with access to Linux on Arm The Cortex-A5 offers other configurability options such as optional Floating-Point Unit (FPU) and NEON for accelerated audiovisual signal processing. It functions as the NU4000’s main core, running the operating system, hardware accelerators, power management, communication and additional code optimized for imaging and vision.

“This is a chip we expect our customers to develop really interesting software for,” says Zepeniuk. “But not all of our customers want to develop software for the chip itself, so we decided early-on that this should be a Linux-based device. The Cortex-A5 is ideal for this due to the huge software ecosystem. It’s a huge value point for us, and why we imagine so many interesting uses for the NU4000.”

Dor Zepeniuk Zepeniuk notes how easy it was for the team to integrate the Cortex-A5 with different Co-founder, CTO and VP of R&D IPs. “We needed DMA, AXI Matrix, a flash controller and so on. All of these integrated extremely easily with the Arm processor, which made development a whole lot easier for us. The Arm matrix bus and AMBA designer tool was also highly effective as it enabled us to tailor the architecture and adapt to certain challenges.”

4 “It’s my hope that by Choice and flexibility with a supportive ecosystem this time next year, the “We got excellent support from the Arm team here in Israel,” says Zepeniuk. “While there were very few hurdles when we came to implement the Cortex-A5 itself, it’s always NU4000 is recognized challenging to fit so many different pieces of IP onto one SoC and get them to play nicely as a true powerhouse for together. The support we received from Arm’s technical team was fundamental in ensuring driving the augmented and we reached our development milestones.” virtual reality revolution.” Once Inuitive had decided on the NU4000’s architectural design and created the RTL, it passed its designs to Arm Approved design partner Inomize in order to translate this into a physical design. “Inomize ensured that the chip met the size, speed and power consumption we needed to hit,” says Zepeniuk. “From verification to tape-out, Inomize played a vital role in rapidly moving the project from design to production.”

With verification and tape-out now successful and evaluations underway, the NU4000 is set to enter mass production in the second half of 2019. “We already have a number of customers evaluating the chip – Tier 1 players in the area of AR/VR and robotics,” says Zepeniuk. “It’s my hope that by this time next year, the NU4000 is recognized as a true powerhouse for driving the augmented and virtual reality revolution.”

Learn more about Intuitive and get start on your Linux-capable, Cortex-A5 based custom SoC with Arm DesignStart. Start quickly with a low-cost access fee, simplified contract and at least 1 year of design support to ensure success – visit designstart.arm.com

5 Application-specific Integrated Circuit (ASIC) Design for Extreme Environments RelChip Turns Up the Heat with Arm DesignStart Case Study

RelChip’s Goal Company snapshot Develop a high-performance microcontroller with the widest operating temperature Name: RelChip range in the world to be used for applications in extreme conditions, from geothermal Vertical: Extreme environment to aerospace. microprocessors Size: <50 employees Arm Solution HQ: Utah, USA Through Arm DesignStart, the fast, low-cost IP access program used to accelerate Arm Solution: Arm Cortex-M CPU ASIC design, RelChip was able to access the Arm Cortex-M0 CPU for no upfront fee. through Arm DesignStart This enabled the company to reimagine its existing microcontroller (MCU) and accelerate time to market. The project resulted in an enhanced MCU that offers 10 times the performance, with a die area one-hundredth the size of the original.

Benefits of Arm Solution 10 times the speed of the preceding RC10001 microcontroller 100 times area reduction in the processor Lower power at 10 times the frequency of the previous design A 5-year life at 225°C

Extreme Processor Design

Aiming to push the boundaries of high-temperature electronics, RelChip has established itself in a niche market with the first microcontroller for extreme environmental products — those that operate from a chilly -55°C to a scorching 225°C.

RelChip used Arm technology for its original product design, and now is using Arm technology to enhance that product based on Arm Cortex-M0 processor IP to meet even higher performance ratings. They gained fast, easy access to the Cortex-M0 CPU for no upfront fee through Arm DesignStart, which helped them get to market quickly and cost- effectively.

“After surveying the field, we realized the Cortex-M0 was the perfect fit,” explains Marshall Soares, Relchip’s founder and president. “We used Arm Cortex-M0 technology for our first product, and that’s what made our company possible. Our customers know how to program

6 on Arm—it’s proven technology. Plus, working with DesignStart offered access to the IP at no upfront cost, so we could reduce risk and design right away, rather than wait on funding for upfront licensing before production.”

