The Golden Age of Domestic MCUs in China

The Golden Age of China Domestic MCU

Table of Contents

1. Global and China MCU Market Trends

2. Six Major Application Markets for MCUs

3. Six Directions for Future MCU Design

4. Five Driving Forces for the Development of Domestic MCUs

5. Top 10 Domestic MCU Listed Companies

6. Top 10 Domestic MCU Chips

7. Top 10 Automotive Grade MCU Chips

8. Top 10 AIoT MCU/SoC Chips

9. Top 10 RISC-V MCU/SoC Chips

10. Overview of 50 Domestic MCU Manufacturers

Note: Due to space limitations, this report only lists summaries of each section. Please click on the title links of each section to enter the webpage and read the full version of the report. You can also download the PDF version of this report at the end.

Global and China MCU Market Trends

According to authoritative market research institutions, the global MCU market size is approximately $17.79 billion in 2020, with shipments exceeding 28 billion units. It is expected that this year’s MCU size will reach $18.48 billion, with a compound annual growth rate (CAGR) of 10.1% from 2021 to 2028, growing to $36.16 billion by 2028.

According to IC Insights, global MCU revenue reached a record high of $17.6 billion in 2018; it fell by 7% in 2019; and continued to decline by 8% in 2020, down to $14.9 billion. In 2020, automotive MCU sales were $6 billion; industrial MCU revenue accounted for 29%, approximately $4.3 billion. It is expected that the total global MCU market sales in 2021 could reach $19 billion, with automotive and industrial markets accounting for 70%, and consumer electronics, home appliances, and computers and communications markets accounting for the remaining 30%.

It is expected that global automotive MCU sales will surge by 23% in 2021, reaching a record $7.6 billion, followed by a 14% increase in 2022 and about a 16% increase in 2023.

In terms of automotive MCU types, over three-quarters are 32-bit MCUs, with expected sales of about $5.8 billion this year; 16-bit MCUs account for 17.6%, about $1.34 billion; and 8-bit MCUs account for less than 6%, at $441 million. The higher average selling price (ASP) of 32-bit MCUs will also push up this year’s sales, partly due to tight market supply, leading MCU suppliers to start raising prices. The average selling price of 32-bit MCUs is expected to rise by 13% in 2021, increasing to $0.72.

Now let’s look at the domestic MCU market. According to IHS data, from 2008 to 2018, the annual average compound growth rate (CAGR) of the Chinese MCU market was 7.2%, which is four times the global MCU market growth rate during the same period. In 2019, the Chinese MCU market size reached 25.6 billion yuan, and in 2020 it reached 26.9 billion yuan, but it was mainly dominated by overseas giants, with a low penetration rate of domestic MCUs.

The application market for domestic MCUs is mainly concentrated in home appliances/consumer electronics, computer networks and communications, automotive electronics, smart cards, as well as industrial control/medical fields. Among them, the demand for MCUs in automotive electronics and industrial control applications is growing the fastest, and it is expected that by 2023, the market share of industrial/medical electronics will catch up with that of consumer electronics, reaching 9.2 billion yuan.

Domestic MCU manufacturers are mainly concentrated in the consumer electronics and home appliance segments, while the automotive and industrial control mid-to-high-end markets are dominated by international giants. However, in the emerging IoT application field, domestic MCU manufacturers are almost on the same starting line as international giants.

Six Major Application Markets for MCUs

From the perspective of competition in the MCU industry, the major global suppliers are still dominated by international manufacturers, including STMicroelectronics, Renesas, NXP, Texas Instruments, and Infineon (including Cypress), which occupy a leading position. These five international manufacturers mainly focus on automotive electronics and industrial applications, while in the markets of consumer electronics, home appliances, computers, and network communications, as well as emerging IoT, Silicon Labs, Taiwanese manufacturers, and many domestic manufacturers occupy a larger share.

We roughly divide the Chinese MCU market into six application categories: home appliances and consumer electronics, IoT, smart metering/IC cards and security, computer and network communications, industrial control, and automotive electronics. The table below lists the main domestic MCU manufacturers in each application category.

1. Home Appliances and Consumer Electronics

According to IDC data, the scale of the smart home market in China is expected to grow from $11.6 billion in 2018 to $36.8 billion in 2024, with a six-year average compound growth rate of 21.22%; shipments are expected to grow from 156 million units in 2018 to 453 million units in 2023, with a five-year average compound growth rate of 21.5%.

