Thing Done

← The fleet

PIC18F57Q84 vs PIC24FJ64GU205

PIC18F (PIC18F57Q84) at 64 MHz against PIC24F (PIC24FJ64GU205) at 32 MHz, on real hardware, same C kernels, clocks verified.

PIC18F57Q84 Curiosity Nano

PIC18F57Q84

PIC18F (PIC18F57Q84)
Family
PIC / dsPIC
Board
PIC18F57Q84 Curiosity Nano
Clock
64 MHz
float / double
32-bit / 32-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
4.30×
PIC24FJ64GU205 faster, geometric mean of 25 cells
Biggest win for PIC18F57Q84
0.85×
uint32_t modulo
Biggest win for PIC24FJ64GU205
27.3×
uint32_t multiply
Clock for clock
8.59×
PIC24FJ64GU205 needs fewer cycles per op (2.00× clock gap)

Operation by operation

Every cell

How many times faster PIC18F57Q84 is than PIC24FJ64GU205, operations per second. Blue: PIC18F57Q84 ahead. Orange: PIC24FJ64GU205 ahead.

AddSubtractMultiplyDivideModulo
uint8_t 1/2.65× 538 k/s · 1.43 M/s 1/2.83× 504 k/s · 1.43 M/s 1/2.52× 520 k/s · 1.31 M/s 1/3.67× 135 k/s · 497 k/s 1/4.68× 106 k/s · 497 k/s
uint16_t 1/3.49× 328 k/s · 1.15 M/s 1/3.49× 328 k/s · 1.15 M/s 1/3.99× 287 k/s · 1.15 M/s 1/7.81× 62.1 k/s · 485 k/s 1/6.88× 70.6 k/s · 485 k/s
uint32_t 1/3.72× 240 k/s · 890 k/s 1/3.71× 240 k/s · 890 k/s 1/27.3× 23.5 k/s · 641 k/s 1/1.40× 24.5 k/s · 34.3 k/s 1/1.18× 28.6 k/s · 33.8 k/s
uint64_t not measurednot measurednot measurednot measurednot measured
float 1/3.55× 37.4 k/s · 133 k/s 1/3.84× 31.7 k/s · 122 k/s 1/7.43× 22.6 k/s · 168 k/s 1/7.67× 10.2 k/s · 78.2 k/s 1/4.86× 19.8 k/s · 96.3 k/s
double 1/3.46× 37.4 k/s · 130 k/s 1/3.75× 31.7 k/s · 119 k/s 1/7.20× 22.6 k/s · 162 k/s 1/7.56× 10.2 k/s · 77.1 k/s 1/4.78× 19.8 k/s · 94.5 k/s

Where they sit in the fleet

Keep comparing

Swap sides