Intel® Agilex™ SoC FPGAs provide the agility and flexibility to address a broad range of markets with tailored solutions.
To meet the needs of high-end applications with the most demanding performance requirements, Intel offers the Intel® Stratix® series. The Intel® Arria® series balances cost and power with performance for midrange applications. The Intel® Cyclone® series provides low system cost and power coupled with performance levels that make the device family ideal for differentiating high-volume applications.
The Intel® Agilex™ SoC FPGA family manufactured on Intel’s 10nm technology, integrates the quad-core Arm* Cortex*-A53 processor, features a hardened variable precision DSP, and delivers significant improvements in power and performance1 for a wide array of applications which require high system integration.
Intel® Stratix® 10 SoC FPGAs offer breakthrough advantages in bandwidth and system integration, including a next-generation hard processor system (HPS). Intel® Stratix® 10 SoC FPGAs feature the revolutionary Intel® Hyperflex™ FPGA Architecture and are manufactured on the Intel 14 nm Tri-Gate process, delivering breakthrough levels of performance and power efficiencies that were previously unimaginable. When coupled with 64 bit quad-core ARM* Cortex*-A53 processor and advanced heterogeneous development and debug tools such as the Intel® SDK for OpenCL™ 2 and SoC Embedded Design Suite (EDS), Intel® Stratix® 10 SoC FPGAs offer the industry’s most versatile heterogeneous computing platform.
The 20 nm ARM-based Intel® Arria® 10 SoC FPGAs deliver optimal performance, power efficiency, small form factor, and low cost for midrange applications. The Intel® Arria® 10 SoC FPGAs, based on TSMC’s 20 nm process technology, combine a dual-core ARM* Cortex*-A9 MPCore* HPS with industry-leading programmable logic technology that includes hardened floating-point digital signal processing (DSP) blocks. By utilizing the same dual-core ARM* Cortex*-A9 processor as the Arria® V SoC FPGA, the Intel® Arria® 10 SoC FPGA offers an easy performance upgrade and software migration path for Arria® V SoC FPGA designs.
Arria® V SoC FPGAs provide the highest bandwidth with the lowest total power for midrange applications such as remote radio units, 10G/40G line cards, medical imaging, and broadcast studio equipment. The combination of a HPS consisting of a dual-core ARM* Cortex*-A9 processor, peripherals, and memory interfaces with our flexible 28 nm FPGA fabric lets you reduce system power, cost, and board space.
Cyclone® V SoC FPGAs provide the industry's lowest system cost and power. The SoC FPGA high performance levels are ideal for differentiating high-volume applications such as industrial motor control drives, protocol bridging, video converter and capture cards, and handheld devices. SoC FPGAs come in a wide range of programmable logic densities with many system-level functions hardened in silicon-a dual-core ARM* Cortex*-A9 HPS, embedded peripherals, multiport memory controllers, serial transceivers, and PCI Express* (PCIe*) ports.
Benefits of SoC FPGAs
Intel® SoC FPGAs integrate an ARM*-based hard processor system (HPS) consisting of processor, peripherals, and memory interfaces with the FPGA fabric using a high-bandwidth interconnect backbone. It combines the performance and power savings of hard intellectual property (IP) with the flexibility of programmable logic. These devices include additional hard logic such as PCI Express* Gen2 and Gen3, multiport memory controllers, error correction code (ECC), memory protection, and high-speed serial transceivers. The ARM-compatible software provides unmatched target visibility, control, and productivity using our FPGA-adaptive debugging.
Building a product with a strong architecture is key to ensuring that your system design meets its performance requirements now and into the future. With our SoCs for embedded systems, you begin with a solid foundation that brings your design:
- Improved system performance through a higher hard processor system (HPS) to FPGA bandwidth interconnect, hardware acceleration, and increased memory performance
- Increased reliability through ECC and memory protection that help protect systems against potential hardware or software errors and warm/cold CPU reset that initiates without affecting or reprogramming the FPGA
- More flexibility through hardware and software differentiation, system boot and configuration options, and multiple hardened memory controllers
- Lower system cost through single-chip integration, integrated PCIe* controller, and no power off sequencing
- Increased productivity through our FPGA-adaptive debugging tools with unmatched target visibility, control, and productivity
- Path for the future through our roadmap for high-end, mid-range, and low-end applications, forward migration of software, and products with average life cycles of 15 years or more
Not All SoC FPGAs Are Created Equal. Architecture Matters.
