FPGA & adaptive SoC databook · online editionRev. 2026.10

GW5AS-138 vs XCZU2EG

A comparison of the Gowin Arora V GW5AS-138 and the AMD Zynq UltraScale+ MPSoC XCZU2EG using the data from both vendors' selection guides.

Side-by-side specifications

GW5AS-138XCZU2EG
Vendor / familyGowin Arora VAMD Zynq UltraScale+ MPSoC
Logic (LUT4-equivalent)138,24075,200
LUTs as published138,240 LUT447,000 LUT6
Vendor logic figure— LC103,000 SLC
DSP298 27×18 DSP blocks240 DSP slices
Block RAM6 Mb5.3 Mb
Large RAM——
Max. user I/O312252
TransceiversNoneNone
Hard PCI ExpressNoneNone
Hard processor1× RISC-V AE3504× Arm Cortex-A53
Process22 nm16 nm
Configuration memorySRAMSRAM
Packages14
Design toolGowin EDAAMD Vivado

Highlighted values are higher. Logic is compared in LUT4 equivalents; see how FPGA vendors count logic.

Key differences

The GW5AS-138 has about 84% more logic capacity (138K vs 75.2K LUT4 equivalents).

They use different logic architectures: the GW5AS-138 is built from LUT4 and the XCZU2EG from LUT6, so the raw LUT counts (138K vs 47K) are not directly comparable.

For arithmetic, the GW5AS-138 has 298 27×18 DSP blocks against 240 DSP slices on the XCZU2EG.

The GW5AS-138 has more on-chip memory (6 Mb vs 5.3 Mb).

The GW5AS-138 is designed with Gowin EDA and the XCZU2EG with AMD Vivado.

Full specifications

See the complete spec sheets with packages and grades: GW5AS-138 (Arora V) and XCZU2EG (Zynq UltraScale+ MPSoC).