Among domestic MCU manufacturers, Zhongying Electronics and Shenzhen Zhongwei Semiconductor have laid out their market presence in the home appliance sector earlier, gaining relative market advantages. As home appliance giants such as Midea, Gree, and Haier shift towards smart home appliances and even start their own chip R&D projects, domestic MCU manufacturers will achieve continuous growth in this market and will hold a dominant position in the global market.

As technological bottlenecks are continuously broken, the scale of the wireless charging market is growing year by year. According to Zhiyan Consulting data, the global wireless charging market was $8.7 billion in 2019 and is expected to grow to $15 billion by 2024, with an average annual compound growth rate of 12%. With the strong development of the global wireless charging market, the scale of the Chinese wireless charging market is also continuously increasing. From 360 million yuan in 2018 to 23.94 billion yuan in 2026, market scale growth is expected. In the consumer electronics field, mobile phones are the largest market for wireless charging. Domestic manufacturers such as Voda Semiconductor and Shenzhen Zhongwei Semiconductor are leading in wireless charging transmitter and receiver chips, as well as corresponding MCU and SoC chips.

2. IoT

With the rapid development of the IoT, the demand for sensors is continuously growing. In 2019, the global sensor market size was $226.3 billion, with a year-on-year growth rate of 11.63%; it is expected to reach $298.5 billion by 2023, with an average compound growth rate of 7.14% compared to 2019. The Chinese sensor market still has a significant gap compared to the US, Japan, and Germany, but its overall growth rate is leading globally. In 2019, the Chinese sensor market size was $24.3 billion, with a year-on-year growth rate of 13.45%; it is expected to reach $38.8 billion by 2023, with an average compound growth rate of 12.44% compared to 2019.

In the field of IoT applications, the combination of sensors and MCUs to form smart sensors will have considerable development space. Domestic manufacturers such as Goodix Technology, Chipsea Technology, GigaDevice (which has acquired Silan Micro) and Awinic have certain technical strength and market application advantages in touch, smart sensors, and high-precision ADC.

3. Smart Meters, IC Cards, and Security

After nearly 20 years of development, the domestic smart meter industry has basically achieved full domestic production of core components such as energy measurement chips, smart meter MCUs, and carrier communication chips. Taking smart meter MCUs as an example, the current main control MCU chips generally use 32-bit ARM Cortex-M cores, with operating frequencies ranging from tens to hundreds of MHz, typically manufactured using 180-90nm embedded flash processes, integrating 128KB – 512KB of large-capacity embedded flash memory, 8KB – 64KB of embedded SRAM, and integrating rich peripheral functions including ADC, temperature sensors, LCD drivers, UART/SPI/I2C communication interfaces, and high-precision real-time clocks, with extremely low operating and sleep power consumption.

In terms of smart cards and security chips, National Technology, Fudan Microelectronics, Unisoc, and China Electronics Huada Technology are the mainstream manufacturers in the domestic related fields. NXP holds a leading position in this field internationally, but due to the lack of products containing commercial encryption algorithms, its competitiveness in the domestic security market is gradually decreasing. Unisoc and China Electronics Huada Technology, having entered the smart card and security chip industry earlier, have certain first-mover advantages and occupy a market share advantage in the domestic financial card field compared to other companies.

4. Computers and Network Communications

5. Industrial Control

The scale of China’s industrial control and automation market continues to expand, expected to reach 208.5 billion yuan by 2022. Industrial MCU products are mainly used in motor control, instruments, low-voltage distribution, power tools, industrial robots, and other application scenarios, with functions mainly focused on motor control calculations, data acquisition, and control. As the complexity of industrial equipment increases, the number of industrial MCUs used per unit continues to grow. For example, in industrial robots, a single machine typically uses more than ten MCU chips.

