Diagrams.TwoD.Apollonian:initialConfig from diagrams-contrib-1.3.0.5, A

Percentage Accurate: 69.3% → 99.9%

Time: 8.5s

Alternatives: 8

Speedup: 1.2×

• 42.1% of points produce a very large (infinite) output. You may want to add a precondition.

Specification

Math

\[\begin{array}{l} \\ \frac{\left(x \cdot x + y \cdot y\right) - z \cdot z}{y \cdot 2} \end{array} \]

FPCore

(FPCore (x y z)
 :precision binary64
 (/ (- (+ (* x x) (* y y)) (* z z)) (* y 2.0)))

C

double code(double x, double y, double z) {
	return (((x * x) + (y * y)) - (z * z)) / (y * 2.0);
}

Fortran

real(8) function code(x, y, z)
    real(8), intent (in) :: x
    real(8), intent (in) :: y
    real(8), intent (in) :: z
    code = (((x * x) + (y * y)) - (z * z)) / (y * 2.0d0)
end function

Java

public static double code(double x, double y, double z) {
	return (((x * x) + (y * y)) - (z * z)) / (y * 2.0);
}

Python

def code(x, y, z):
	return (((x * x) + (y * y)) - (z * z)) / (y * 2.0)

Julia

function code(x, y, z)
	return Float64(Float64(Float64(Float64(x * x) + Float64(y * y)) - Float64(z * z)) / Float64(y * 2.0))
end

MATLAB

function tmp = code(x, y, z)
	tmp = (((x * x) + (y * y)) - (z * z)) / (y * 2.0);
end

Wolfram

code[x_, y_, z_] := N[(N[(N[(N[(x * x), $MachinePrecision] + N[(y * y), $MachinePrecision]), $MachinePrecision] - N[(z * z), $MachinePrecision]), $MachinePrecision] / N[(y * 2.0), $MachinePrecision]), $MachinePrecision]

Initial Program: 69.3% accurate, 1.0× speedup

Math

\[\begin{array}{l} \\ \frac{\left(x \cdot x + y \cdot y\right) - z \cdot z}{y \cdot 2} \end{array} \]

FPCore

(FPCore (x y z)
 :precision binary64
 (/ (- (+ (* x x) (* y y)) (* z z)) (* y 2.0)))

C

double code(double x, double y, double z) {
	return (((x * x) + (y * y)) - (z * z)) / (y * 2.0);
}

Fortran

real(8) function code(x, y, z)
    real(8), intent (in) :: x
    real(8), intent (in) :: y
    real(8), intent (in) :: z
    code = (((x * x) + (y * y)) - (z * z)) / (y * 2.0d0)
end function

Java

public static double code(double x, double y, double z) {
	return (((x * x) + (y * y)) - (z * z)) / (y * 2.0);
}

Python

def code(x, y, z):
	return (((x * x) + (y * y)) - (z * z)) / (y * 2.0)

Julia

function code(x, y, z)
	return Float64(Float64(Float64(Float64(x * x) + Float64(y * y)) - Float64(z * z)) / Float64(y * 2.0))
end

MATLAB

function tmp = code(x, y, z)
	tmp = (((x * x) + (y * y)) - (z * z)) / (y * 2.0);
end

Wolfram

code[x_, y_, z_] := N[(N[(N[(N[(x * x), $MachinePrecision] + N[(y * y), $MachinePrecision]), $MachinePrecision] - N[(z * z), $MachinePrecision]), $MachinePrecision] / N[(y * 2.0), $MachinePrecision]), $MachinePrecision]

Alternative 1: 99.9% accurate, 1.2× speedup

Math

\[\begin{array}{l} z_m = \left|z\right| \\ x_m = \left|x\right| \\ 0.5 \cdot \left(\left(z\_m + x\_m\right) \cdot \frac{x\_m - z\_m}{y} + y\right) \end{array} \]

FPCore

z_m = (fabs.f64 z)
x_m = (fabs.f64 x)
(FPCore (x_m y z_m)
 :precision binary64
 (* 0.5 (+ (* (+ z_m x_m) (/ (- x_m z_m) y)) y)))

C

z_m = fabs(z);
x_m = fabs(x);
double code(double x_m, double y, double z_m) {
	return 0.5 * (((z_m + x_m) * ((x_m - z_m) / y)) + y);
}

