Thing Done

← The fleet

Feather M4 vs PIC24FJ64GU205

Cortex-M4F (SAMD51) at 120 MHz against PIC24F (PIC24FJ64GU205) 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
PIC24FJ64GU205 Curiosity Nano

PIC24FJ64GU205

PIC24F (PIC24FJ64GU205)
Family
PIC / dsPIC
Board
PIC24FJ64GU205 Curiosity Nano
Clock
32 MHz
float / double
32-bit / 32-bit
Typical gap
20.2×
Feather M4 faster, geometric mean of 30 cells
Biggest win for Feather M4
405×
uint64_t modulo
Closest cell
4.50×
double modulo
Clock for clock
5.39×
Feather M4 needs fewer cycles per op (3.75× clock gap)

Operation by operation

Every cell

How many times faster Feather M4 is than PIC24FJ64GU205, operations per second. Blue: Feather M4 ahead. Orange: PIC24FJ64GU205 ahead.

AddSubtractMultiplyDivideModulo
uint8_t 6.30× 8.98 M/s · 1.43 M/s 6.30× 8.98 M/s · 1.43 M/s 6.86× 8.98 M/s · 1.31 M/s 14.1× 7.01 M/s · 497 k/s 12.8× 6.36 M/s · 497 k/s
uint16_t 7.84× 8.98 M/s · 1.15 M/s 7.84× 8.98 M/s · 1.15 M/s 7.84× 8.98 M/s · 1.15 M/s 12.6× 6.12 M/s · 485 k/s 11.1× 5.36 M/s · 485 k/s
uint32_t 10.9× 9.71 M/s · 890 k/s 10.9× 9.71 M/s · 890 k/s 15.1× 9.71 M/s · 641 k/s 190× 6.53 M/s · 34.3 k/s 183× 6.19 M/s · 33.8 k/s
uint64_t 10.8× 6.19 M/s · 572 k/s 9.28× 6.19 M/s · 668 k/s 42.2× 5.61 M/s · 133 k/s 405× 885 k/s · 2.19 k/s 405× 885 k/s · 2.18 k/s
float 67.6× 8.98 M/s · 133 k/s 73.8× 8.98 M/s · 122 k/s 53.6× 8.98 M/s · 168 k/s 60.4× 4.73 M/s · 78.2 k/s 9.00× 866 k/s · 96.3 k/s
double 9.24× 1.2 M/s · 130 k/s 10.0× 1.19 M/s · 119 k/s 10.5× 1.7 M/s · 162 k/s 16.3× 1.26 M/s · 77.1 k/s 4.50× 425 k/s · 94.5 k/s

Where they sit in the fleet

Keep comparing

Swap sides