Learn how to choose the right SoC FPGA for your application from our extensive set of resources, including a short series of videos from processor expert Jim Turley.
ARM* Cortex*-A53 MPCore* Processor
Intel® Stratix® 10 SoCs that are manufactured on Intel’s 14 nm FinFET process technology, feature our third-generation hard processor system (HPS) based on a quad-core ARM* Cortex*–A53 MPCore* processor cluster. The hard processor system (HPS) also includes a deep feature set of peripherals and is combined with the ground-breaking Intel® Hyperflex™ FPGA Architecture to create the industry's highest performance SoC FPGA product family.
ARM* Cortex*-A9 MPCore* Processor
A dual-core ARM* Cortex*-A9 MPCore* processor is the heart of the Cyclone® V SoC FPGA, Arria® V SoC FPGA, and Intel® Arria® 10 SoC FPGA. All three devices make use of the same high-performance processor, but with increased clock speeds and performance in the Arria® V SoC FPGA and even more so in the Intel® Arria® 10 SoC FPGA. Because the three devices use essentially the same processor, the Cyclone® V SoC FPGA can effectively be used for early prototyping and software development for systems based on any of the three SoC variants.
Product Documentation and Support
- Get the latest product documentation on the Intel® Stratix® 10 SoC FPGA
- Get the latest product documentation on the Arria® V SoC FPGA
- Get the latest product documentation on the Cyclone® V SoC FPGA
- Get the latest product documentation on the Intel® Arria® 10 SoC FPGA
The hardware design flow for the Intel® SoC FPGA includes configuring the hard processor system (HPS) and adding logic to the FPGA portion of the device. The Platform Designer (formerly Qsys), part of the Intel® Quartus® Prime Design Software, performs both tasks. The Platform Designer (formerly Qsys) automatically generates an optimized network on a chip (NoC) within the FPGA, including interfaces to the HPS, to create a custom system on a chip (SoC).
- Learn about the Platform Designer (formerly Qsys)
- Learn about the Intel® Quartus® Prime Design Software
- Download the Intel® Quartus® Prime Design Software (includes the Platform Designer)
SoC FPGAs leverage the rich ARM* embedded software ecosystem including operating systems, middleware, and software development tools.
- Learn more about the embedded software tools available for SoC FPGAs with the integrated ARM*-based hard processor systems (HPS)
Development Kits and Board
The SoC FPGA Development Kits are preconfigured with Linux and a reference design example called the Golden System Reference Design. As such, it is simple to unpack the board and contents, connect the power supply, and any required communication cables, such as Ethernet, UART, or USB. After powering-up the board, it will immediately boot and run useful examples. There is no need to download any additional tools or software to perform the initial power-up of the board.
After the initial power-up, there are a number of steps to follow. You can download software, tools, and additional examples and begin building and running applications on the board. These options are covered in the board-specific Quick Start Guide.
Select the specific development kit to view the detailed Quick Start Guide for that board.
Frequently Asked Questions
Processors in SoC FPGAs can be “hard” or “soft." Hard processors are implemented in the fixed silicon logic of the SoC FPGA similar to serial transceivers. On SoC FPGAs, however, the processor is surrounded by programmable logic that you can use for custom or application-specific functions. Hard processors offer higher CPU performance than soft processors, depending on factors such as processor architecture, clock rate, and process technology. As the name implies, hard processor feature sets are fixed and typically offered only as a variation of a particular SoC FPGA. The number and type of hard processors within an SoC FPGA are also fixed as a function of that particular SoC FPGA. Altera® offers hard processors in Intel® Stratix® 10 SoC FPGA, Intel® Arria® 10 SoC FPGA, Arria® V SoC FPGA, and Cyclone® V SoC FPGA families.
