Average Error: 2.2 → 2.2
Time: 950.0ms
Precision: binary64
\[\left(\left(\frac{1}{x + 1} - \frac{2}{x}\right) + \frac{1}{x - 1}\right) - timeout1000\]
↓
\[\left(\left(\frac{1}{x + 1} - \frac{2}{x}\right) + \frac{1}{x - 1}\right) - timeout1000\]
\left(\left(\frac{1}{x + 1} - \frac{2}{x}\right) + \frac{1}{x - 1}\right) - timeout1000
↓
\left(\left(\frac{1}{x + 1} - \frac{2}{x}\right) + \frac{1}{x - 1}\right) - timeout1000
double code(double x, double timeout1000) {
	return ((double) (((double) (((double) (((double) (1.0 / ((double) (x + 1.0)))) - ((double) (2.0 / x)))) + ((double) (1.0 / ((double) (x - 1.0)))))) - timeout1000));
}
↓
double code(double x, double timeout1000) {
	return ((double) (((double) (((double) (((double) (1.0 / ((double) (x + 1.0)))) - ((double) (2.0 / x)))) + ((double) (1.0 / ((double) (x - 1.0)))))) - timeout1000));
}

Error

Bits error versus x

Bits error versus timeout1000

Try it out

Your Program's Arguments

Results

Enter valid numbers for all inputs

Derivation

  1. Initial program 2.2

    \[\left(\left(\frac{1}{x + 1} - \frac{2}{x}\right) + \frac{1}{x - 1}\right) - timeout1000\]

  2. Final simplification2.2

    \[\leadsto \left(\left(\frac{1}{x + 1} - \frac{2}{x}\right) + \frac{1}{x - 1}\right) - timeout1000\]

Reproduce

herbie shell --seed 2020152 
(FPCore (x timeout1000)
  :name "(- (+ (- (/ 1 (+ x 1)) (/ 2 x)) (/ 1 (- x 1))) timeout1000)"
  :precision binary64
  (- (+ (- (/ 1.0 (+ x 1.0)) (/ 2.0 x)) (/ 1.0 (- x 1.0))) timeout1000))