Top-Gun Performance With the Arm solution, RelChip increased memory capacity from 4K bytes on the current chip to 32K bytes, which offers more flexibility in high-temperature environments that limit memory size.

“The numbers we’ve been getting have been very consistent, very predictable to the point of silicon,” Soares says. “It’s been a very straightforward design process for us, and the results, compared to the original design were nothing short of fantastic.”

Fantastic may be an understatement for the microcontroller’s jet engine application. A controller with increased memory that can withstand the heat of the engine at the site of the sensor or actuator and perform vital calculations helps decrease the weight on the engine and improve its efficiency.

“That’s a big deal when it comes to paying the fuel bill,” Soares explains. “A one percent efficiency increase equals a one percent decrease in the cost of fuel for a flight. That adds up to a lot when you do the math for every flight, every day. Furthermore, reducing the weight of the engine means more room for people and cargo in the jet, further increasing efficiency.”

Efficiency Takes Flight By moving the sensor and actuator electronics to the sensor/actuator location, RelChip found it could increase efficiency, decrease connections, and modularize designs. The Cortex-M0 based chip design offers the perfect size and performance for RelChip’s target applications.

The result is a revolutionary improvement in sensor and actuator systems for extreme conditions that has applications for a number of industries, from geothermal to aerospace.

“The new microcontroller developed on Arm improves speed by 10 times, the area by 100 times (in the processor), and power is lower at 10 times the frequency than our previous design,” Soares says.

Onboard for the Future RelChip plans to continue its development work using Arm solutions through the DesignStart program.

“We’re looking to build entirely new families of system solutions, and since not all of them will be high volume for the markets we serve,” Soares adds. “DesignStart lets us reuse software to speed that development, and move forward on the right Arm IP for the project.”

Learn more about RelChip and get started on your Arm-based custom SoC with faster access and lower costs through Arm DesignStart today.

7 Faster Take Off for Secure IC Company Accelerated Chip Design for Drones and Cameras with Arm DesignStart Case Study

NeoWine Goal Company snapshot Develop a powerful, secure drone chipset with fully encrypted signal processing for both Name: NeoWine camera video and drone control. Vertical: Semiconductor - Security Size: 20 employees Arm Solution HQ: Korea NeoWine turned to the Arm DesignStart program to create a secure encryption Founded: 2002 ASIC and high-performance MCU for drones, based on the Arm Cortex-M0 and Arm Solution: Arm Cortex-M3 Cortex-M3 processors. CPU through DesignStart

Arm Solution Benefits High-performance design compliant with latest AES encryption standards Power-efficiency for long drone flights Low-cost for mass production volumes Small area chip size for minimum onboard payload

At some point in almost every action movie or TV show, an evil villain hacks into a camera, surveillance system or drone to wreak mayhem on unwitting victims. That’s exactly the kind of activity that NeoWine, a security integrated circuit (IC) company based in Korea, is out to stop happening in real life.

Already, NeoWine has shipped more than 130 million of its secure chipsets across China, Taiwan, and Korea, specializing in the video security market. Its latest product, the DALPU-4, is a specially designed ASIC capable of both controlling drones and managing onboard camera functions. The DALPU-4 relies on Arm technology for both high-performance function and small chip size.

“Drones challenge every design limitation for both processing power and size,” explains Hyo Seung, CEO of NeoWine. “Only Arm offered the performance, size, and power efficiency we needed for the functionality, security, and encryption speeds we want to offer to drone manufacturers. Even better, we could get started on designing and prototyping our Arm- based solution for no upfront fee through Arm DesignStart.”

8 Powerful Data Encryption The market for drones has exploded over the last few years, with new use cases constantly emerging and innovative designs permeating the market. However, like any other Internet of Things (IoT) device, drones also are vulnerable to hackers and security breaches, including data theft and loss of control.

The NeoWine ASIC provides a key exchange with ECDH 256, AES 256 encryption, on both video and control channels at the same time. AES encryption speed is 30 Mbps and control is 100 kbps.

“The Arm DesignStart program gave us the advantage we needed to get our design to market quickly. Using Arm solutions meant we could access a wealth of technical documentation and compatible compiler code that was readily available,” Hyo Seung Lee says. “Security in our market is a little different than others, such as electronic payment cards, but Arm helped us apply payment-level security with low-power, high-performance CPU technology to our ASIC.”