The traditional industrial control MCU field has long been dominated by international giants such as TI, ST, ADI, and Renesas, with only Huada Semiconductor’s MCU division occupying a place in this niche market among domestic MCU manufacturers. In recent years, manufacturers like Wanggao, Hangshun, and Jihai Semiconductor have also begun to make strides in the industrial MCU market. For example, Jihai Semiconductor currently has MCUs based on M0+, M3, and M4, covering working temperatures from -45℃ to +125℃, already applied in frequency converters, motor drivers, servers, inverters, and BMS management in the industrial control sector. Shenzhen-based Fengqian Technology, a design company focused on high-performance BLDC motor drive control chips, covers all key chips for motor drive control, including motor control chips MCU/ASIC, motor drive chips HVIC, motor-specific power devices MOSFET, and intelligent power modules IPM.

6. Automotive Electronics

As the core device for vehicle control, MCUs are mainly used in body control, driving control, infotainment, and driver assistance systems. The automotive MCU market is relatively mature, with a stable competitive landscape. NXP, Infineon, Renesas, STMicroelectronics, and Texas Instruments have long occupied the top five positions in the global automotive MCU market, with a market share exceeding 95% in 2020.

Among numerous domestic MCU manufacturers, a few companies such as AutoChips, ChipON, Saiteng Micro, Chipways, and BYD Semiconductor have already mass-produced MCUs and entered the automotive OEM supply chain, but their products are still limited to simple control applications, such as windows, lighting, and cooling systems, while complex applications such as powertrain control, intelligent cockpits, and ADAS are still rare.

Six Directions for Future MCU Design

With the development and integration of AI and IoT, the design of microcontrollers (MCUs) has become more complex, gradually shifting from traditional single-function microcontrollers to system-on-chip (SoC) with more integrated functional features and stronger computing performance. The ASPENCORE “Electronic Engineering Magazine” analyst team has identified the following six development directions for MCU design.

1. More Intelligent (AI)

Since 2017, MCU manufacturers have attempted to add AI capabilities to MCUs. For example, ST’s Project Orlando serves as an experimental ultra-low-power AI accelerator unit for MCUs, while Renesas released a programmable reconfigurable coprocessor DRP for MCUs in 2018. After three years of development, adding AI accelerators to MCUs is becoming increasingly mainstream. In application scenarios requiring AI-related computing power, using dedicated AI accelerators is often more effective than enhancing processor performance.

From an application perspective, the main reason for the mainstream adoption of AI accelerators paired with MCUs is the increasing prevalence of applications requiring AI. From specific algorithms and models, the focus is concentrated on a few models, such as convolutional neural networks for machine vision (facial recognition, object recognition) and recurrent neural networks (RNN) needed for some advanced speech recognition.

2. Stronger Performance

The Cortex-M series is based on the ARMv7-M architecture (used for Cortex-M3 and Cortex-M4), while the lower-end Cortex-M0+ is based on the ARMv6-M architecture. The first Cortex-M processor was released in 2004, and as some mainstream MCU suppliers began to mass-produce MCU chips based on this core, the Cortex-M processor quickly gained favor in the market. One could say that the Cortex-M for 32-bit MCUs is like the 8051 for 8-bit MCUs, quickly becoming the industry standard microprocessor core, with various MCU suppliers developing their own products based on this core, offering differentiated products in the market.

For cost-sensitive applications or those transitioning from 8-bit to 32-bit, the lowest-end products of the Cortex-M series may be the best choice. Although the performance of the Cortex-M0+ is only 0.95 DMIPS/MHz, which is lower than that of Cortex-M3 and Cortex-M4, it is still compatible with other high-end products in the same series. The Cortex-M0+ uses a subset of the Thumb-2 instruction set, and most of these instructions are 16-bit operands (although all data operations are 32-bit), making them well-suited for the two-level pipeline service provided by Cortex-M0+.

Cortex-M3 and Cortex-M4 are very similar cores, both with a performance of 1.25 DMIPS/MHz, equipped with a three-level pipeline, multiple 32-bit bus interfaces, clock speeds of up to 200MHz, and very efficient debugging options. The biggest difference between them is that the Cortex-M4’s core performance is optimized for DSP. Cortex-M3 and Cortex-M4 share the same architecture and instruction set (Thumb-2). However, Cortex-M4 adds a series of operations and SIMD instructions specifically optimized for processing DSP algorithms.

As the open road pioneer of RISC-V microprocessors, SiFive’s RISC-V cores are directly comparable to Arm’s three series of cores: E core – 32-bit embedded core, aimed at edge computing, AI, and IoT applications, comparable to ARM Cortex-M series; S core – 64-bit embedded core, aimed at storage, AR/VR, and robotics applications, comparable to ARM Cortex-R series; U core – 64-bit application processor, aimed at data centers, communication networks, and other fields, comparable to ARM Cortex-A series.

