XCVC1902 vs XCVM2902
A comparison of the AMD Versal AI Core XCVC1902 and the AMD Versal Prime XCVM2902 using the data from both vendors' selection guides.
Side-by-side specifications
| XCVC1902 | XCVM2902 | |
|---|---|---|
| Vendor / family | AMD Versal AI Core | AMD Versal Prime |
| Logic (LUT4-equivalent) | 1,439,744 | 1,633,485 |
| LUTs as published | 899,840 LUT6 | 1,020,928 LUT6 |
| Vendor logic figure | 1,968,000 SLC | 2,233,000 SLC |
| DSP | 1,968 DSP Engines | 2,672 DSP Engines |
| Block RAM | 34 Mb | 70 Mb |
| Large RAM | 130 Mb | 181 Mb |
| Max. user I/O | 692 | 346 |
| Transceivers | 44 × GTY 32.75 Gb/s | 8 × GTYP 32.75 Gb/s, 36 × GTM 58 Gb/s |
| Hard PCI Express | Gen4 x16 | Gen5 x4 |
| Hard processor | 2× Arm Cortex-A72 | 2× Arm Cortex-A72 |
| Process | 7 nm | 7 nm |
| Configuration memory | SRAM | SRAM |
| Packages | 3 | 1 |
| Design tool | AMD Vivado | AMD Vivado |
Highlighted values are higher. Logic is compared in LUT4 equivalents; see how FPGA vendors count logic.
Key differences
The XCVM2902 has about 13% more logic capacity (1633K vs 1440K LUT4 equivalents).
For arithmetic, the XCVM2902 has 2,672 DSP Engines against 1,968 DSP Engines on the XCVC1902.
The XCVM2902 has more on-chip memory (251 Mb vs 164 Mb).
Both have transceivers: 44 at up to 32.75 Gb/s on the XCVC1902 and 44 at up to 58 Gb/s on the XCVM2902.
The XCVC1902 is designed with AMD Vivado and the XCVM2902 with AMD Vivado.
Full specifications
See the complete spec sheets with packages and grades: XCVC1902 (Versal AI Core) and XCVM2902 (Versal Prime).