GW1NRF-4B vs XC7S6
A comparison of the Gowin LittleBee GW1NRF-4B and the AMD Spartan-7 XC7S6 using the data from both vendors' selection guides.
Side-by-side specifications
| GW1NRF-4B | XC7S6 | |
|---|---|---|
| Vendor / family | Gowin LittleBee | AMD Spartan-7 |
| Logic (LUT4-equivalent) | 4,606 | 6,003 |
| LUTs as published | 4,606 LUT4 | 3,752 LUT6 |
| Vendor logic figure | — LC | 6,000 LC |
| DSP | 16 18×18 multipliers | 10 DSP slices |
| Block RAM | 0.2 Mb | 0.2 Mb |
| Large RAM | — | — |
| Max. user I/O | 25 | 100 |
| Transceivers | None | None |
| Hard PCI Express | None | None |
| Hard processor | 1× ARC (32-bit) | None |
| Process | 55 nm | 28 nm |
| Configuration memory | Embedded flash | SRAM |
| Packages | 0 | 3 |
| Design tool | Gowin EDA | AMD Vivado |
Highlighted values are higher. Logic is compared in LUT4 equivalents; see how FPGA vendors count logic.
Key differences
The XC7S6 has about 30% more logic capacity (6K vs 4.6K LUT4 equivalents).
They use different logic architectures: the GW1NRF-4B is built from LUT4 and the XC7S6 from LUT6, so the raw LUT counts (4.6K vs 3.8K) are not directly comparable.
For arithmetic, the GW1NRF-4B has 16 18×18 multipliers against 10 DSP slices on the XC7S6.
The XC7S6 has more on-chip memory (0.2 Mb vs 0.2 Mb).
The GW1NRF-4B includes a hard ARC (32-bit) processor, while the XC7S6 does not.
Configuration memory: embedded flash on the GW1NRF-4B, sram on the XC7S6.
The GW1NRF-4B is designed with Gowin EDA and the XC7S6 with AMD Vivado.
Full specifications
See the complete spec sheets with packages and grades: GW1NRF-4B (LittleBee) and XC7S6 (Spartan-7).