Thing Done

← The fleet

Feather M4 vs PIC18F57Q84

Cortex-M4F (SAMD51) at 120 MHz against PIC18F (PIC18F57Q84) at 64 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
PIC18F57Q84 Curiosity Nano

PIC18F57Q84

PIC18F (PIC18F57Q84)
Family
PIC / dsPIC
Board
PIC18F57Q84 Curiosity Nano
Clock
64 MHz
float / double
32-bit / 32-bit
Typical gap
70.2×
Feather M4 faster, geometric mean of 25 cells
Biggest win for Feather M4
464×
float divide
Closest cell
16.7×
uint8_t add
Clock for clock
37.5×
Feather M4 needs fewer cycles per op (1.88× clock gap)

Operation by operation

Every cell

How many times faster Feather M4 is than PIC18F57Q84, operations per second. Blue: Feather M4 ahead. Orange: PIC18F57Q84 ahead.

AddSubtractMultiplyDivideModulo
uint8_t 16.7× 8.98 M/s · 538 k/s 17.8× 8.98 M/s · 504 k/s 17.3× 8.98 M/s · 520 k/s 51.8× 7.01 M/s · 135 k/s 59.9× 6.36 M/s · 106 k/s
uint16_t 27.4× 8.98 M/s · 328 k/s 27.4× 8.98 M/s · 328 k/s 31.3× 8.98 M/s · 287 k/s 98.4× 6.12 M/s · 62.1 k/s 76.0× 5.36 M/s · 70.6 k/s
uint32_t 40.5× 9.71 M/s · 240 k/s 40.5× 9.71 M/s · 240 k/s 414× 9.71 M/s · 23.5 k/s 267× 6.53 M/s · 24.5 k/s 216× 6.19 M/s · 28.6 k/s
uint64_t not measurednot measurednot measurednot measurednot measured
float 240× 8.98 M/s · 37.4 k/s 283× 8.98 M/s · 31.7 k/s 398× 8.98 M/s · 22.6 k/s 464× 4.73 M/s · 10.2 k/s 43.8× 866 k/s · 19.8 k/s
double 32.0× 1.2 M/s · 37.4 k/s 37.6× 1.19 M/s · 31.7 k/s 75.5× 1.7 M/s · 22.6 k/s 123× 1.26 M/s · 10.2 k/s 21.5× 425 k/s · 19.8 k/s

Where they sit in the fleet

Keep comparing

Swap sides