Thing Done

← The fleet

PIC24FJ64GU205 vs PIC18F57Q84

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

PIC24FJ64GU205 Curiosity Nano

PIC24FJ64GU205

PIC24F (PIC24FJ64GU205)
Family
PIC / dsPIC
Board
PIC24FJ64GU205 Curiosity Nano
Clock
32 MHz
float / double
32-bit / 32-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
4.30×
PIC24FJ64GU205 faster, geometric mean of 25 cells
Biggest win for PIC24FJ64GU205
27.3×
uint32_t multiply
Closest cell
1.18×
uint32_t modulo
Clock for clock
8.59×
PIC24FJ64GU205 needs fewer cycles per op (2.00× clock gap)

Operation by operation

Every cell

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

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

Where they sit in the fleet

Keep comparing

Swap sides