Soft processors, such as the Nios® II processor, are implemented in programmable logic, use on-chip resources such as logic elements, multipliers, and memory, and can be instantiated in almost any FPGA family. The performance and cost of a soft processor depend mainly on the FPGA in which the processor is instantiated, but performance and cost are typically lower than in hard processors. The number of soft processors that can be instantiated in a single device is limited only by the device’s resources (that is, its logic and memory). High-density FPGAs, for example, can contain hundreds of soft processors. Likewise, different types of soft processors can be implemented: 16 or 32 bit, performance optimized, logic-area optimized, and so on. You can choose to migrate your soft processor designs to hard processor implementations when moving to gate arrays or cell-based designs. One or more soft processors can likewise be used in the FPGA portion of an SoC FPGA.
There are many ways to use FPGAs in an embedded system. Typical uses include:
- I/O and peripheral expansion-Add peripherals missing from your current processor such as LCD or memory controllers, or increase the number of I/O channels in your system by adding Ethernet, general-purpose I/O (GPIO), or UART ports.
- Coprocessing-Boost system performance by moving compute-intensive algorithms from software running on a processor to hardware in the FPGA. Signal processing, image processing, and packet processing applications achieve orders of magnitude performance improvement running in hardware rather than software.
- Custom embedded controller-You decide which (and how many) processors, peripherals, interfaces, direct memory access (DMA) channels, and memories to include in your custom embedded controller.
- Multiprocessor-Accelerate your software development, improve code reliability, and increase maintainability by distributing tasks across several CPUs. You can design a multiprocessor system as a custom system inside a single FPGA or to augment an external CPU or digital signal processor.
FPGA developers enjoy several benefits not available to traditional embedded solutions:
- Protect your software investment from processor obsolescence-Because you own the hardware design for an FPGA-based embedded processor, your software investment is protected from processor obsolescence. In a worst case scenario, you can migrate your embedded design to a new FPGA family, requiring a board redesign. Your software investment, however, remains intact because the processor subsystem does not change.
- Reduce time to market-By adding an FPGA to your design you can release new products earlier, with modest feature sets, and then upgrade the hardware over time. Altera® provides an easy path to remotely update FPGA hardware designs over the Internet. In some cases, entire product lines can be based on a single board design; the variable content is contained within the FPGA.
- Adapt to changing requirements-FPGAs let you adapt to last-minute changes or evolving standards by adding or changing hardware features even after the PCB is complete.
- Increase system performance without board redesign-Sometimes you discover late in the design process that the system doesn’t meet performance. By including an FPGA in the system, you can add more performance without having to redesign the board, purchase a faster speed-grade device, or rewrite code in assembly. You can add multiple processors, custom instructions, and hardware accelerators to the FPGA to boost system performance without requiring a board redesign.
The Simulink*, Embedded Coder* and HDL Coder* tools from MathWorks* provide a hardware/software workflow spanning simulation, prototyping, verification, and implementation on Intel® SoC FPGAs. Go here for more information.
Subscribe to the Intel® FPGA Newsletter
Receive updates on Intel® FPGA products and technology, news, and upcoming events.
Infos sur le produit et ses performances
Cette comparaison oppose la famille de SoC et FPGA Intel® Agilex™ et les FPGA Intel® Stratix® 10 à l'aide des résultats de simulation et est susceptible d'être modifiée. Ce document contient des informations sur des produits, des services ou des processus en cours de développement. Toutes les informations susmentionnées sont modifiables à tout moment sans préavis. Contactez votre représentant Intel pour obtenir les toutes dernières informations concernant les prévisions, le calendrier, les caractéristiques et les feuilles de route.
Les fonctionnalités et avantages des technologies Intel® dépendent de la configuration du système et peuvent nécessiter du matériel et des logiciels compatibles, ou l'activation de services. Obtenez plus d'informations sur intel.fr, auprès du fabricant OEM ou du revendeur. Aucun ordinateur ne saurait être totalement sécurisé. Les logiciels et charges de travail utilisés dans les tests de performance ont peut-être été optimisés uniquement pour les microprocesseurs Intel®.
Intel, le logo Intel, Atom, Xeon et toute autre marque sont des marques commerciales d'Intel Corporation, aux États-Unis ou dans d'autres pays. *Les autres noms et marques peuvent être revendiqués comme la propriété de tiers.
© Intel Corporation 2019.
OpenCL et le logo OpenCL sont des marques de commerce d'Apple Inc. utilisées avec autorisation par Khronos.