Scalable processor technology As the demand for high-quality video from low-cost, low-powered drones continues to grow, NeoWine plans to expand its product line and believes the scalability and flexibility of the Arm Cortex-M0 and Cortex-M3 processors allow the company to innovate more quickly.

Arm’s upward-compatible processors means software code can be reused, opening the door for more features and advanced capability in future.

“When we tell our customers that the ASIC is based on Arm architecture, they know straight away that they’re getting a high-quality chipset that will deliver the performance and security they need,” Hyo Seung Lee concludes. “Our company slogan is ‘For the more secure world,’ and Arm is the perfect partner to help us deliver on that.”

Learn more about NeoWine and get started on your Arm-based custom SoC with faster, lower costs through Arm DesignStart today.

9 Bringing Intelligence to Endpoint Devices Eta Compute Preconfigures ASIC with Neural Networks for Optimum Machine Learning https://etacompute.com Case Study

Eta Compute Goal Company snapshot To deliver machine learning (ML) capabilities to embedded systems in low-power intelligent Name: Eta Compute applications, such as image and speech recognition, face detection, and heartrate monitoring. Industry: ASICs for AI/ML

Size: Less than 50 employees Arm Solution HQ: Westlake Village, Calif., USA Arm Cortex-M3 meets all Eta Compute requirements for performance and low-power Arm Solution: Arm Cortex-M3 consumption. The Arm DesignStart program gave the company easy access to Cortex-M3 processor through Arm DesignStart reference designs, documentation, and more for no upfront fees. DesignStart helped Eta Compute accelerate its development cycle, minimize risk with a proven reference example and reduce the upfront costs of building a microcontroller (MCU) for neural networks with low-power consumption.

Benefits of the Arm Solution Reference design that helps reduce design complexity and time-to-market. Quick, open access to a high-performance, energy-efficient processor with a simple contract. A vast ecosystem of design services, tools, software and training, including IAR Systems, an Arm tools partner, for an optimized code compiler that helps to reduce power consumption. Proven processor technology on a well-established architecture, optimized for ML workloads. Low IP startup costs for accelerated market entry.

As the market for always-on, intelligent mobile and edge devices expands, companies that are looking to develop products for AI applications, such as speech, image recognition, and heartrate monitoring, need processors with machine learning capabilities that consume minimal power.

To address this need, Eta Compute developed EtaCore, an energy-efficient application- specific integrated circuit (ASIC) with neural network-based AI software to run on hardware that is optimized for low power consumption. EtaCore has unparalleled power efficiency for speech activity and voice recognition that lets OEMs deliver more features with significantly longer battery life.

10 “We chose to partner with The ASIC was developed with a CPU and system IP available through the fast-access Arm Arm because it offers DesignStart Pro program and preconfigured with a variety of neural networks running on Eta Compute’s asynchronous implementation. proven technology and a huge ecosystem of support.” “We chose to partner with Arm because it offers proven technology and a huge ecosystem of support,” says Paul Washkewicz of Eta Compute. “We didn’t want to introduce any risk factors for our customers. With Arm we can test the software and see straight away that the architecture and algorithms work as required.”

The ASIC can perform complex video and imaging tasks, such as image recognition or transcription of handwritten text, and enables sensor hubs to perform more extensive sensor algorithms, providing real-time data and updates from mobile and IoT networked devices.

Flexibility and Expertise from the Arm Ecosystem To help reduce power consumption on the ASIC even further, Eta Compute chose to partner with one of Arm’s ecosystem partners, IAR Systems, for their powerful development tools. IAR Embedded Workbench helps ensure code on EtaCore runs faster, smaller, and smarter, while optimizing power efficiency.

“We developed the chip using IAR Systems’ tools to squeeze every last drop of power consumption out of the ASIC and get great results,” Washkewicz explains. “IAR Embedded Workbench is fast and lightweight, resulting in an IDE that helps accelerate development.”

IAR Systems also offered top-notch support, with the expertise and response times Eta Compute needed to get from prototype to production quickly, within six months.

“The Arm ecosystem offers choice,” explains Robert DeOliveira of IAR Systems. “More than 150,000 companies have migrated to Arm using our toolchain. By using DesignStart Pro, Eta Compute has also been able to take advantage of our powerful optimization and debugging technologies to help it be successful in the AI space.”