Fortran

z_m = abs(z)
x_m = abs(x)
real(8) function code(x_m, y, z_m)
    real(8), intent (in) :: x_m
    real(8), intent (in) :: y
    real(8), intent (in) :: z_m
    code = 0.5d0 * (((z_m + x_m) * ((x_m - z_m) / y)) + y)
end function

Java

z_m = Math.abs(z);
x_m = Math.abs(x);
public static double code(double x_m, double y, double z_m) {
	return 0.5 * (((z_m + x_m) * ((x_m - z_m) / y)) + y);
}

Python

z_m = math.fabs(z)
x_m = math.fabs(x)
def code(x_m, y, z_m):
	return 0.5 * (((z_m + x_m) * ((x_m - z_m) / y)) + y)

Julia

z_m = abs(z)
x_m = abs(x)
function code(x_m, y, z_m)
	return Float64(0.5 * Float64(Float64(Float64(z_m + x_m) * Float64(Float64(x_m - z_m) / y)) + y))
end

MATLAB

z_m = abs(z);
x_m = abs(x);
function tmp = code(x_m, y, z_m)
	tmp = 0.5 * (((z_m + x_m) * ((x_m - z_m) / y)) + y);
end

Wolfram

z_m = N[Abs[z], $MachinePrecision]
x_m = N[Abs[x], $MachinePrecision]
code[x$95$m_, y_, z$95$m_] := N[(0.5 * N[(N[(N[(z$95$m + x$95$m), $MachinePrecision] * N[(N[(x$95$m - z$95$m), $MachinePrecision] / y), $MachinePrecision]), $MachinePrecision] + y), $MachinePrecision]), $MachinePrecision]

Derivation

1. Initial program 71.0%

   \[\frac{\left(x \cdot x + y \cdot y\right) - z \cdot z}{y \cdot 2} \]
2. Add Preprocessing

3. Taylor expanded in y around inf

   \[\leadsto \color{blue}{\frac{1}{2} \cdot y} \]
4. Step-by-step derivation

   1. *-commutative N/A

      \[\leadsto \color{blue}{y \cdot \frac{1}{2}} \]

   2. lower-*.f64 37.2

      \[\leadsto \color{blue}{y \cdot 0.5} \]

5. Applied rewrites 37.2%

   \[\leadsto \color{blue}{y \cdot 0.5} \]

6. Taylor expanded in x around 0

   \[\leadsto \color{blue}{\frac{1}{2} \cdot \frac{{y}^{2} - {z}^{2}}{y}} \]
7. Step-by-step derivation

   1. *-commutative N/A

      \[\leadsto \color{blue}{\frac{{y}^{2} - {z}^{2}}{y} \cdot \frac{1}{2}} \]

   2. unpow2 N/A

      \[\leadsto \frac{\color{blue}{y \cdot y} - {z}^{2}}{y} \cdot \frac{1}{2} \]

   3. unpow2 N/A

      \[\leadsto \frac{y \cdot y - \color{blue}{z \cdot z}}{y} \cdot \frac{1}{2} \]

   4. difference-of-squares N/A

      \[\leadsto \frac{\color{blue}{\left(y + z\right) \cdot \left(y - z\right)}}{y} \cdot \frac{1}{2} \]

   5. lower-*.f64 N/A

      \[\leadsto \color{blue}{\frac{\left(y + z\right) \cdot \left(y - z\right)}{y} \cdot \frac{1}{2}} \]

8. Applied rewrites 67.8%

   \[\leadsto \color{blue}{\mathsf{fma}\left(-z, \frac{z}{y}, y\right) \cdot 0.5} \]

9. Taylor expanded in x around 0

   \[\leadsto \color{blue}{\frac{1}{2} \cdot \frac{{y}^{2} - {z}^{2}}{y} + \frac{1}{2} \cdot \frac{{x}^{2}}{y}} \]
10. Step-by-step derivation

    1. distribute-lft-out N/A

       \[\leadsto \color{blue}{\frac{1}{2} \cdot \left(\frac{{y}^{2} - {z}^{2}}{y} + \frac{{x}^{2}}{y}\right)} \]

    2. *-commutative N/A

       \[\leadsto \color{blue}{\left(\frac{{y}^{2} - {z}^{2}}{y} + \frac{{x}^{2}}{y}\right) \cdot \frac{1}{2}} \]

