Teensy 4.0 vs Pi Pico 2W (RISC-V)
Cortex-M7 (i.MX RT1062) at 600 MHz against Hazard3 RV32IMAC (RP2350) at 150 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 2W (RISC-V)
Hazard3 RV32IMAC (RP2350)
- Family
- RISC-V
- Board
- Pi Pico 2W (RISC-V)
- Clock
- 150 MHz
- float / double
- 32-bit / 64-bit
Typical gap
11.0×
Teensy 4.0 faster, geometric mean of 30 cells
Biggest win for Teensy 4.0
64.5×
double multiply
Closest cell
5.65×
uint8_t multiply
Clock for clock
2.75×
Teensy 4.0 needs fewer cycles per op (4.00× clock gap)
Operation by operation
Every cell
How many times faster Teensy 4.0 is than Pi Pico 2W (RISC-V), operations per second. Blue: Teensy 4.0 ahead. Orange: Pi Pico 2W (RISC-V) ahead.
| Add | Subtract | Multiply | Divide | Modulo | |
|---|---|---|---|---|---|
| uint8_t | 6.33× 77.4 M/s · 12.2 M/s | 5.65× 75.3 M/s · 13.3 M/s | 5.65× 75.3 M/s · 13.3 M/s | 9.88× 52.3 M/s · 5.3 M/s | 8.74× 46.3 M/s · 5.3 M/s |
| uint16_t | 6.66× 75.3 M/s · 11.3 M/s | 6.63× 75 M/s · 11.3 M/s | 6.52× 68.6 M/s · 10.5 M/s | 8.39× 41.5 M/s · 4.94 M/s | 7.83× 37.5 M/s · 4.79 M/s |
| uint32_t | 6.77× 82.8 M/s · 12.2 M/s | 7.58× 85.7 M/s · 11.3 M/s | 6.07× 68.6 M/s · 11.3 M/s | 7.96× 39.3 M/s · 4.94 M/s | 7.05× 34.8 M/s · 4.94 M/s |
| uint64_t | 7.70× 63.2 M/s · 8.2 M/s | 7.70× 63.2 M/s · 8.2 M/s | 5.92× 43.7 M/s · 7.39 M/s | 6.74× 5.54 M/s · 822 k/s | 6.98× 5.54 M/s · 794 k/s |
| float | 32.0× 77.7 M/s · 2.43 M/s | 32.0× 77.7 M/s · 2.43 M/s | 26.1× 77.7 M/s · 2.98 M/s | 35.6× 27 M/s · 758 k/s | 5.83× 6.67 M/s · 1.14 M/s |
| double | 46.2× 56 M/s · 1.21 M/s | 44.0× 54.7 M/s · 1.24 M/s | 64.5× 43.7 M/s · 678 k/s | 39.8× 15.7 M/s · 394 k/s | 6.62× 4.84 M/s · 731 k/s |