Choosing Domestic FPGA Chips: An Overview of Alternatives to Xilinx

01

Introduction

This article discusses the current level of FPGA development in China, the companies that are doing well, and the factors to consider for domestic FPGA alternatives.

Some content and data are referenced from the internet and official websites. If there are inaccuracies, I hope to discuss and learn from experts.

FPGA, The Universal Chip! With its powerful parallel computing capabilities and customizable functions, it is widely used in communications, medical, power, and military industries, as well as in the early prototype verification systems for IC and ASIC design.

Since its invention by one of the founders of Xilinx, Ross Freeman, in 1985, 90% of the global FPGA market has been occupied by foreign manufacturers.

The market prospects for FPGA are extremely enticing, yet the barriers to entry in the chip industry are unparalleled. Since the invention of FPGA, over 60 companies worldwide have invested billions of dollars to attempt to conquer this high ground, including industry giants like Intel, IBM, TI, Motorola, Philips, TOSHIBA, and SAMSUNG. However, only four companies located in Silicon Valley, USA—Xilinx, Altera, Lattice, and Microsemi—have successfully reached the summit, with the first two holding nearly 90% of the global market share and possessing the vast majority of patents in the FPGA field, forming an insurmountable technological barrier.

Due to the global pandemic, both upstream and downstream FPGA chip production companies have been affected, leading to a significant increase in FPGA chip prices. Some mid-range FPGA products with high usage have seen prices increase by dozens of times, resulting in a situation where they are available but in short supply! Interestingly, some high-end, high-performance FPGA chips are not as expensive, proving that the market determines prices!

02

Which Domestic FPGA Manufacturers Are There?

Although the FPGA chip industry has a high technological barrier, China has not ceased its exploration of FPGA technology, transitioning from reverse engineering to independent research and development, and from military applications to the broader industrial and civilian sectors.

The mainstream FPGA manufacturers currently include the following companies, whose products are primarily used in communications, industrial, military, and consumer electronics fields.

1. Unisoc (Shenzhen)

https://www.pangomicro.com/

Choosing Domestic FPGA Chips: An Overview of Alternatives to Xilinx

Unisoc’s FPGA products are divided into three main series:Titan, Logos, and Compa series.

Choosing Domestic FPGA Chips: An Overview of Alternatives to Xilinx
  • Titan Series

High-speed, High-performance. The Titan series is China’s first domestically produced high-performance FPGA product with independent intellectual property rights, featuring 174K equivalent LUT4 units, a maximum frequency of 500MHz, 5.0Gbps SERDES interface, 800Mbps DDR3, and LVDS, PCIe Gen2x4, suitable for communication networks, information security, data centers, and industrial control.

  • Logos Series

High cost-performance ratio. With a new LUT5 structure, it integrates rich on-chip resources such as RAM, DSP, ADC, SERDES, and DDR3, supporting various standard IO, LVDS, and MIPI interfaces, widely used in industrial control, communications, and consumer products, making it an ideal choice for large-volume, cost-sensitive projects.

  • Compact Series

CPLD products, low power consumption, low cost. The Compa series CPLD products are low-power, low-cost, and small-sized, supporting MIPI, LVDS, I2C, SPI, OSC, RAM, PLL, etc., with up to 383 user IO, supporting 3.3/2.5V cores or 1.2V low-voltage cores, suitable for system configuration, interface expansion and bridging, board-level power management, power-up timing management, and sensor fusion applications.

Unisoc’s FPGA product ecosystem is quite good, with chip documentation manuals, evaluation boards and downloaders, EDA software and licenses, IP core resources, offline training, university programs, and competition sponsorships all performing well.

The following image shows the development board jointly produced by Unisoc and the Black Gold Power Community:

Choosing Domestic FPGA Chips: An Overview of Alternatives to Xilinx

2. Anlogic (Shanghai)

http://www.anlogic.com/

FPGA has two main series: SALEAGLE and SALELF series

  • SALEAGLE Series

Divided into two product models: EG4 and AL3.

Choosing Domestic FPGA Chips: An Overview of Alternatives to Xilinx

EG4: 19600 LUTs, 55nm process, minimum static power consumption of 5.5mA, DSP, BRAM, high-speed differential IO, with user IO ranging from 71 to 193, on-chip 8-bit ADC, 1M sampling rate, 8-channel input.

AL3: 8640 LUTs, 65nm process, minimum static power consumption of 4mA, user IO ranging from 60 to 184, with rich DSP, BRAM, high-speed differential IO resources, and strong pin compatibility replacement performance.

  • SALELF Series

Small Elf series FPGA, with three generations of products, single-chip solutions, instant startup without external Flash, supports OTP mode, 55nm process, some product models have embedded hard-core MCU.