    3. lower-*.f64 N/A

       \[\leadsto \color{blue}{\left(\frac{{y}^{2} - {z}^{2}}{y} + \frac{{x}^{2}}{y}\right) \cdot \frac{1}{2}} \]

11. Applied rewrites 100.0%

    \[\leadsto \color{blue}{\left(y + \frac{x - z}{y} \cdot \left(z + x\right)\right) \cdot 0.5} \]

12. Final simplification 100.0%

    \[\leadsto 0.5 \cdot \left(\left(z + x\right) \cdot \frac{x - z}{y} + y\right) \]
13. Add Preprocessing

Alternative 2: 39.0% accurate, 0.3× speedup

Math

\[\begin{array}{l} z_m = \left|z\right| \\ x_m = \left|x\right| \\ \begin{array}{l} t_0 := \frac{\left(y \cdot y + x\_m \cdot x\_m\right) - z\_m \cdot z\_m}{2 \cdot y}\\ \mathbf{if}\;t\_0 \leq -1 \cdot 10^{-50}:\\ \;\;\;\;-0.5 \cdot \frac{z\_m \cdot z\_m}{y}\\ \mathbf{elif}\;t\_0 \leq 10^{+148}:\\ \;\;\;\;0.5 \cdot y\\ \mathbf{elif}\;t\_0 \leq \infty:\\ \;\;\;\;\left(\frac{x\_m}{y} \cdot x\_m\right) \cdot 0.5\\ \mathbf{else}:\\ \;\;\;\;\left(\frac{z\_m}{y} \cdot z\_m\right) \cdot -0.5\\ \end{array} \end{array} \]

FPCore

z_m = (fabs.f64 z)
x_m = (fabs.f64 x)
(FPCore (x_m y z_m)
 :precision binary64
 (let* ((t_0 (/ (- (+ (* y y) (* x_m x_m)) (* z_m z_m)) (* 2.0 y))))
   (if (<= t_0 -1e-50)
     (* -0.5 (/ (* z_m z_m) y))
     (if (<= t_0 1e+148)
       (* 0.5 y)
       (if (<= t_0 INFINITY)
         (* (* (/ x_m y) x_m) 0.5)
         (* (* (/ z_m y) z_m) -0.5))))))

C

z_m = fabs(z);
x_m = fabs(x);
double code(double x_m, double y, double z_m) {
	double t_0 = (((y * y) + (x_m * x_m)) - (z_m * z_m)) / (2.0 * y);
	double tmp;
	if (t_0 <= -1e-50) {
		tmp = -0.5 * ((z_m * z_m) / y);
	} else if (t_0 <= 1e+148) {
		tmp = 0.5 * y;
	} else if (t_0 <= ((double) INFINITY)) {
		tmp = ((x_m / y) * x_m) * 0.5;
	} else {
		tmp = ((z_m / y) * z_m) * -0.5;
	}
	return tmp;
}

Java

z_m = Math.abs(z);
x_m = Math.abs(x);
public static double code(double x_m, double y, double z_m) {
	double t_0 = (((y * y) + (x_m * x_m)) - (z_m * z_m)) / (2.0 * y);
	double tmp;
	if (t_0 <= -1e-50) {
		tmp = -0.5 * ((z_m * z_m) / y);
	} else if (t_0 <= 1e+148) {
		tmp = 0.5 * y;
	} else if (t_0 <= Double.POSITIVE_INFINITY) {
		tmp = ((x_m / y) * x_m) * 0.5;
	} else {
		tmp = ((z_m / y) * z_m) * -0.5;
	}
	return tmp;
}

Python

z_m = math.fabs(z)
x_m = math.fabs(x)
def code(x_m, y, z_m):
	t_0 = (((y * y) + (x_m * x_m)) - (z_m * z_m)) / (2.0 * y)
	tmp = 0
	if t_0 <= -1e-50:
		tmp = -0.5 * ((z_m * z_m) / y)
	elif t_0 <= 1e+148:
		tmp = 0.5 * y
	elif t_0 <= math.inf:
		tmp = ((x_m / y) * x_m) * 0.5
	else:
		tmp = ((z_m / y) * z_m) * -0.5
	return tmp

