Processor/DSP/FPGA-2024 World Electronics Achievement Awards-EE Times

Product：AXera Tongyuan AI Processor

Recommended reasons：

More Vote

Product：AMD Versal Gen2 adaptive SoCs product portfolio.

Recommended reasons： AMD launched new Versal™ Series Gen 2 devices, Versal AI Edge Series Gen 2 and Versal Prime Series Gen 2 adaptive SoCs integrate preprocessing, AI inference, and postprocessing together in a single device, enabling end-to-end acceleration of AI-driven embedded systems. The Versal AI Edge Series Gen 2 portfolio feature a powerful new AI engine, expected to deliver up to 3x higher TOPs-per-watt than first generation Versal AI Edge Series devices. While new high-performance integrated Arm^® CPUs are expected to offer up to 10x more scalar compute than first gen Versal AI Edge and Prime series devices. Balancing performance, power, area, together with advanced functional safety and security, Versal Series Gen 2 devices deliver new capabilities and features that enable the design of high-performance, edge-optimized products for the automotive, aerospace and defense, industrial, vision, healthcare, broadcast and pro AV markets.

More Vote

SHANGHAI ANLOGIC INFOTECH CO.,LTD.

Product：SALPHOENIX^® PH1A90

Recommended reasons： The PH1A90 device uses advanced 28nm technology, providing best-in-class transceiver and signal processing functions while maintaining low power consumption. It offers a rich set of sensor interfaces (MIPI, LVDS, Caramelink, CoXpress, etc.) and video interfaces (HDMI, SDI, DP, eDP, VBO, etc.), focusing on application areas such as TV/monitors, industrial cameras, audio and video broadcasting, and communication infrastructure, providing developers with a better solution to address more complex market demands. Key Features: - Logic scale of 115,776 LUTs - Supports two-input adders, three-input adders, multipliers, and arithmetic logic units - Maximum channel rate of 10.3125Gbps, supporting multiple protocols - Supports GEN1/2/3, as well as X1, X2, and X4 modes - Maximum support of 1066Mbps, with bit widths of x8/x16/x32 - Supports single-bit error detection and correction, as well as dual-bit error detection - Maximum support for MIPI 2.5Gbps DPHY-RX Compared to other products in the same category, the PH1A90 device considers cost control from chip design, packaging design, to production control, making it one of the most cost-effective devices supporting the 10G SERDES level.

More Vote

GOWIN Semiconductor Corp.

Product：GW5AT-LV60UG225

Recommended reasons： GW5AT-LV60UG225 is a high-speed integrated SERDES FPGA chip of Gowin's Arora V series. It adopts 22nm SRAM technology and has abundant internal resources, high-speed LVDS interface and rich BSRAM memory resources, at the same time integrated with self-developed DDR3, integrated 4-way 12.5 Gbps SERDES, support for PCIe 2.0 Hard Core, support for high-speed MIPI Hard Core, support for EDP, DP, SLVS-EC, SDI, USB 3.0 and many other protocols. It is mainly used in many fields such as multi-screen display, CMS system, laser radar, automatic driving/assistant driving, vehicle power control, domain control and so on. --Ultra-efficient low power advanced 22nm lp process low core voltage – 1.0v & 0.9v dynamic clock gating --Hardcore PCIe 3.0 supports x1, x2, x4, x8 lanes including Root Complex and End Point dual-mode --Hardcore MIPI D-PHY MIPI transmission rate can reach 2.5 Gbps (RX/TX) MIPI bandwidth 10Gbps with up to 4 data-lanes and 1 clock-lane --Hardcore MIPI C-PHY MIPI transmission rate can reach 2.5 Gbps (RX/TX) with up to 3 data-lanes -- Integrates two ADC: SARADC and ADC Sensor -- supports multiple SDRAM interfaces up to ddr3 1333Mbps --supports multiple I/O level standards

