Thing Done

← The fleet

Feather M4 vs PIC16F17146

Cortex-M4F (SAMD51) at 120 MHz against PIC16F (PIC16F17146) at 32 MHz, on real hardware, same C kernels, clocks verified.

Feather M4 Express

Feather M4

Cortex-M4F (SAMD51)
Family
Arm Cortex-M
Board
Feather M4 Express
Clock
120 MHz
float / double
32-bit / 64-bit
PIC16F17146 Curiosity Nano

PIC16F17146

PIC16F (PIC16F17146)
Family
PIC / dsPIC
Board
PIC16F17146 Curiosity Nano
Clock
32 MHz
float / double
32-bit / 32-bit
Typical gap
176×
Feather M4 faster, geometric mean of 25 cells
Biggest win for Feather M4
1,508×
float multiply
Closest cell
30.1×
uint8_t add
Clock for clock
47.1×
Feather M4 needs fewer cycles per op (3.75× clock gap)

Operation by operation

Every cell

How many times faster Feather M4 is than PIC16F17146, operations per second. Blue: Feather M4 ahead. Orange: PIC16F17146 ahead.

AddSubtractMultiplyDivideModulo
uint8_t 30.1× 8.98 M/s · 298 k/s 32.4× 8.98 M/s · 277 k/s 138× 8.98 M/s · 64.8 k/s 113× 7.01 M/s · 62 k/s 316× 6.36 M/s · 20.1 k/s
uint16_t 36.9× 8.98 M/s · 244 k/s 36.9× 8.98 M/s · 244 k/s 354× 8.98 M/s · 25.4 k/s 259× 6.12 M/s · 23.6 k/s 196× 5.36 M/s · 27.3 k/s
uint32_t 92.1× 9.71 M/s · 105 k/s 109× 9.71 M/s · 89 k/s 980× 9.71 M/s · 9.9 k/s 708× 6.53 M/s · 9.23 k/s 573× 6.19 M/s · 10.8 k/s
uint64_t not measurednot measurednot measurednot measurednot measured
float 361× 8.98 M/s · 24.9 k/s 482× 8.98 M/s · 18.6 k/s 1,508× 8.98 M/s · 5.95 k/s 858× 4.73 M/s · 5.51 k/s 118× 866 k/s · 7.33 k/s
double 48.1× 1.2 M/s · 24.9 k/s 64.0× 1.19 M/s · 18.6 k/s 286× 1.7 M/s · 5.95 k/s 229× 1.26 M/s · 5.51 k/s 58.1× 425 k/s · 7.33 k/s

Where they sit in the fleet

Keep comparing

Swap sides