Julia

z_m = abs(z)
x_m = abs(x)
function code(x_m, y, z_m)
	t_0 = Float64(Float64(Float64(Float64(y * y) + Float64(x_m * x_m)) - Float64(z_m * z_m)) / Float64(2.0 * y))
	tmp = 0.0
	if (t_0 <= -1e-50)
		tmp = Float64(-0.5 * Float64(Float64(z_m * z_m) / y));
	elseif (t_0 <= 1e+148)
		tmp = Float64(0.5 * y);
	elseif (t_0 <= Inf)
		tmp = Float64(Float64(Float64(x_m / y) * x_m) * 0.5);
	else
		tmp = Float64(Float64(Float64(z_m / y) * z_m) * -0.5);
	end
	return tmp
end

MATLAB

z_m = abs(z);
x_m = abs(x);
function tmp_2 = code(x_m, y, z_m)
	t_0 = (((y * y) + (x_m * x_m)) - (z_m * z_m)) / (2.0 * y);
	tmp = 0.0;
	if (t_0 <= -1e-50)
		tmp = -0.5 * ((z_m * z_m) / y);
	elseif (t_0 <= 1e+148)
		tmp = 0.5 * y;
	elseif (t_0 <= Inf)
		tmp = ((x_m / y) * x_m) * 0.5;
	else
		tmp = ((z_m / y) * z_m) * -0.5;
	end
	tmp_2 = tmp;
end

Wolfram

z_m = N[Abs[z], $MachinePrecision]
x_m = N[Abs[x], $MachinePrecision]
code[x$95$m_, y_, z$95$m_] := Block[{t$95$0 = N[(N[(N[(N[(y * y), $MachinePrecision] + N[(x$95$m * x$95$m), $MachinePrecision]), $MachinePrecision] - N[(z$95$m * z$95$m), $MachinePrecision]), $MachinePrecision] / N[(2.0 * y), $MachinePrecision]), $MachinePrecision]}, If[LessEqual[t$95$0, -1e-50], N[(-0.5 * N[(N[(z$95$m * z$95$m), $MachinePrecision] / y), $MachinePrecision]), $MachinePrecision], If[LessEqual[t$95$0, 1e+148], N[(0.5 * y), $MachinePrecision], If[LessEqual[t$95$0, Infinity], N[(N[(N[(x$95$m / y), $MachinePrecision] * x$95$m), $MachinePrecision] * 0.5), $MachinePrecision], N[(N[(N[(z$95$m / y), $MachinePrecision] * z$95$m), $MachinePrecision] * -0.5), $MachinePrecision]]]]]

Derivation

1. Split input into 4 regimes

2. if (/.f64 (-.f64 (+.f64 (*.f64 x x) (*.f64 y y)) (*.f64 z z)) (*.f64 y #s(literal 2 binary64))) < -1.00000000000000001e-50

   1. Initial program 77.7%

      \[\frac{\left(x \cdot x + y \cdot y\right) - z \cdot z}{y \cdot 2} \]
   2. Add Preprocessing

   3. Taylor expanded in z around inf

      \[\leadsto \color{blue}{\frac{-1}{2} \cdot \frac{{z}^{2}}{y}} \]
   4. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{\frac{{z}^{2}}{y} \cdot \frac{-1}{2}} \]

      2. lower-*.f64 N/A

         \[\leadsto \color{blue}{\frac{{z}^{2}}{y} \cdot \frac{-1}{2}} \]

      3. lower-/.f64 N/A

         \[\leadsto \color{blue}{\frac{{z}^{2}}{y}} \cdot \frac{-1}{2} \]

      4. unpow2 N/A

         \[\leadsto \frac{\color{blue}{z \cdot z}}{y} \cdot \frac{-1}{2} \]

      5. lower-*.f64 34.0

         \[\leadsto \frac{\color{blue}{z \cdot z}}{y} \cdot -0.5 \]

   5. Applied rewrites 34.0%

      \[\leadsto \color{blue}{\frac{z \cdot z}{y} \cdot -0.5} \]

   if -1.00000000000000001e-50 < (/.f64 (-.f64 (+.f64 (*.f64 x x) (*.f64 y y)) (*.f64 z z)) (*.f64 y #s(literal 2 binary64))) < 1e148

   1. Initial program 83.1%

      \[\frac{\left(x \cdot x + y \cdot y\right) - z \cdot z}{y \cdot 2} \]
   2. Add Preprocessing

