Nucleo G474RE vs Teensy 4.0
Cortex-M4F (STM32G474) at 170 MHz against Cortex-M7 (i.MX RT1062) at 600 MHz, on real hardware, same C kernels, clocks verified.
Nucleo G474RE
Cortex-M4F (STM32G474)
- Family
- Arm Cortex-M
- Board
- Nucleo G474RE
- Clock
- 170 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
7.03×
Teensy 4.0 faster, geometric mean of 30 cells
Biggest win for Nucleo G474RE
0.24×
uint32_t modulo
Biggest win for Teensy 4.0
35.5×
double add
Clock for clock
1.99×
Teensy 4.0 needs fewer cycles per op (3.53× clock gap)
Operation by operation
Every cell
How many times faster Nucleo G474RE is than Teensy 4.0, operations per second. Blue: Nucleo G474RE ahead. Orange: Teensy 4.0 ahead.
| Add | Subtract | Multiply | Divide | Modulo | |
|---|---|---|---|---|---|
| uint8_t | 1/6.59× 11.7 M/s · 77.4 M/s | 1/6.41× 11.7 M/s · 75.3 M/s | 1/6.41× 11.7 M/s · 75.3 M/s | 1/5.84× 8.96 M/s · 52.3 M/s | 1/5.71× 8.1 M/s · 46.3 M/s |
| uint16_t | 1/6.41× 11.7 M/s · 75.3 M/s | 1/6.39× 11.7 M/s · 75 M/s | 1/5.84× 11.7 M/s · 68.6 M/s | 1/5.06× 8.2 M/s · 41.5 M/s | 1/4.96× 7.56 M/s · 37.5 M/s |
| uint32_t | 1/6.56× 12.6 M/s · 82.8 M/s | 1/6.79× 12.6 M/s · 85.7 M/s | 1/5.43× 12.6 M/s · 68.6 M/s | 1/4.51× 8.73 M/s · 39.3 M/s | 1/4.20× 8.3 M/s · 34.8 M/s |
| uint64_t | 1/8.35× 7.56 M/s · 63.2 M/s | 1/8.35× 7.56 M/s · 63.2 M/s | 1/6.56× 6.67 M/s · 43.7 M/s | 1/4.78× 1.16 M/s · 5.54 M/s | 1/4.78× 1.16 M/s · 5.54 M/s |
| float | 1/6.62× 11.7 M/s · 77.7 M/s | 1/6.62× 11.7 M/s · 77.7 M/s | 1/6.62× 11.7 M/s · 77.7 M/s | 1/4.21× 6.42 M/s · 27 M/s | 1/4.98× 1.34 M/s · 6.67 M/s |
| double | 1/35.5× 1.58 M/s · 56 M/s | 1/34.7× 1.58 M/s · 54.7 M/s | 1/19.7× 2.22 M/s · 43.7 M/s | 1/9.39× 1.67 M/s · 15.7 M/s | 1/7.04× 688 k/s · 4.84 M/s |