Choosing Domestic FPGA Chips: An Overview of Alternatives to Xilinx

Anlogic also independently developed the FPGA integrated development environment—TangDynasty(TD), supporting standard design input methods, a complete circuit optimization process, and rich analysis and debugging tools, providing strong support for application design based on Anlogic FPGA products.

3. Gaoyun Semiconductor (Guangzhou)

http://www.gowinsemi.com.cn/

FPGA products mainly include two series: Morning Series and Little Bee Series.

Choosing Domestic FPGA Chips: An Overview of Alternatives to Xilinx
  • Morning Series

55nm SRAM process, with high-performance DSP resources, high-speed LVDS interfaces, and rich B-SRAM memory resources, suitable for high-speed low-cost applications.

  • Little Bee Family

Low power consumption, low cost, instant startup, and high security non-volatile editable logic devices. Including SoC products and non-SoC products, SoC products embed ARM Cortex-M3 hard-core processors, GW1NS series products integrate USB2.0 PHY, user flash memory, and ADC converters.

SoC products achieve seamless connection between programmable logic devices and embedded processors, compatible with various peripheral device standards, significantly reducing user costs, and can be widely used in industrial control, communications, IoT, servo drives, consumer markets, and more.

GW1NRF series Bluetooth FPGA devices, centered around a 32-bit hard-core microprocessor, support Bluetooth 5.0 low-power RF functions, with rich logic units, embedded B-SRAM and DSP resources, abundant IO resources, and internal power management and security encryption modules.

4. Fudan Micro (Shanghai)

http://www.fmsh.com

Fudan Micro is one of the leading companies in the domestic FPGA field, currently offering products in three series: tens of millions of gate-level FPGA chips, hundreds of millions of gate-level FPGA chips, and embedded programmable devices (PSoC). Fudan Micro’s hundreds of millions of gate-level FPGA chips are based on a 28nm process and utilize advanced CMOS technology, being the first domestically developed hundreds of millions of gate-level FPGA chips, which are now in mass production and sales.

In 2019, Fudan Micro launched its independently developed tool software Procise, which is the first ultra-large-scale full-process EDA design tool in the domestic FPGA field, with a user-friendly interface, powerful functions, and ease of use. However, due to its later release compared to Xilinx’s Vivado, its version iteration is relatively few, and the functionality of the development tool and available IP library still need further improvement.

The company is also actively developing the next generation of 14/16nm process 1 billion gate-level products while combining CPU and AI technology to develop PSoC chips, expanding into new markets and maintaining the company’s leading position in domestic FPGA technology.

However, compared to industry leader Xilinx’s revenue of $3.148 billion as of April 2021, the company’s revenue in 2020 was only $15.3 million, indicating a significant gap in operational scale.

The following image shows a development board based on the Fudan Micro FMQL45T900 chip, which can replace the Xilinx ZYNQ FPGA 7045, integrating FPGA and a quad-core ARM Cortex-A7.

Choosing Domestic FPGA Chips: An Overview of Alternatives to Xilinx

5. Zhiduojing (Xi’an)

http://www.isilicontech.com/

Divided into three series: Seagull 1000, Seagull 2000, and Seagull 5000 series.

Choosing Domestic FPGA Chips: An Overview of Alternatives to Xilinx
  • Seagull 1000 Series

64, 128, and 256 logic units available, 0.162um process, maximum frequency Fmax=322MHz

  • Seagull 2000 Series

5K, 12K, and 25K logic units available, low-power 55nm process, built-in hard-core DSP, built-in Flash, LVDS maximum transmission rate of 840Mbps, receiving rate of 875Mbps, supports 400Mbps DDR2 SDRAM interface, and supports common LVDS, LVCMOS, LVTTL IO standards.

  • Seagull 5000 Series

Devices with 30K to 325K logic units, up to 500 user IO, LUT6 structure, advanced 28nm copper CMOS process, maximum frequency 500MHz, hardware multipliers, LVDS interfaces up to 1.6Gbps, embedded hard-core ARM, ADC, DDR2/3 controllers.

6. Jingwei Qili (Beijing)

http://hercules-micro.com/

Divided into: HME-R (He), HME-M (Huashan), HME-P (Feima), and HME-H (Hercules) four series. Currently, it has fully achieved mass production at 65/55/40nm, with 22nm scale mass production starting in 2022.

  • HME-R Series

Low power consumption, high cost-performance ratio, 40nm process, LUT4 structure, logic capacity of 1-3K, mainly aimed at low power consumption application fields, embedded memory, minimum support for 1.5mmx1.5mm packaging.

  • HME-H Series

Integrating ARM Cortex-M3 hard-core controller and high-performance FPGA, ARM core maximum clock 300MHz, logic performance maximum 200MHz, hardware integration of Ethernet, USB, CAN, DMA controllers, and DDR controllers, meeting the “customizable, reconfigurable, programmable” design needs of different applications, realizing the SoC of FPGA.

