Arduino Uno vs Teensy 4.0
ATmega328P at 16 MHz against Cortex-M7 (i.MX RT1062) at 600 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
Typical gap
388×
Teensy 4.0 faster, geometric mean of 30 cells
Biggest win for Arduino Uno
0.02×
double modulo
Biggest win for Teensy 4.0
1,586×
uint32_t divide
Clock for clock
10.4×
Teensy 4.0 needs fewer cycles per op (37.5× clock gap)
Operation by operation
Every cell
How many times faster Arduino Uno is than Teensy 4.0, operations per second. Blue: Arduino Uno ahead. Orange: Teensy 4.0 ahead.
| Add | Subtract | Multiply | Divide | Modulo | |
|---|---|---|---|---|---|
| uint8_t | 1/158× 490 k/s · 77.4 M/s | 1/154× 490 k/s · 75.3 M/s | 1/168× 448 k/s · 75.3 M/s | 1/368× 142 k/s · 52.3 M/s | 1/325× 142 k/s · 46.3 M/s |
| uint16_t | 1/144× 522 k/s · 75.3 M/s | 1/144× 522 k/s · 75 M/s | 1/166× 413 k/s · 68.6 M/s | 1/597× 69.5 k/s · 41.5 M/s | 1/540× 69.5 k/s · 37.5 M/s |
| uint32_t | 1/242× 342 k/s · 82.8 M/s | 1/251× 342 k/s · 85.7 M/s | 1/508× 135 k/s · 68.6 M/s | 1/1,586× 24.8 k/s · 39.3 M/s | 1/1,405× 24.8 k/s · 34.8 M/s |
| uint64_t | 1/847× 74.6 k/s · 63.2 M/s | 1/847× 74.6 k/s · 63.2 M/s | 1/1,218× 35.9 k/s · 43.7 M/s | 1/363× 15.3 k/s · 5.54 M/s | 1/365× 15.2 k/s · 5.54 M/s |
| float | 1/791× 98.3 k/s · 77.7 M/s | 1/787× 98.7 k/s · 77.7 M/s | 1/894× 87 k/s · 77.7 M/s | 1/838× 32.2 k/s · 27 M/s | 1/64.9× 103 k/s · 6.67 M/s |
| double | 1/571× 98.1 k/s · 56 M/s | 1/555× 98.6 k/s · 54.7 M/s | 1/504× 86.8 k/s · 43.7 M/s | 1/487× 32.2 k/s · 15.7 M/s | 1/47.1× 103 k/s · 4.84 M/s |