Domestic RISC-V processor core developer Chiplet Technology has also launched four processor cores of different performance levels, namely: N100 series processor core designed for ultra-low power and ultra-small area scenarios; N200 series 32-bit ultra-low power RISC-V processor designed for sensing, connectivity, control, and lightweight intelligent applications for IoT terminal devices; N300 series 32-bit ultra-low power RISC-V processor designed for high energy efficiency and requiring DSP and FPU characteristics, suitable for IoT and industrial control scenarios; N600 series 32-bit RISC-V processor aimed at real-time control or high-performance embedded application scenarios, suitable for AIoT edge computing, storage, or other real-time control applications.

Taiwan’s Andes Technology RISC-V processor series includes: 22 cores for entry-level MCUs used in small IoT and wearable devices, achieving the highest performance of 3.95 Coremark/MHz among peers, with its high performance and streamlined design suitable for processing protocol packets running at high data transfer rates; N25F core suitable for floating-point intensive applications, such as audio processing, advanced motor controllers, satellite navigation, high-precision sensor fusion, and advanced smart meters; 45 series cores adopt ordered 8-stage dual-issue superscalar technology, N-series supports RTOS applications, while D-series supports RISC-V’s SIMD / DSP instruction set (P extension instruction set).

3. Lower Power Consumption

Consumer electronics, wearable devices, and other battery-powered IoT terminals have strict requirements for low power consumption. System power consumption is one of the main considerations for IoT deployment, as many IoT devices in application scenarios are battery-powered and require sustainable use for more than 10 years. In many applications, the MCU spends most of its time in low-power sleep mode, only occasionally waking up to read sensor data or process and transmit data.

The power consumption of the MCU subsystem includes two parts – dynamic power consumption when the MCU is working (proportional to the processor’s main frequency) and static power consumption related to leakage current when the MCU is in sleep mode (mostly constant). Therefore, total power consumption is affected by working mode current, sleep mode current, and the duration of working mode. If the application remains off for most of the time, sleep current may be more critical than working current. Generally, 32-bit MCUs have higher main frequencies, resulting in higher working current, but their processing speed is also fast, allowing for faster completion of processing tasks and quicker entry into sleep mode to save power. Additionally, the current consumption of wireless transceivers in sleep mode, send, and receive modes is also a crucial factor determining overall system power consumption.

4. More Security

Beginning to focus on security issues reflects a field’s development to a higher stage, as the investment and yield of security technology are often disproportionate, and many times it also affects the development schedule. Integrating security technology from the MCU level allows downstream developers in the supply chain to incorporate security into the development life cycle, which is a relatively positive signal for the MCU ecosystem.

5. Wireless Connectivity

In recent years, there has been explosive growth in IoT (IoT) devices and wireless connected products (such as wireless sensors, smart meters, smart homes, and wearable devices). The costs of electronic devices such as sensors and processors have gradually decreased, while the addition of wireless connectivity and AI capabilities has made many products more “intelligent,” allowing them to communicate with each other without human intervention. However, successful AIoT products must meet specific application requirements, such as low power consumption, long wireless connection range, and higher computing processing capabilities.

Wireless MCUs will become standard processor chips in the AIoT era. International manufacturers like Silicon Labs have focused on the IoT application market for years, even spinning off other businesses to fully invest in IoT, and their integration and support of various wireless communication protocols are noteworthy. Domestic chip manufacturers like Espressif Technology and Lian Sheng De have also begun to integrate more wireless connectivity features into their chips, and they are expected to seize emerging IoT opportunities and become leading manufacturers in the AIoT era.

6. Smaller Size

To meet the needs of IoT applications, MCU developers need to achieve the best balance in performance, power consumption, and size (PPA). We have roughly introduced high performance and low power consumption designs, now let’s discuss the design challenges of small size. The basic requirement for IoT terminal nodes is small size, as these devices are usually confined to a very small footprint. For example, in wearable device design, small size and light weight are key to gaining customer acceptance.