Arm Cortex-M and Design Support Arm’s easy, low-cost licensing of Cortex-M technology was a big factor in Eta Compute’s decision to use DesignStart Pro. It gives the team access to proven CPU and system IP used in billions of devices to date, as well as training, support and resources, access to EDA partners, and a global network of Arm Approved design partners. All of this helps provide the simplest, smoothest, most cost-efficient path to a custom ASIC.

11 “We wanted an innovative “To meet the needs of the intelligent IoT market, we have to keep costs low and support solution we could the many use cases of different customers,” Eta Compute’s Washkewicz says. “DesignStart Pro gave us fast access to the Cortex-M3 CPU and system IP, as well as everything else deliver quickly, as we’re we needed to develop an ASIC that appeals to a broad swath of industries – and streamline implementing logic the path to help us deliver quickly without high upfront costs.” specifically to reduce power The ASIC, EtaCore, is the world’s lowest power microcontroller IP, operating on the lowest on a given application.” power supply without compromising functionality or performance. It offers a powerful foundation for a wide range of applications that support a more intelligent IoT.

“We wanted an innovative solution we could deliver quickly, as we’re implementing logic specifically to reduce power on a given application,” Washkewicz says. “Our latest revision achieved this, as we could focus our energy on innovating and differentiating our product by making the most of everything Arm and its ecosystem offer.”

Intelligent Expansion Ahead Now, Arm’s reputation is helping Eta Compute attract new customers in battery- constrained markets, such as hearables and wearables, who are looking for machine learning features at the edge.

“Currently, we’re focusing on using Arm Cortex-M technology to make tiny devices more intelligent with features such as speech command and wake-up sensors,” Washkewicz says.

“But we’re also looking at higher-class Arm processor technology for the next generation of image recognition and object detection. With one platform and one ecosystem, we’re set up to go from strength to strength.”

Learn more about Eta Compute and IAR, and get started on your Arm-based custom SoC with faster access and lower costs through Arm DesignStart today.

12 Case Study

Hailo Goal Company snapshot Design, prototype and secure funding for a new embedded SoC, capable of both traditional Name: Hailo and neural network processing at the edge based on Arm Cortex processors. Vertical: IoT - automotive Size: 40 employees Arm Solution HQ: Tel Aviv, Israel Hailo used Arm DesignStart Pro in order to quickly gain access to the IP necessary to create Founded: 2017 a prototype device, pairing a high-efficiency Arm Cortex-M3 CPU with its own proprietary hardware IP. Creating a viable proof of concept using Arm IP with no upfront license fee helped Hailo to secure $16 million in funding.

Benefits for Hailo High-efficiency, high-performance compute acceleration for complex machine learning (ML) workloads Developers can use an established Arm toolchain to program the processor Enables independent real-time ML in devices at the network edge Supports all current neural network models Enables embedded ML in various applications including drones, cars, smart home appliances and cameras

Proof of concept through DesignStart Pro helps secure $16 million in funding Hailo has a single vision: to enable edge devices such as drones, cars and smart home appliances to think for themselves. Founded by Orr Danon, the Israeli-based startup does this by combining traditional compute processes with intuitive machine learning capability.

“Like the two hemispheres of the human brain, when it comes to solving a real-life problem, it is the objective, logical side working together with the subjective, intuitive side that results in a decision,” Danon says. Hailo’s designs use tried-and-tested processor technology as a foundation for its own proprietary neural network hardware. In selecting a foundation, however, a low barrier to entry was paramount for the start-up.

13 With $0 upfront license fees and a success-based, royalty-only model, the Arm DesignStart Pro program provided the ideal development environment in which Hailo could quickly and reliably prototype its SoC (system-on-chip) design.

“DesignStart Pro has proven invaluable to us,” Danon says. “It tackles the pain points typically experienced by early stage startups when funding is extremely tight. At this point in our journey we had to spend our money wisely, taking the custom approach in areas we wanted to really differentiate in, while using readily available, high-quality IP

Hailo Co-founders: and tools everywhere else.” Orr Danon, Hadar Zeitlin and Avi Baum To build its prototype, Hailo selected the Arm Cortex-M3 processor, designed to enable high-performance, low-cost platforms for use in a broad range of devices.

“We chose the Arm Cortex-M3 for a number of reasons,” Danon explains. “It’s an incredibly capable processor in its class, with low power requirements. By properly allocating what is running where and joining forces between it and our own IP, we managed to create a highly capable yet also extremely power-efficient prototype.”