More Vote

XI’AN Intelligence Silicon Technology Limited Inc

Product：Seal_30k

Recommended reasons： At present, there is no benchmark product in the market for Seal-30k. This product is independently developed by our company, and the technical parameters are shown in the table below: Equivalent logical unit: 32666 Register: 40832 Distributed memory/shift register (Kbit): 653 Dedicated IO:8 User IO#:244 IO module (BANK): 4 Embedded DSP (18x18): 46 Embedded DSP (25x18): 23 Number of embedded memory (18Kbit): 69 Total capacity of embedded memory (Kbit): 1242 PLL:2 Global clock network: 24 GCLK PINs:24 ADC:1 (2 lines) DDR2/3 controller hardcore: 1 (16bits) MCU:M3 compatible Maximum user HR IO: 244 Note: No benchmarking products

More Vote

Mobileye Global Inc.

Product：EyeQ™6 Lite (EyeQ™6L)

Recommended reasons： Features, competitive advantages Benefits and Features for Drivers Worldwide The EyeQ6L combines Mobileye’s experience in designing automotive-grade processors and artificial intelligence/machine learning algorithms with its expertise in custom integrated software that optimizes performance and energy efficiency. Designed with two CPU cores and five high-compute density accelerators, the EyeQ6L provides 4.5 times more computing power than the EyeQ4M, at roughly half the physical space, with similar levels of power consumption – key elements for automotive use. The chip also improves pixel segmentation capabilities through a dynamic neural network with more than double the point density of EyeQ4M. Innovations of Mobileye’s new EyeQ6L • Our most powerful & cost-efficient SoC for basic driving-assist to date* • 4.5 times more powerful than previous high-volume ADAS chip • 8MP camera with 120° FoV for improved sensing and range capabilities • Sensing wet roads and adjusting AEB • Detecting and reacting to objects at greater distances • Reading key text phrases on road signs Target applications As Mobileye’s most powerful & cost-efficient SoC for basic driving-assist to date, the EyeQ6L launches to • Speed ADAS upgrades globally The EyeQ6L will power advanced driver-assistance systems in multiple models launching this year. This milestone marks the beginning of the EyeQ6 family, with the EyeQ6L already set to be installed in 46 million vehicles over the next few years – becoming the global auto industry’s ADAS solution of choice from the start. • Enable safety and convenience in 46 million vehicles “With the EyeQ6L, the team has once again delivered a system-on-chip that enables sizable gains in performance, safety, and comfort features to our customers, without a material increase in price, as has been our pledge for many years,” said Prof. Amnon Shashua, founder and CEO of Mobileye. “We know the power of ADAS to save lives and reduce traffic accidents globally, and with the EyeQ6L, automakers can meet regulatory requirements while delivering meaningful technology improvements to end users.”

More Vote

Moffett AI

Product：Moffett S40 Computing Card

Recommended reasons： Moffett S40 Compute Card is built for AI inference applications in the data center. As a general-purpose deep learning inference gas pedal, the form factor is in dual-slot, full-height, full-length PCIe x16 gold finger physical form factor. Moffett S40 Compute Card supports 80GB of LPDDR4x memory, with a theoretical peak memory bandwidth of 256GB/s, and a maximum power consumption of 300W.The passively cooled board design keeps the compute card operating within its thermal limitations and through system airflow. Moffett S40 Compute Card is built on Moffett AI Antoum^®️ architecture. The architecture is designed to emphasize balanced structured sparsity through tight integration of hardware and software, supporting high sparsity rates of up to 32x. Based on Antoum^®️ architecture, Moffett S40 compute card supports BF16 and INT8 computation. At the same time, Moffett S40 Compute Card supports end-to-end software solutions including software toolchain, compiler and runtime with integrated model sparser, ensuring that mainstream AI inference jobs can be quickly realized. The tightly integrated hardware and software design makes Antoum^®️ an efficient AI system-on-chip processor. In addition, Moffett S40 Compute Card supports hardware video codecs and JPEG decoders, enabling it to handle a variety of video and image application scenarios. Also, Moffett S40 Compute Card ships with the device with ECC enabled for the system DDR to prevent detectable errors in memory.

