XAZU5EV-1SFVC784Q
| Parameters | |
|---|---|
| Series | Zynq® UltraScale+™ MPSoC EV |
| Moisture Sensitivity Level (MSL) | 3 (168 Hours) |
| RoHS Status | ROHS3 Compliant |
| Package / Case | 784-BFBGA, FCBGA |
| Mfr | AMD |
| Operating Temperature | -40°C ~ 125°C (TJ) |
| Supplier Device Package | 784-FCBGA (23x23) |
| Speed | 500MHz, 600MHz, 1.2GHz |
| Core Processor | Quad ARM® Cortex®-A53 MPCore™ with CoreSight™, Dual ARM®Cortex™-R5 with CoreSight™, ARM Mali™-400 MP2 |
| RAM Size | 256KB |
| Number of I/O | 252 |
| Architecture | MCU, FPGA |
| Peripherals | DMA, WDT |
| Primary Attributes | Zynq®UltraScale+™ FPGA, 256K+ Logic Cells |
| Flash Size | - |
| Connectivity | CANbus, EBI/EMI, Ethernet, I²C, MMC/SD/SDIO, SPI, UART/USART, USB OTG |
| Package | Tray |
| Product Status | Active |
| Base Product Number | XAZU5 |
| Other Names | 122-XAZU5EV-1SFVC784Q |
| Standard Package | 1 |
XAZU5EV-1SFVC784Q Automotive-Grade Zynq UltraScale+ MPSoC
The XAZU5EV-1SFVC784Q is a high-performance automotive-grade System-on-Chip manufactured by AMD Xilinx, built on the Zynq UltraScale+ MPSoC architecture. This device fuses a rich processing system with massive programmable logic in a single package, making it one of the most capable SoC FPGAs for safety-critical automotive, aerospace, and industrial edge applications. The processing system integrates a quad-core 64-bit Arm Cortex-A53 application processing unit (APU) running up to 1.2GHz with 1MB shared L2 cache and 32KB L1 instruction plus 32KB L1 data caches per core, alongside a dual-core Arm Cortex-R5 real-time processing unit (RPU) clocked up to 500MHz featuring lockstep or independent operation modes for deterministic real-time control. The embedded Mali-400 MP2 GPU provides hardware-accelerated graphics and compute for instrument cluster and infotainment displays.
The programmable logic fabric delivers 256,200 logic elements organized into 14,640 adaptive logic modules (ALMs), supporting complex custom datapath designs for sensor fusion, protocol bridging, and real-time signal processing. On-chip memory resources include 256KB of OCM with ECC, up to 27Mb of UltraRAM with ECC in the PL, 21.1Mb of block RAM, and 3.5Mb of distributed RAM, providing ample storage for buffering high-bandwidth video streams and neural network inference workloads without relying on external DRAM. The device offers 252 user I/O pins supporting LVDS, LVCMOS, and HSTL standards, plus 8 multi-gigabit transceivers capable of 12.5 Gb/s for PCIe Gen3, Gigabit Ethernet, and JESD204B connectivity.
Alternative FPGA SoC ICs
● XCZU4EV-2SFVC784I (AMD Xilinx): Lower-density 192K LE sibling in the same 784-pin package for cost-sensitive automotive designs.
● XCZU9EG-2FFVB1156E (AMD Xilinx): 1.2M LE Monster device with 36 DSP slices and HBM2 memory for data center AI inference.
● AGFB014R24A2E3V (Intel Agilex): 14nm automotive FPGA with Arm Cortex-A53 and RISC-V cores for cross-vendor safety-critical migration.
● MPFS250T-1FCG484I (Microchip SmartFusion2): 250K LE FPGA with hardened ARM Cortex-M3 for industrial motor control and IoT gateway applications.
● LFE5U-85F-8BG756C (Lattice CertusPro-NX): 85K LE low-power FPGA with MIPI and PCIe support for automotive camera and sensor bridging.
