Feather M4 vs Teensy 4.0
Cortex-M4F (SAMD51) at 120 MHz against Cortex-M7 (i.MX RT1062) at 600 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
Teensy 4.0
Cortex-M7 (i.MX RT1062)
- Family
- Arm Cortex-M
- Board
- Teensy 4.0
- Clock
- 600 MHz
- float / double
- 32-bit / 64-bit
Typical gap
9.28×
Teensy 4.0 faster, geometric mean of 30 cells
Biggest win for Feather M4
0.18×
uint32_t modulo
Biggest win for Teensy 4.0
46.8×
double add
Clock for clock
1.86×
Teensy 4.0 needs fewer cycles per op (5.00× clock gap)
Operation by operation
Every cell
How many times faster Feather M4 is than Teensy 4.0, operations per second. Blue: Feather M4 ahead. Orange: Teensy 4.0 ahead.
| Add | Subtract | Multiply | Divide | Modulo | |
|---|---|---|---|---|---|
| uint8_t | 1/8.62× 8.98 M/s · 77.4 M/s | 1/8.38× 8.98 M/s · 75.3 M/s | 1/8.38× 8.98 M/s · 75.3 M/s | 1/7.46× 7.01 M/s · 52.3 M/s | 1/7.28× 6.36 M/s · 46.3 M/s |
| uint16_t | 1/8.38× 8.98 M/s · 75.3 M/s | 1/8.35× 8.98 M/s · 75 M/s | 1/7.64× 8.98 M/s · 68.6 M/s | 1/6.78× 6.12 M/s · 41.5 M/s | 1/6.99× 5.36 M/s · 37.5 M/s |
| uint32_t | 1/8.52× 9.71 M/s · 82.8 M/s | 1/8.83× 9.71 M/s · 85.7 M/s | 1/7.06× 9.71 M/s · 68.6 M/s | 1/6.02× 6.53 M/s · 39.3 M/s | 1/5.62× 6.19 M/s · 34.8 M/s |
| uint64_t | 1/10.2× 6.19 M/s · 63.2 M/s | 1/10.2× 6.19 M/s · 63.2 M/s | 1/7.79× 5.61 M/s · 43.7 M/s | 1/6.26× 885 k/s · 5.54 M/s | 1/6.26× 885 k/s · 5.54 M/s |
| float | 1/8.66× 8.98 M/s · 77.7 M/s | 1/8.66× 8.98 M/s · 77.7 M/s | 1/8.66× 8.98 M/s · 77.7 M/s | 1/5.71× 4.73 M/s · 27 M/s | 1/7.70× 866 k/s · 6.67 M/s |
| double | 1/46.8× 1.2 M/s · 56 M/s | 1/45.9× 1.19 M/s · 54.7 M/s | 1/25.7× 1.7 M/s · 43.7 M/s | 1/12.5× 1.26 M/s · 15.7 M/s | 1/11.4× 425 k/s · 4.84 M/s |