Feather M4 vs XIAO MG24
Cortex-M4F (SAMD51) at 120 MHz against Cortex-M33 (EFR32MG24) at 39 MHz, on real hardware, same C kernels, clocks verified.
Feather M4
Cortex-M4F (SAMD51)
- Family
- Arm Cortex-M
- Board
- Feather M4 Express
- Clock
- 120 MHz
- float / double
- 32-bit / 64-bit
XIAO MG24
Cortex-M33 (EFR32MG24)
- Family
- Arm Cortex-M
- Board
- XIAO MG24 Sense
- Clock
- 39 MHz
- float / double
- 32-bit / 64-bit
Typical gap
3.04×
Feather M4 faster, geometric mean of 30 cells
Biggest win for Feather M4
3.56×
uint8_t multiply
Closest cell
2.58×
double modulo
Clock for clock
1.01×
XIAO MG24 needs fewer cycles per op (3.08× clock gap)
Operation by operation
Every cell
How many times faster Feather M4 is than XIAO MG24, operations per second. Blue: Feather M4 ahead. Orange: XIAO MG24 ahead.
| Add | Subtract | Multiply | Divide | Modulo | |
|---|---|---|---|---|---|
| uint8_t | 3.31× 8.98 M/s · 2.71 M/s | 3.31× 8.98 M/s · 2.71 M/s | 3.56× 8.98 M/s · 2.52 M/s | 2.87× 7.01 M/s · 2.44 M/s | 3.32× 6.36 M/s · 1.92 M/s |
| uint16_t | 3.05× 8.98 M/s · 2.94 M/s | 3.07× 8.98 M/s · 2.93 M/s | 3.07× 8.98 M/s · 2.93 M/s | 2.93× 6.12 M/s · 2.09 M/s | 3.03× 5.36 M/s · 1.77 M/s |
| uint32_t | 3.04× 9.71 M/s · 3.19 M/s | 3.04× 9.71 M/s · 3.19 M/s | 3.31× 9.71 M/s · 2.93 M/s | 2.96× 6.53 M/s · 2.21 M/s | 3.15× 6.19 M/s · 1.96 M/s |
| uint64_t | 2.62× 6.19 M/s · 2.36 M/s | 2.62× 6.19 M/s · 2.36 M/s | 2.85× 5.61 M/s · 1.97 M/s | 3.11× 885 k/s · 285 k/s | 3.11× 885 k/s · 285 k/s |
| float | 2.83× 8.98 M/s · 3.18 M/s | 2.83× 8.98 M/s · 3.18 M/s | 2.82× 8.98 M/s · 3.18 M/s | 3.06× 4.73 M/s · 1.54 M/s | 2.72× 866 k/s · 318 k/s |
| double | 3.24× 1.2 M/s · 370 k/s | 3.23× 1.19 M/s · 368 k/s | 3.38× 1.7 M/s · 503 k/s | 3.33× 1.26 M/s · 379 k/s | 2.58× 425 k/s · 165 k/s |