LAV-AT-G70 vs XCKU9P
A comparison of the Lattice Avant-G LAV-AT-G70 and the AMD Kintex UltraScale+ XCKU9P using the data from both vendors' selection guides.
Side-by-side specifications
| LAV-AT-G70 | XCKU9P | |
|---|---|---|
| Vendor / family | Lattice Avant-G | AMD Kintex UltraScale+ |
| Logic (LUT4-equivalent) | 397,000 | 438,400 |
| LUTs as published | 397,000 LUT4 | 274,000 LUT6 |
| Vendor logic figure | 637,000 LC | 600,000 SLC |
| DSP | 1,800 18×18 multipliers | 2,520 DSP slices |
| Block RAM | 35.6 Mb | 32.1 Mb |
| Large RAM | — | — |
| Max. user I/O | 554 | 304 |
| Transceivers | 28 × SERDES 12.5 Gb/s | 28 × GTH 16.3 Gb/s |
| Hard PCI Express | None | None |
| Hard processor | None | None |
| Process | 16 nm | 16 nm |
| Configuration memory | SRAM | SRAM |
| Packages | 3 | 1 |
| Design tool | Lattice Radiant | AMD Vivado |
Highlighted values are higher. Logic is compared in LUT4 equivalents; see how FPGA vendors count logic.
Key differences
The XCKU9P has about 10% more logic capacity (438K vs 397K LUT4 equivalents).
They use different logic architectures: the LAV-AT-G70 is built from LUT4 and the XCKU9P from LUT6, so the raw LUT counts (397K vs 274K) are not directly comparable.
For arithmetic, the XCKU9P has 2,520 DSP slices against 1,800 18×18 multipliers on the LAV-AT-G70.
Both have transceivers: 28 at up to 12.5 Gb/s on the LAV-AT-G70 and 28 at up to 16.3 Gb/s on the XCKU9P.
The LAV-AT-G70 is designed with Lattice Radiant and the XCKU9P with AMD Vivado.
Full specifications
See the complete spec sheets with packages and grades: LAV-AT-G70 (Avant-G) and XCKU9P (Kintex UltraScale+).