AMD / Xilinx Artix™ UltraScale+™ FGPAs
AMD/Xilinx Artix™ UltraScale+™ FGPAs are cost-optimized FPGAs based on an advanced, production-proven 16nm architecture for exceptional performance/watt. These devices are enhanced with packaging innovation for ultra-compact form factor and compute density.
With up to 16Gb/s transceivers for advanced protocols and the highest FPGA DSP performance against competitive FPGA products, AMD/Xilinx Artix UltraScale+ FPGAs match I/O bandwidth to compute to help maximize system performance for cost-sensitive and low-power applications in machine vision, networking, 4K broadcast, and a range of Industrial IoT and edge markets.
Features
- Enhanced programmable logic architecture
- Based on TSMC's 16nm FinFET+ process
- 2.4X performance/watt vs. Artix 7 FPGAs
- Voltage scaling to tune power and performance on the same device
- Enhanced CLB/LUTs, routing, and ASIC-class clocking for high utilization
- High-performance transceivers
- Up to 16Gbps transceiver line rates (minimum of 12Gbps across the family)
- Power-optimized architecture vs. Artix 7 FPGAs
- Single oscillator for fabric and SerDes eliminates extra clocking components
- PCI Express® Gen3 and Gen4 support
- PCI Express Gen3 x8, Gen4 x8 compatible
- DMA IP for complete end-to-end solution
- Exceptional DSP compute in its class
- Highest bandwidth in a cost-optimized FPGA
- 1,860 GOP/s, 620 GFLOPs (FP32) in the largest device
- Up to 50% fewer resources for equivalent computation vs. Artix 7 FPGAs
- Safety and multi-level security features
- RSA-4096 authentication to verify design source
- NIST AES-CGM decryption approved for faster configuration
- Permanent tamper penalty to prevent adversaries from accessing security features
- Security monitoring IP to adapt to security threats across the product life cycle
- DDR4-2400 performance
- DDR4-2400 for highest memory interface performance in a cost-optimized FPGA
- Memory bandwidth to match on-chip compute
- Reduced memory controller fabric utilization and power vs. Artix 7 FPGAs
- Outstanding MIPI and LVDS performance
- Up to 2500Mbps MIPI and LVDS performance
- Support for advanced vision sensors (MIPI, SLVS-EC)
- Analog mixed-signal monitoring block
- Voltage, current, and temperature tracking for safe, secure, and reliable operation
- Helps meet requirements for key standards (FIPS 140-2, IEC 61508, and ISO26262)
- Allows for integration of low-amplitude sensors
Applications
- Industrial
- Machine vision
- Industrial networking (time-sensitive networking)
- Industrial controllers
- Medical
- Portable and desktop ultra-sound
- Surgical vision
- Endoscopy
- Networking
- Cost-sensitive Nx10 G and 25 G networking
- Network bridging for Nx100 G systems
- AV Broadcasting
- LED video walls
- Digital signage
- KVM switch
- Video mini-converters
- Aerospace and defense
- MILCOM radio
- Missiles and munitions
Videos
Related FPGAs
AMD / Xilinx Spartan-6 LXT FPGAs
Extends LX devices adding up to eight 3.2Gb/s GTP transceivers and an integrated block.
AMD / Xilinx Spartan™ 7 FPGAs
Feature a MicroBlaze™ soft processor running over 200DMIPs with 800Mb/s DDR3 support.
AMD / Xilinx Artix™ 7 FPGAs
Ideal for cost-sensitive applications that need high-end features.
Design Software
AMD / Xilinx Vivado™ Design Suite
Software used for adaptive SoCs and FPGAs.