  • HME-P Series

TSMC 40nm CMOS process, new LUT6 structure, 6.5Gbps Serdes high-speed I/O, 1333Mbps hard-core DDR2/3 control and PHY, high-speed AXI, PCIe, DDR2/3 hard-core IP, targeting the high-speed, high-performance large-capacity FPGA market.

  • HME-H Series

Integrating high-performance FPGA, enhanced MCU, and MIPI interfaces into intelligent video bridging devices, enhanced 8051 MCU, embedded DSP units, mainly aimed at the video processing field, widely applicable in mobile phones, tablets, wearables, VR, AR, drones, and smart homes.

FX-Fuxi software is an EDA software independently developed by Jingwei Qili, capable of FPGA and SoC application design, and supports collaboration with other common third-party design tools. It features a user-friendly graphical interface, rich IP, command line operation support, remote download programming, bitrate encryption, and on-chip debugging, etc.

Domestic FPGA manufacturers are still in the development phase, with limited market share and revenue scale, but they can be said to be the hope and future of the country.

03

Factors to Consider for Domestic Substitution

For the large FPGA market in China, if we are to implement FPGA domestication solutions, the following factors need to be considered:

1. Process Technology, Gate Scale, and SerDes Rate

These three parameters are also important indicators of FPGA’s basic performance. The process technology directly affects the chip’s power consumption, performance, and cost. If the power consumption is too high, then power and heat dissipation issues must be considered in hardware design. The process also affects gate scale; higher process technology allows for larger gate-level circuits on the same wafer size, thus providing more design space. The transmission rate of SerDes affects the performance of FPGA in high-speed data transmission and processing.

Taking Xilinx, which has a high market share, as an example, the latest UltraSCALE+ series FPGA chips have reached a process technology of 16nm, while most domestic FPGA manufacturers’ products are still at a 28nm process.

Choosing Domestic FPGA Chips: An Overview of Alternatives to Xilinx

2. Reliability, Stability, and Consistency

FPGA is typically applied in scenarios requiring high-speed, real-time processing, where reliability and stability are crucial. How does the chip perform under different temperatures, humidity, vibrations, and salt mist? How long can the chip’s lifespan last? Are the performance parameters of each chip consistent within a certain range? These are issues that need to be considered when implementing domestic alternatives.

3. Compatibility of Chips and Intellectual Property Issues

Domestic FPGA has transitioned from reverse engineering to independent research and development, with some domestic FPGA manufacturers providing fully compatible FPGA chips for certain models of Xilinx and Altera. For such products, the compatibility of FPGA hardware and software must be considered.

In terms of hardware, is it pin-to-pin compatible, allowing direct replacement without circuit modifications, or does it require some changes, such as impedance matching for high-speed interfaces and trace lengths?

On the software side, compatibility of development tools must be considered, such as FPGA development, debugging, and downloading tools, MCU development tools, IP cores, RTL-level code, and primitives compatibility, etc., whether additional patches need to be installed on the original development environment for adaptation.

Moreover, if your products need to be exported for sales abroad, using compatible FPGA chip models may involve intellectual property issues, so communication with chip manufacturers is necessary for confirmation.

4. Ecosystem of Self-Developed Chips

If it is a completely independently developed FPGA chip, the ecosystem of the chip needs to be considered, including the use of graphical EDA development tools, support for third-party tools like ModelSim, whether it supports mixed programming of Verilog/VHDL, the richness of provided IP cores, development boards, chip manuals/application documents, etc.

5. Cost Performance and Supply Sources

The cost performance of chips is a very important factor. Compared to Xilinx and Altera, if domestic FPGA chips have a price advantage for comparable performance, I believe many users will choose to implement domestic alternatives.

From a design perspective, it is also necessary to consider whether the chip’s power requirements, peripheral circuits, impedance traces, and packaging are in line with common design requirements.

From a supply chain perspective, factors such as the stability of the chip’s supply and supply cycle need to be considered.

04

Conclusion

Compared to FPGA giants like Xilinx and Altera, domestic FPGA research and development started relatively late, but the gap is gradually narrowing, and the difference with leading manufacturers has been reduced from three generations to two. With the current strong support from the state for the chip industry, and the widespread application of FPGA in 5G and AI fields, these circumstances present a rare opportunity for domestic FPGA manufacturers.

05 Carefully Recommended A training class has been developed in collaboration with friends, suitable for beginners. For students or friends who are confused about their future, this is worth a try, whether for a career change or those in the same field. As long as you complete the course seriously, getting a good offer is absolutely within reach. Interested students please add the WeChat below for inquiries:Choosing Domestic FPGA Chips: An Overview of Alternatives to XilinxCourse Outline:Choosing Domestic FPGA Chips: An Overview of Alternatives to Xilinx

Leave a Comment