More Vote

SemiDrive

Product：Central Computing Processor X9CC

Recommended reasons： Evolution of Automotive Cockpits: From Digital to Integrated Information Display to AI Cockpits.The automotive cockpit has evolved from digital to integrated information display and is gradually achieving AI cockpits. SemiDrive, with its X9 series of cockpit processors, comprehensively covers the needs of cockpit processors across different eras. This includes applications ranging from entry-level to flagship-level, such as 3D instrument panels, IVI (In-Vehicle Infotainment), cockpit domain controllers, integrated cockpit-parking, integrated cockpit-driving-parking, and central computing platforms, actively leading the development of AI cockpit products. The X9SP, released in 2023, is the first-generation AI cockpit product from SemiDrive, featuring 8 TOPS of NPU computing power, enabling the local deployment and acceleration of AI algorithms. The new generation X9CC offers a higher-performance NPU unit capable of hybrid deployment of large models locally and in the cloud. Released in April 2024, the X9CC boasts up to 200 KDMIPS of computing power and integrates multiple high-performance computing cores in a single chip, including 24 Cortex-A55 CPUs, 12 Cortex-R5F CPUs, 2 NPUs, 4 GPUs, 4 Vision DSPs, and a Crypto engine supporting national cryptographic algorithms. To date, the shipment volume of SemiDrive's X9 series cockpit processors has exceeded 3.5 million units. Models equipped with the X9 series chips from automakers such as Chery, SAIC, Changan, GAC, BAIC, Dongfeng Nissan, and Dongfeng Honda have all entered mass production. With the advent of the AI cockpit era, SemiDrive's intelligent cockpit processors are expected to have broader development opportunities, providing innovative products and services to more automakers.

More Vote

Synaptics

Product：Synaptics Astra™ Platform

Recommended reasons： The award-winning Synaptics Astra platform with the SL-Series of embedded AI-native Internet of Things (IoT) processors and the Astra MachinaTM Foundation Series development kit provides the structure, scalability, and flexibility to meet the demands of OEMs and ODMs looking to incorporate AI in virtually any product from smart conferencing systems to wearable devices to intuitive automotive display panels, and beyond. The Astra Platform has been architected to address the needs of customers building a new generation of smart IoT devices. Benefits include: • Compute Solutions – allows product differentiation with power-efficient multi-modal AI-enabled MPUs and high-performance AI and connectivity MCUs. • Adaptive Framework – expedites the move from 'making AI work' to 'working with AI' with open, cross-platform, tooling, modeling & optimization infrastructure. • Unified Software Experience – leverages an open, consistent cross platform experience, supporting Linux, Android and the leading RTOS offerings. • Partner Solution Ecosystem – facilitates the development of market ready systems & solution stacks with industry-leading ODMs, ISVs, service providers and system integrators. • Wireless Connectivity – allows for the pairing of best-in-class, industry proven options spanning Wi-Fi, Bluetooth, BLE, 802.15.4, & GPS/GNSS. The Astra Platform’s SL-Series of processors allows designers to bring AI directly to their products independent of the data center, solving for data privacy and latency. This family of embedded processors provides industry-leading compute capability at power-consumption levels that enable a wide range of consumer, enterprise, and industrial edge IoT applications. The SL-Series multi-core Linux or Android systems on chip (SoCs) are based on Arm^®Cortex^®A-series CPUs and features hardware accelerators for edge inferencing and multimedia processing on audio, video, vision, image, voice, and speech. Each of the three devices in the SL-Series offers unique features and benefits, catering to different multi-modal application requirements and feature ‘out-of-the-box’ AI, including: • The SL1680 processor is based on a quad-core Arm Cortex-A73 64-bit CPU, a 7.9 TOPS NPU, a high-efficiency, feature-rich GPU, and a multimedia accelerator pipeline. It is ideal for home and industrial control, smart appliances, home security gateways, digital signage, displays, point-of-sale systems, and scanners. • The SL1640 processor is optimized for cost and power and is based on a quad-core Arm Cortex-A55 processor, a 1.6+ TOPS NPU, and a GE9920 GPU. It is ideal for smart home appliances, enterprise conferencing, smart speakers, displays and signage, consumer, and industrial control panels. • The SL1620 processor is also based on a quad-core Arm Cortex-A55 CPU subsystem, a feature-rich GPU for advanced graphics and AI acceleration, superior audio algorithms, and dual displays. Applications include enterprise multimedia conferencing, smart appliances, home security gateways, digital signage, displays, point-of-sale systems, and smart speakers. The Astra Machina Foundation Series development kit supports the Astra Platform’s SL-Series of processors. The kit helps AI beginners and experts quickly unlock the processors’ superior AI capabilities, powerful processing and graphics performance, and matching wireless connectivity, including Synaptics’ Wi-Fi^® and Bluetooth^® combo SoCs.