   3. Taylor expanded in y around inf

      \[\leadsto \color{blue}{\frac{1}{2} \cdot y} \]
   4. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{y \cdot \frac{1}{2}} \]

      2. lower-*.f64 66.2

         \[\leadsto \color{blue}{y \cdot 0.5} \]

   5. Applied rewrites 66.2%

      \[\leadsto \color{blue}{y \cdot 0.5} \]

   if 1e148 < (/.f64 (-.f64 (+.f64 (*.f64 x x) (*.f64 y y)) (*.f64 z z)) (*.f64 y #s(literal 2 binary64))) < +inf.0

   1. Initial program 74.0%

      \[\frac{\left(x \cdot x + y \cdot y\right) - z \cdot z}{y \cdot 2} \]
   2. Add Preprocessing

   3. Taylor expanded in y around inf

      \[\leadsto \color{blue}{\frac{1}{2} \cdot y} \]
   4. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{y \cdot \frac{1}{2}} \]

      2. lower-*.f64 27.6

         \[\leadsto \color{blue}{y \cdot 0.5} \]

   5. Applied rewrites 27.6%

      \[\leadsto \color{blue}{y \cdot 0.5} \]

   6. Taylor expanded in x around inf

      \[\leadsto \color{blue}{\frac{1}{2} \cdot \frac{{x}^{2}}{y}} \]
   7. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{\frac{{x}^{2}}{y} \cdot \frac{1}{2}} \]

      2. lower-*.f64 N/A

         \[\leadsto \color{blue}{\frac{{x}^{2}}{y} \cdot \frac{1}{2}} \]

      3. unpow2 N/A

         \[\leadsto \frac{\color{blue}{x \cdot x}}{y} \cdot \frac{1}{2} \]

      4. associate-*l/ N/A

         \[\leadsto \color{blue}{\left(\frac{x}{y} \cdot x\right)} \cdot \frac{1}{2} \]

      5. lower-*.f64 N/A

         \[\leadsto \color{blue}{\left(\frac{x}{y} \cdot x\right)} \cdot \frac{1}{2} \]

      6. lower-/.f64 49.8

         \[\leadsto \left(\color{blue}{\frac{x}{y}} \cdot x\right) \cdot 0.5 \]

   8. Applied rewrites 49.8%

      \[\leadsto \color{blue}{\left(\frac{x}{y} \cdot x\right) \cdot 0.5} \]

   if +inf.0 < (/.f64 (-.f64 (+.f64 (*.f64 x x) (*.f64 y y)) (*.f64 z z)) (*.f64 y #s(literal 2 binary64)))

   1. Initial program 0.0%

      \[\frac{\left(x \cdot x + y \cdot y\right) - z \cdot z}{y \cdot 2} \]
   2. Add Preprocessing

   3. Taylor expanded in z around inf

      \[\leadsto \color{blue}{\frac{-1}{2} \cdot \frac{{z}^{2}}{y}} \]
   4. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{\frac{{z}^{2}}{y} \cdot \frac{-1}{2}} \]

      2. lower-*.f64 N/A

         \[\leadsto \color{blue}{\frac{{z}^{2}}{y} \cdot \frac{-1}{2}} \]

      3. lower-/.f64 N/A

         \[\leadsto \color{blue}{\frac{{z}^{2}}{y}} \cdot \frac{-1}{2} \]

      4. unpow2 N/A

         \[\leadsto \frac{\color{blue}{z \cdot z}}{y} \cdot \frac{-1}{2} \]

      5. lower-*.f64 45.8

         \[\leadsto \frac{\color{blue}{z \cdot z}}{y} \cdot -0.5 \]

   5. Applied rewrites 45.8%

      \[\leadsto \color{blue}{\frac{z \cdot z}{y} \cdot -0.5} \]
   6. Step-by-step derivation

   7. Applied rewrites 55.2%

      \[\leadsto \left(\frac{z}{y} \cdot z\right) \cdot \color{blue}{-0.5} \]
3. Recombined 4 regimes into one program.

