Teensy 4.0 vs Pi Pico 2W
Cortex-M7 (i.MX RT1062) at 600 MHz against Cortex-M33 (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
Cortex-M33 (RP2350)
- Family
- Arm Cortex-M
- Board
- Pi Pico 2W
- Clock
- 150 MHz
- float / double
- 32-bit / 64-bit
Typical gap
6.34×
Teensy 4.0 faster, geometric mean of 30 cells
Biggest win for Teensy 4.0
16.3×
uint64_t divide
Closest cell
3.84×
float modulo
Clock for clock
1.58×
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, operations per second. Blue: Teensy 4.0 ahead. Orange: Pi Pico 2W ahead.
| Add | Subtract | Multiply | Divide | Modulo | |
|---|---|---|---|---|---|
| uint8_t | 6.34× 77.4 M/s · 12.2 M/s | 6.17× 75.3 M/s · 12.2 M/s | 6.67× 75.3 M/s · 11.3 M/s | 5.51× 52.3 M/s · 9.49 M/s | 5.80× 46.3 M/s · 7.97 M/s |
| uint16_t | 6.14× 75.3 M/s · 12.3 M/s | 6.15× 75 M/s · 12.2 M/s | 5.59× 68.6 M/s · 12.3 M/s | 4.77× 41.5 M/s · 8.7 M/s | 5.06× 37.5 M/s · 7.4 M/s |
| uint32_t | 6.23× 82.8 M/s · 13.3 M/s | 6.42× 85.7 M/s · 13.4 M/s | 5.59× 68.6 M/s · 12.3 M/s | 4.28× 39.3 M/s · 9.2 M/s | 4.25× 34.8 M/s · 8.19 M/s |
| uint64_t | 6.42× 63.2 M/s · 9.84 M/s | 6.42× 63.2 M/s · 9.84 M/s | 5.34× 43.7 M/s · 8.19 M/s | 16.3× 5.54 M/s · 340 k/s | 16.3× 5.54 M/s · 340 k/s |
| float | 5.82× 77.7 M/s · 13.4 M/s | 5.82× 77.7 M/s · 13.4 M/s | 5.82× 77.7 M/s · 13.4 M/s | 4.20× 27 M/s · 6.43 M/s | 3.84× 6.67 M/s · 1.74 M/s |
| double | 9.81× 56 M/s · 5.71 M/s | 9.61× 54.7 M/s · 5.69 M/s | 12.6× 43.7 M/s · 3.46 M/s | 7.47× 15.7 M/s · 2.1 M/s | 3.97× 4.84 M/s · 1.22 M/s |