More Vote

UNISOC (Shanghai) Technology Co., Ltd.

Product：UNISOC first 5G IoT-NTN satellite communication SoC V8821

Recommended reasons： UNISOC first 5G IoT-NTN satellite communication SoC V8821, is based on the 3GPP NTN R17 standard and leverages the IoT NTN network as its infrastructure, facilitating seamless integration with terrestrial core networks. By utilizing L-band maritime satellites and S-band Tiantong satellites, V8821 offers functionalities such as data transmission, text messaging, voice calls, and location sharing. It also supports expansion to access other high-orbit satellite systems, making it ideally suited for communication needs in areas with limited cellular network coverage, including oceans, urban fringes, remote mountainous regions, and more. V8821 supports the TCP/IP protocol, enabling compatibility with various upper-layer services and inheriting common IoT protocols for quick integration into existing data platforms. In terms of hardware configuration, V8821 flexibly supports customers in reusing hardware resources from existing solutions, enhancing deployment efficiency and avoiding issues like high hardware costs and lengthy timelines. UNISOC participated in and completed China's first 5G NTN mobile phone direct-to-satellite field validation. By adopting vivo's highly integrated RF antenna solution and a vivo satellite communication terminal prototype embedded with UNISOC's V8821 chip, the team achieved 5G NTN mobile phone direct-to-satellite uplink and downlink connections under the Tiantong satellite environment. This demonstration successfully enabled intercommunication between vivo phones, showcasing the transmission and reception of text messages via satellite communication, with functionality and performance meeting expectations. IoT based on satellite communication addresses pain points such as the inability to transmit information monitoring data in areas with no cellular communication signal coverage. UNISOC participated in and completed multiple field tests of IoT application scenarios based on Tiantong satellites in regions like Yanqing, Beijing, and Dianshan Lake, Suzhou. These scenarios included water surface buoys, satellite messengers, and satellite-terrestrial dual-mode trackers. By leveraging the satellite communication chip V8821 embedded in the terminals and Quectel's latest satellite communication module CC950U-LS, the collected sensor data, emergency messages, and terminal location data were successfully transmitted to cloud platforms in real-time via satellite, achieving satisfactory results.

More Vote

Black Sesame Technologies Co.,Ltd.

Product：Wudang series C1296 cross-domain computing chip for intelligent vehicles

Recommended reasons： Through heterogeneous isolation technology, different computing power can be selected to use according to different scenarios, as well as different specifications and safety requirements, which can support the flexible development of automotive electronic and electrical architectures, and support various architectural schemes such as compute clusters, Cockpit-AD integration, and central compute.Target a massive L2+ level cross-domain computing market. To balance performance and cost efficiency, our design goal is to provide human-machine interaction, driving-and-parking integration, and data exchange capabilities through a single chip. Based on the most advanced semiconductor technology in the industry, we ensure the balance of computing power, power consumption, and cost. We platformilized the product development and met the needs of each scenario through a set of SDK, saving development time and subsequent long-term maintenance costs. At the same time, the development tools are designed and applied towards the whole product family, and the deep learning toolchain based on the Huashan series is reused, and one-click migration of the algorithm ensures that the software assets of existing Huashan series customers are inherited.