Five Driving Forces for the Development of Domestic MCUs

For domestic MCU manufacturers, “domestic substitution” and “chip shortage” have become the driving forces pushing their MCU product lines into the supply chains of large and medium-sized OEM manufacturers, even the automotive supply chain. According to the ASPENCORE “Electronic Engineering Magazine” analyst team’s investigation, any domestic MCU manufacturer that can ensure normal supply from foundries and packaging and testing partners has enjoyed simultaneous growth in sales and profits.

In addition to the two macroeconomic and market driving forces mentioned above, the rise of emerging IoT applications, RISC-V processor architecture, and edge AI will also greatly promote the development of domestic MCUs. Below we explain these five driving factors one by one.

1. Domestic Substitution: Since the “ZTE incident” and Huawei’s ban, the US has gradually tightened export restrictions on China in semiconductor technology and products. Even with Trump’s departure, the Biden administration has shown no signs of easing in this regard. Many Chinese manufacturers, such as Hikvision, have been placed on the US government’s “entity list,” causing great panic in the domestic electronics and semiconductor industries. At the same time, domestic manufacturers have begun seeking “domestic substitution,” with procurement personnel from companies that previously never considered domestic chips actively inviting domestic IC design companies to participate in their product designs, engineering testing, and even mass supply. As a general-purpose basic device, MCU chips have a wide range of applications, and there are more opportunities for “domestic substitution.”

2. Chip Shortage: According to semiconductor industry experts, the global chip shortage is expected to ease by 2023. The chip shortage is most severe in application markets with a high semiconductor content, particularly in the automotive industry. Among automotive semiconductors, the shortage of automotive-grade MCUs, which serve as the automotive micro “brain,” is especially serious. The MCU usage in the automotive market accounts for 40% of the global MCU market, which indicates the large market size of automotive MCUs. With the development of new energy smart connected vehicles, the application of automotive-grade MCUs is becoming more widespread, covering everything from cockpit safety, engine transmission and braking control to dashboard, environmental control, in-car entertainment communication, and even ADAS and autonomous driving.

3. IoT: Emerging IoT applications have also brought many new opportunities for domestic MCUs, such as smart home appliances, consumer electronics, wearable devices, as well as industrial IoT markets like smart cards and smart meters. Traditional MCUs (such as 8051 and general-purpose 32-bit MCUs) have relatively simple functions, and their products are quite homogeneous, making it difficult to stand out in already mature markets, forcing domestic manufacturers to rely on price wars to survive. However, the application scenarios of IoT are relatively fragmented, and traditional general-purpose MCUs are often unable to meet the specific needs of different applications. Domestic manufacturers have begun attempting to integrate more peripheral functions onto microprocessor cores, such as ADCs, sensors, RF, drivers, and various peripherals. Responding quickly to market and customer needs, with strong flexibility, and the ability to add various functional features according to customer requirements is a strength of domestic MCU manufacturers.

4. RISC-V: The microprocessor core is the core of MCUs, and adding RISC-V alongside Arm allows domestic MCU manufacturers to achieve independence and control, increase flexibility in choices, and reduce development costs, while developing differentiated MCU products for specific application needs. Currently, the domestic RISC-V ecosystem is developing rapidly and healthily, with RISC-V core IP suppliers such as Alibaba Pingtouge, Chiplet Technology, Sifang, and Andes Technology actively expanding into the IoT market and domestic MCU manufacturer clientele. Organizations such as the China RISC-V Industry Alliance, the China Open Instruction Ecology (RISC-V) Alliance (CRVA), and the International RISC-V Foundation are also actively participating in the construction of the Chinese market and developer community.

5. Edge AI: Chip designs related to AI mainly involve powerful CPUs, GPUs, FPGAs, and specialized AI chips. How does a relatively low-frequency and computing performance MCU relate to AI? As AI expands from the cloud to the edge and terminals, AI computing engines can enable MCUs to break through the limits of embedded applications, not only enhancing the on-site processing performance of MCUs but also improving real-time response capabilities against network attacks and device security. MCUs equipped with AI algorithms and functional modules are penetrating emerging application fields such as facial recognition, intelligent voice services, and natural language processing. AI also helps improve the accuracy and data privacy of battery-powered devices in IoT, wearable, and medical applications.

Overview of 50 Domestic MCU Manufacturers

Click to read the original text and download the report PDF version!