Cortex-M3 also opens the door for Hailo to augment its design with further Arm-based IP, such as sensors and memory processors. Most importantly, the Arm processor acts as the ‘security guard’ for the whole device: secure boot ensures that firmware cannot be tampered with, and all further processing occurs within the CPU’s trusted execution environment.

An unparalleled ecosystem Hailo’s multi-disciplinary team is made up of industry experts with system-oriented backgrounds, committed to develop end-to-end solutions. An important part of this, explains Danon, is co-designing the software and hardware.

“Creating robust hardware requires an equally robust software ecosystem with a tried-and- tested software toolchain. We’re not interested in forcing the industry to use non-standard toolsets; our toolchain combines the standard Arm tools seamlessly with our own software development kit. “Arm’s extensive software ecosystem, combined with the variety of support and resources that exist, made it the obvious choice for us to ensure successful adoption in the market.”

14 “Prototyping using Arm Getting funded DesignStart Pro enabled us “Prototyping using Arm DesignStart Pro enabled us to design and demonstrate a comprehensive proof of concept to potential investors, helping to gather more to design and demonstrate than $16 million in funding from stakeholders,” Danon says. a comprehensive proof of concept to potential This funding, which comes from Israeli crowdfunding and a number of angel investors, investors, helping to gather lets Hailo build on its prototype to create a finished product that demonstrates the success of this technology. more than $16 million in

funding from stakeholders.” While the Arm Cortex-M3 processor is more than sufficient for Hailo’s current use cases, Danon believes that the continuity of the Arm ecosystem provides a great springboard into more advanced Arm processors for higher performance applications.

Hailo’s key market, Danon says, is the automotive industry — specifically, its move toward autonomous driving technology.

“Vehicle autonomy requires a lot of compute from an edge perspective; currently, an autonomous vehicle needs to be a tiny datacenter on wheels. Our solution can accommodate high-level requirements for autonomous driving, and for that we may need a more powerful Arm solution such as dual or quad-core Cortex-A,or even Cortex-R, CPUs. Conversely, some solutions could incorporate an even lower-power Arm Cortex-M0 processor. Using Arm, we have complete flexibility.”

Learn how Arm is enabling machine learning at the edge. Interested in creating your own design? Learn more about Arm DesignStart.

15 Connecting a new generation of IoT devices

Case Study

EigenCOMM Goal Company snapshot Develop a low-power, high-performance chipset solution based on the NB-IoT standard Name: EigenCOMM to help developers affordably offer communication services for connected devices for the Vertical: Semiconductor - IoT Internet of Things (IoT). Size: 60 employees HQ: Shanghai Arm Solution Founded: 2017 At the heart of the industry-leading EigenCOMM EC616 SoC is the Arm Cortex-M3 with Arm Corstone-100 foundation IP. The solution was developed using theArm DesignStart program, and implemented with Arm Artisan Physical IP.

Benefits to EigenCOMM Die size nearly 1/3 of other solutions 50 percent less power consumption than other options Rapid development time in about six months Successful tape out first time Access to Arm experts through its rich ecosystem Powerful foundation for future innovation

As the world prepares for more than a trillion connected devices by 2035, semiconductor company EigenCOMM is paving the way with a cellular-based chipset solution designed to meet the needs of a growing IoT. Its newest SoC, the EC616 was taped out in about six months, a major boost to time to market for IoT innovation.

Founded in 2017, the Shanghai-based startup is focusing on an NB-IoT system-on-chip (SoC) based on the Arm Cortex-M3 processor. NarrowBand-IoT is widely regarded as the best low-power wide area network standard to support cellular devices and services arising from the massive (and growing) volume of connected devices. NB-IoT is specifically designed for efficient, low-bandwidth communication on mass-distributed devices that require a long battery life – perfect for the IoT.

“NB-IoT is poised to enable huge IoT applications, from smart cities to smart agriculture, and across industries into daily life,” says Haifeng Xiong, president of EigenCOMM. “Arm is the perfect partner for EigenCOMM’s expansion into IoT technologies, with its wide ecosystem of tools, software, knowledge and developers.”