More Vote

Ehiway Microelectronic Science and Technology (Suzhou) Co.Ltd.

Product：eFPGA IP Core Compilation Technology Based on Domestic Processes

Recommended reasons： The embedded FPGA IP core compilation technology based on domestic process developed by Ehiway can quickly tailor and form IP cores of different capacities and types according to user application needs on the hard core, and provide SoC users with integrated use; It provides an evaluation process in the software, helps customize IP resources, generates a chip database, and provides EDA whole-process support, filling the gap in the design, integration and EDA technology of China's independent process FPGA IP core. Based on this technology, it has provided solutions for a number of SoC users, formed tens of millions of yuan of technology development orders, and developed a variety of domestic FPGAs have been promoted and applied to more than 300 users, playing an important role in the localization of chips in important markets such as industrial control, oil exploration, secure communication, and financial acceleration.

More Vote

Elitestek (Shenzhen) Technology Co., Ltd.

Product：TJ-Series FPGA

Recommended reasons： English version: Elitestek 16nm TJ-Series FPGA Main specifications and features: low power consumption, high performance, flexible interface support and embedded RISC-V processing. Logic capacity covers the low-end and mid-range FPGA market (35K to 375K); supports multiple MIPI protocols: 1.5G MIPI and 2.5G MIPI hard cores; supports high-bandwidth memory interfaces such as 3.7Gbps LPDDR4/4x; supports 16Gbps Serdes high-speed interface; supports 1/2/4-core soft-core RISC-V configuration, and supports 4-core hard-core embedded RISC-V processors. Main application areas: has successfully obtained market-leading shares in machine vision, smart cameras, thermal imaging sensors and other fields. In portable medical ultrasound applications, it has become the mainstream solution in the market with its power consumption advantage far exceeding that of its competitors. In the automotive LiDAR market, the perfect balance between high performance and low power consumption of the TJ-Series has achieved innovative success for the industry's leading customers. Main advantages and competitive product analysis: Through the optimized Quantum architecture, the TJ-Series has extremely low power consumption while maintaining high performance (the main frequency can reach more than 400MHz), and the power consumption is 1/2~1/3 of the same-level competitors. Compared with 28nm competitors of the same scale, the performance is improved by 40~50%. In terms of bandwidth/area ratio, the TJ-Series has a 3~4 times advantage over competitors due to its small size and high bandwidth. In terms of RISC-V support, the entire TJ-Series supports the soft-core RISC-V configuration independently optimized by Elitestek, and some products have built-in hard-core RISC-V cores (without any authorization/license risk). The mainstream competitors in the industry still use ARM as the mainstream choice for embedded systems. The TJ-Series has significant advantages in open source ecology, independent innovation, and supply chain security.

More Vote

Hercules Microelectronics Co. Ltd.

