Average Error: 8.0 → 8.0
Time: 779.0ms
Precision: binary64
\[\frac{M \cdot c - x}{c - 1}\]
↓
\[\frac{M \cdot c - x}{c - 1}\]
\frac{M \cdot c - x}{c - 1}
↓
\frac{M \cdot c - x}{c - 1}
double code(double M, double c, double x) {
	return ((double) (((double) (((double) (M * c)) - x)) / ((double) (c - 1.0))));
}
↓
double code(double M, double c, double x) {
	return ((double) (((double) (((double) (M * c)) - x)) / ((double) (c - 1.0))));
}

Error

Bits error versus M

Bits error versus c

Bits error versus x

Try it out

Your Program's Arguments

Results

Enter valid numbers for all inputs

Derivation

  1. Initial program 8.0

    \[\frac{M \cdot c - x}{c - 1}\]

  2. Final simplification8.0

    \[\leadsto \frac{M \cdot c - x}{c - 1}\]

Reproduce

herbie shell --seed 2020152 
(FPCore (M c x)
  :name "(/ (- (* M c) x) (- c 1))"
  :precision binary64
  (/ (- (* M c) x) (- c 1.0)))