16 “Arm is the perfect partner SoC for a wide range of applications for EigenCOMM’s expansion The EigenCOMM team brings deep expertise to the IoT industry, covering protocols such as PHS, TD-SCDMA, LTE, L1/L2/L3 and RF, as well as SoC design experience focused in the into IoT technologies, with mobile market. Its Arm-based product, the EC616, is an ultra-low power, low-cost NB-IoT SoC. its wide ecosystem of tools, software, knowledge The EC616 supports the connectivity and performance required in designs for billions of and developers.” IoT devices across a wide range of markets, including smart buildings, smart cities, supply chain logistics, medical devices, wearables, and many more emerging IoT innovations. Its reduced size increases the chipset’s flexibility as the foundation for the smallest communications-enabled devices.

“The EC616 was built using the Arm DesignStart program with Arm Cortex-M3 and Corstone foundation IP,” Xiong explains. “The innovative architecture and sophisticated communication system enabled our experts to achieve a die size nearly 1/3 of existing solutions in the market. In addition, we were able to reduce overall power consumption to at least 50 percent of that offered by comparable devices by optimizing work node power consumption.”

First tape out success Arm’s wide foundry support for Artisan Physical IP made it possible for EigenCOMM to switch foundries easily for optimum production needs and costs, as the development team quickly reached an early production phase.

“We taped out our test chip in one foundry, and moved to another for mass production,” Xiong explains. “Due to Arm’s proven physical IP, our first tape out was successful and took less than 6 months from beginning the project. The time savings and reduced upfront investment gave us a huge advantage.”

The Arm DesignStart advantage Arm DesignStart lets developers turn their business ideas into products, without worrying about increased risk or the cost and time associated with custom SoC designs. “DesignStart offers us easy and instant access to proven Arm CPU IP, plus an extensive ecosystem of software, debuggers, OS, partners and tools, Xiong says. “Not only could we accelerate production time, but we could also use our resources to focus on the main value add - the transceiver.”

Support for IoT vision “As the EC616 goes to market, we anticipate it setting a whole new benchmark for the industry,” Xiong says. “Not just for its low power and low cost, but for its comprehensive feature set, such as integrated PA, sawless and DCXO support, as well as peripherals that include ADC, PWM, I2C, SPI, UART, and Crypto. Thanks to Arm’s high-performance technology, we can make our vision as an IoT industry leader a reality.”

Learn more about EigenCOMM and get started on your Arm-based custom SoC for no upfront fee by registering for Arm DesignStart today.

17 Choose a Licensing Plan That’s Right for Your Business

As the world’s leading semiconductor intellectual DesignStart property (IP) supplier, Arm licenses its technology to a $0 for the Cortex-M0, $75K entry package annual Upfront license fees based on vast network of partners, Cortex-M1 and Cortex-M3 access fee license terms from leading semiconductor CPUs companies to niche custom $200K standard package design companies. $75K for the Cortex-A5 CPU annual access fee

Per product license fees due at Depending on your needs, No additional license fee for the Cortex-M0, Cortex-M1, tape out choose the best way to access and Cortex-M3 CPUs Arm IP, tools and support. $50K license fee for the Cortex-A5 CPU at tape out

Includes

Arm Cortex-A5, A wide range of Access to the most Cortex-M3 and Cortex-A, Cortex-R and advanced Arm IP Cortex-M0 CPUs Cortex-M CPUs Locked-down system- FPGA-optimized Mali GPUs on-chip (SoC) roadmaps Cortex-M3 and Corstone foundation IP with multiple uses of Cortex-M1 soft CPU IP System IP specific Arm IP products Corstone foundation IP Security IP Supporting system IP Artisan physical IP Artisan physical IP libraries libraries

90-day trial of Arm Keil MDK Arm DS Gold One Software, IP tooling and models are all licensed as 90-day trial of IAR Embedded Arm Socrates additional products Workbench Arm System Models Model of Cortex-M3 Standard support Arm Socrates and maintenance for all included products Community-based support

www.arm.com/resources/ www.arm.com/products/ www.arm.com/products designstart flexible-access/product

18 19 Find Out How To Access Arm IP Today: www.arm.com/why-arm/how-licensing-works

All brand names or product names are the property of their respective holders. Neither the whole nor any part of the information contained in, or the product described in, this document may be adapted or reproduced in any material form except with the prior written permission of the copyright holder. The product described in this document is subject to continuous developments and improvements. All particulars of the product and its use contained in this document are given in good faith. All warranties implied or expressed, including but not limited to implied warranties of satisfactory quality or fitness for purpose are excluded. This document is intended only to provide information to the reader about the product. To the extent permitted by local laws Arm shall not be liable for any loss or damage arising from the use of any information in this document or any error or omission in such information.

© Arm Ltd. 2019

20