Average Error: 1.6 → 0.2
Time: 4.1s
Precision: binary64
\[\]
↓
\[\]
↓
double code(double x, double y, double z) {
	return ((double) fabs(((double) (((double) (((double) (x + 4.0)) / y)) - ((double) (((double) (x / y)) * z))))));
}

↓
double code(double x, double y, double z) {
	double VAR;
	if (((x <= -1.3397329542554413e+35) || !(x <= 2.9490020604149927e-58))) {
		VAR = ((double) fabs(((double) (((double) (((double) (x + 4.0)) / y)) - ((double) (x * ((double) (z / y))))))));
	} else {
		VAR = ((double) fabs(((double) (((double) (x + ((double) (4.0 - ((double) (x * z)))))) / y))));
	}
	return VAR;
}

Error
Bits error versus x
Bits error versus y
Bits error versus z
Try it out
Your Program's Arguments
Results
Enter valid numbers for all inputs
Derivation
  1. Split input into 2 regimes
  2. if  x  < -1.3397329542554413e35 or 2.9490020604149927e-58 <  x 
    1. Initial program 0.2
      \[\]
    2. Simplified0.3
      \[\leadsto \]
    if -1.3397329542554413e35 <  x  < 2.9490020604149927e-58
    1. Initial program 2.5
      \[\]
    2. Using strategy rm
    3. Applied associate-*l/0.1
      \[\leadsto \]
    4. Applied sub-div0.1
      \[\leadsto \]
    5. Simplified0.1
      \[\leadsto \]
  3. Recombined 2 regimes into one program.
  4. Final simplification0.2
    \[\leadsto \]
Reproduce
herbie shell --seed 2020181 
(FPCore (x y z)
  :name "fabs fraction 1"
  :precision binary64
  (fabs (- (/ (+ x 4.0) y) (* (/ x y) z))))