Teensy 4.0 vs Pi Pico
Cortex-M7 (i.MX RT1062) at 600 MHz against Cortex-M0+ (RP2040) at 200 MHz, on real hardware, same C kernels, clocks verified.
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
Pi Pico
Cortex-M0+ (RP2040)
- Family
- Arm Cortex-M
- Board
- Pi Pico
- Clock
- 200 MHz
- float / double
- 32-bit / 64-bit
Typical gap
10.1×
Teensy 4.0 faster, geometric mean of 30 cells
Biggest win for Teensy 4.0
43.0×
double multiply
Closest cell
2.76×
uint64_t modulo
Clock for clock
3.38×
Teensy 4.0 needs fewer cycles per op (3.00× clock gap)
Operation by operation
Every cell
How many times faster Teensy 4.0 is than Pi Pico, operations per second. Blue: Teensy 4.0 ahead. Orange: Pi Pico ahead.
| Add | Subtract | Multiply | Divide | Modulo | |
|---|---|---|---|---|---|
| uint8_t | 5.71× 77.4 M/s · 13.5 M/s | 5.18× 75.3 M/s · 14.5 M/s | 5.94× 75.3 M/s · 12.7 M/s | 10.9× 52.3 M/s · 4.78 M/s | 9.68× 46.3 M/s · 4.78 M/s |
| uint16_t | 7.07× 75.3 M/s · 10.7 M/s | 7.04× 75 M/s · 10.7 M/s | 6.09× 68.6 M/s · 11.3 M/s | 8.69× 41.5 M/s · 4.77 M/s | 7.86× 37.5 M/s · 4.77 M/s |
| uint32_t | 7.35× 82.8 M/s · 11.3 M/s | 7.62× 85.7 M/s · 11.3 M/s | 5.75× 68.6 M/s · 11.9 M/s | 7.65× 39.3 M/s · 5.14 M/s | 6.78× 34.8 M/s · 5.14 M/s |
| uint64_t | 9.81× 63.2 M/s · 6.44 M/s | 10.4× 63.2 M/s · 6.05 M/s | 14.0× 43.7 M/s · 3.13 M/s | 2.76× 5.54 M/s · 2.01 M/s | 2.76× 5.54 M/s · 2.01 M/s |
| float | 36.0× 77.7 M/s · 2.16 M/s | 38.8× 77.7 M/s · 2 M/s | 31.7× 77.7 M/s · 2.45 M/s | 13.4× 27 M/s · 2.01 M/s | 5.85× 6.67 M/s · 1.14 M/s |
| double | 35.0× 56 M/s · 1.6 M/s | 36.9× 54.7 M/s · 1.48 M/s | 43.0× 43.7 M/s · 1.02 M/s | 16.9× 15.7 M/s · 926 k/s | 6.60× 4.84 M/s · 734 k/s |