Product：HME-H7P20

Recommended reasons：

More Vote

Qualcomm

Product：Snapdragon 8 Gen 3 Mobile Platform

Recommended reasons： Snapdragon 8 Gen 3 is Qualcomm Technologies, Inc.’s first premium mobile platform meticulously designed with generative AI in mind, launched in October 2023. It infuses high-performance AI across the entire system to deliver premium-level performance and extraordinary experiences to consumers, and unlocks a new era of generative AI. Snapdragon 8 Gen 3 features the most powerful on-device intelligence Qualcomm has ever integrated into a smartphone platform, supports large language and vision models, and generative AI applications. It can run generative AI models up to 10 billion parameters and large language models at up to 20 tokens per second, bringing more responsive, more efficient, and safer AI experiences. The AI features of Snapdragon 8 Gen 3 brings a skyrocketing boost in image quality, and transform the camera into a powerful handheld sensor, continuously empowering premium image experiences. In addition, Snapdragon 8 Gen 3 integrates several industry-leading technology features, including the world's first 5G Advanced-ready modem-RF system to usher in a new phase of cellular communications with breakthrough 5G performance; with Snapdragon Elite Gaming, it brings even more desktop-level gaming features to smartphones. Snapdragon 8 Gen 3 has been widely adopted by leading OEMs and smartphone brands in the Chinese market, including ASUS, Honor, iQOO, MEIZU, Nubia, OnePlus, OPPO, realme, RedMagic, Redmi, Samsung, vivo, and Xiaomi, with more than 20 flagship smartphone products launched. It supports rich GenAI use cases and serves as a strong catalyst for market development and experience innovation for Android flagship smartphones.

More Vote

Renesas Electronics

Product：RZ/V2H MPU

Recommended reasons： The RZ/V2H comes with a new generation of Renesas proprietary AI accelerator, DRP (Dynamically Reconfigurable Processor)-AI3, delivering 10 TOPS/W power efficiency, an impressive 10-fold improvement over previous models. Additionally, pruning technology employed in the DRP-AI3 accelerator significantly improves AI computing efficiency, boosting AI inference performance up to 80 TOPS. This performance boost allows engineers to process vision AI applications directly at edge AI devices without relying on cloud computing platforms. The details of the new DRP-AI3 acceleration technology were recently presented at the International Solid-State Circuits Conference (ISSCC 2024) in San Francisco. The RZ/V2H incorporates four Arm^® Cortex^®-A55 CPU cores with a maximum operating frequency of 1.8 GHz for Linux application processing, two Cortex-R8 cores running at 800 MHz for high-performance real-time processing, and one Cortex-M33 as a sub core. By integrating these cores into a single chip, the device can effectively manage both vision AI and real-time control tasks, making it ideal for demanding robotics applications of the future. Since the RZ/V2H consumes less power, it eliminates the need for cooling fans and other heat-dissipating components. This means engineers can design systems that are smaller in size, less expensive, and more reliable. The RZ/V2H which combines AI technology with real-time control, will facilitate the development of next-generation autonomous robots with vision AI capabilities, that have the ability to think independently and control movements in real time.

More Vote

STMicroelectronics

Product：STM32MP25 Industrial edge computing processor

Recommended reasons： These are ST’s first MPUs to contain a 64-bit central processing unit (CPU), the Arm Cortex^®-A35, which runs up to 1.5GHz to raise the main processing capability compared to the first-generation STM32MP1 devices. The CPU is the center of a true heterogeneous processing engine that also contains a Cortex-M33 core. In addition, there is a graphics processor (GPU), neural processor (NPU), and a video processor (VPU). AI workloads can run on the CPU, GPU, or NPU depending on processor loading and application demands, for optimal performance and energy efficiency. In fact, the high efficiency of these MPUs permits system designs to go without active cooling, gaining advantages such as smaller size, silent operation, greater reliability, and reduced power consumption. The 3D GPU supports up to 1080p displays and the powerful multimedia features also include a full-HD video pipeline with parallel LVDS and DSI interfaces. Together with a MIPI CSI-2 camera interface with ISP, these enhance support for cutting-edge machine-vision applications. There are further powerful industrial interfaces, including up to three Gigabit Ethernet ports with a 2-port switch and support for Ethernet time-sensitive networking (TSN). There are also PCIe Gen2, USB 3.0, and three CAN-FD interfaces that ease integration in a wide variety of communication and control applications. The devices are fully supported with STM32 development resources that are familiar to engineers working with ST’s STM32 microcontrollers (MCUs) and STM32 microprocessors. The STM32Cube ecosystem provides tools and dedicated MPU software packs including OpenSTLinux and OpenSTDroid for Android support. There are also evaluation boards that provide further help to simplify development and accelerate project completion. "

More Vote