JavaScript bit Operation notes

It has been thought that in the bit operation of JS, the double-precision floating-point number will be converted into a 32-bit integer, then perform bit operation and then convert to 64-bit value, so the efficiency is very low. Today's experiment found that the bitwise operation on the floating point will only result in inaccurate values, without causing inefficiency, and the left shift is 66% faster than multiplication (in fact, the gap between floating-point and integer operations).

The following results were tested on IE 7 ~ ie 10

Tip 1: The shift operation is faster than the multiplication method (when the factor is a power of 2)

b = a << 1; // probably fast 6.8% .

Tip 2: Parity and even judgment

b = A & 1; // 0-bit vs. 1 phase, approximately 35% compared to x 2

Tip 3: Determine if the two-digit symbol is the same

 (a ^ b) >= 0//  compared to  if  (A = = 0)  return  1; else  if  (A > 0)  return  b >= 0;  else  return  b <= 0;  //  its efficiency is basically the same, but more concise.  //  compared to   * b) >= 0;  //  There is no overflow problem. 

Tip 4: Determine if a number is a power of 2

(A & (a-1)) = = 0;

JavaScript bit Operation notes