Teensy 4.0 vs ESP32-C6
Cortex-M7 (i.MX RT1062) at 600 MHz against RV32IMAC (ESP32-C6) at 160 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
ESP32-C6
RV32IMAC (ESP32-C6)
- Family
- RISC-V
- Board
- ESP32-C6-DevKitM-1
- Clock
- 160 MHz
- float / double
- 32-bit / 64-bit
Typical gap
9.59×
Teensy 4.0 faster, geometric mean of 30 cells
Biggest win for Teensy 4.0
40.4×
double subtract
Closest cell
4.69×
uint32_t modulo
Clock for clock
2.56×
Teensy 4.0 needs fewer cycles per op (3.75× clock gap)
Operation by operation
Every cell
How many times faster Teensy 4.0 is than ESP32-C6, operations per second. Blue: Teensy 4.0 ahead. Orange: ESP32-C6 ahead.
| Add | Subtract | Multiply | Divide | Modulo | |
|---|---|---|---|---|---|
| uint8_t | 6.03× 77.4 M/s · 12.8 M/s | 5.39× 75.3 M/s · 14 M/s | 5.39× 75.3 M/s · 14 M/s | 6.46× 52.3 M/s · 8.1 M/s | 5.64× 46.3 M/s · 8.2 M/s |
| uint16_t | 7.29× 75.3 M/s · 10.3 M/s | 7.26× 75 M/s · 10.3 M/s | 6.20× 68.6 M/s · 11.1 M/s | 5.64× 41.5 M/s · 7.35 M/s | 5.05× 37.5 M/s · 7.43 M/s |
| uint32_t | 6.45× 82.8 M/s · 12.8 M/s | 7.22× 85.7 M/s · 11.9 M/s | 5.77× 68.6 M/s · 11.9 M/s | 5.36× 39.3 M/s · 7.35 M/s | 4.69× 34.8 M/s · 7.43 M/s |
| uint64_t | 7.70× 63.2 M/s · 8.2 M/s | 7.30× 63.2 M/s · 8.65 M/s | 6.16× 43.7 M/s · 7.1 M/s | 5.48× 5.54 M/s · 1.01 M/s | 5.62× 5.54 M/s · 986 k/s |
| float | 32.4× 77.7 M/s · 2.4 M/s | 33.9× 77.7 M/s · 2.29 M/s | 28.8× 77.7 M/s · 2.7 M/s | 29.9× 27 M/s · 905 k/s | 6.95× 6.67 M/s · 959 k/s |
| double | 40.1× 56 M/s · 1.39 M/s | 40.4× 54.7 M/s · 1.35 M/s | 25.7× 43.7 M/s · 1.7 M/s | 31.9× 15.7 M/s · 492 k/s | 7.49× 4.84 M/s · 647 k/s |