4. Final simplification 44.5%

   \[\leadsto \begin{array}{l} \mathbf{if}\;\frac{\left(y \cdot y + x \cdot x\right) - z \cdot z}{2 \cdot y} \leq -1 \cdot 10^{-50}:\\ \;\;\;\;-0.5 \cdot \frac{z \cdot z}{y}\\ \mathbf{elif}\;\frac{\left(y \cdot y + x \cdot x\right) - z \cdot z}{2 \cdot y} \leq 10^{+148}:\\ \;\;\;\;0.5 \cdot y\\ \mathbf{elif}\;\frac{\left(y \cdot y + x \cdot x\right) - z \cdot z}{2 \cdot y} \leq \infty:\\ \;\;\;\;\left(\frac{x}{y} \cdot x\right) \cdot 0.5\\ \mathbf{else}:\\ \;\;\;\;\left(\frac{z}{y} \cdot z\right) \cdot -0.5\\ \end{array} \]
5. Add Preprocessing

Alternative 3: 39.7% accurate, 0.3× speedup

Math

\[\begin{array}{l} z_m = \left|z\right| \\ x_m = \left|x\right| \\ \begin{array}{l} t_0 := \left(\frac{z\_m}{y} \cdot z\_m\right) \cdot -0.5\\ t_1 := \frac{\left(y \cdot y + x\_m \cdot x\_m\right) - z\_m \cdot z\_m}{2 \cdot y}\\ \mathbf{if}\;t\_1 \leq -1 \cdot 10^{-50}:\\ \;\;\;\;t\_0\\ \mathbf{elif}\;t\_1 \leq 10^{+148}:\\ \;\;\;\;0.5 \cdot y\\ \mathbf{elif}\;t\_1 \leq \infty:\\ \;\;\;\;\left(\frac{x\_m}{y} \cdot x\_m\right) \cdot 0.5\\ \mathbf{else}:\\ \;\;\;\;t\_0\\ \end{array} \end{array} \]

FPCore

z_m = (fabs.f64 z)
x_m = (fabs.f64 x)
(FPCore (x_m y z_m)
 :precision binary64
 (let* ((t_0 (* (* (/ z_m y) z_m) -0.5))
        (t_1 (/ (- (+ (* y y) (* x_m x_m)) (* z_m z_m)) (* 2.0 y))))
   (if (<= t_1 -1e-50)
     t_0
     (if (<= t_1 1e+148)
       (* 0.5 y)
       (if (<= t_1 INFINITY) (* (* (/ x_m y) x_m) 0.5) t_0)))))

C

z_m = fabs(z);
x_m = fabs(x);
double code(double x_m, double y, double z_m) {
	double t_0 = ((z_m / y) * z_m) * -0.5;
	double t_1 = (((y * y) + (x_m * x_m)) - (z_m * z_m)) / (2.0 * y);
	double tmp;
	if (t_1 <= -1e-50) {
		tmp = t_0;
	} else if (t_1 <= 1e+148) {
		tmp = 0.5 * y;
	} else if (t_1 <= ((double) INFINITY)) {
		tmp = ((x_m / y) * x_m) * 0.5;
	} else {
		tmp = t_0;
	}
	return tmp;
}

Java

z_m = Math.abs(z);
x_m = Math.abs(x);
public static double code(double x_m, double y, double z_m) {
	double t_0 = ((z_m / y) * z_m) * -0.5;
	double t_1 = (((y * y) + (x_m * x_m)) - (z_m * z_m)) / (2.0 * y);
	double tmp;
	if (t_1 <= -1e-50) {
		tmp = t_0;
	} else if (t_1 <= 1e+148) {
		tmp = 0.5 * y;
	} else if (t_1 <= Double.POSITIVE_INFINITY) {
		tmp = ((x_m / y) * x_m) * 0.5;
	} else {
		tmp = t_0;
	}
	return tmp;
}

Python

z_m = math.fabs(z)
x_m = math.fabs(x)
def code(x_m, y, z_m):
	t_0 = ((z_m / y) * z_m) * -0.5
	t_1 = (((y * y) + (x_m * x_m)) - (z_m * z_m)) / (2.0 * y)
	tmp = 0
	if t_1 <= -1e-50:
		tmp = t_0
	elif t_1 <= 1e+148:
		tmp = 0.5 * y
	elif t_1 <= math.inf:
		tmp = ((x_m / y) * x_m) * 0.5
	else:
		tmp = t_0
	return tmp

Julia

z_m = abs(z)
x_m = abs(x)
function code(x_m, y, z_m)
	t_0 = Float64(Float64(Float64(z_m / y) * z_m) * -0.5)
	t_1 = Float64(Float64(Float64(Float64(y * y) + Float64(x_m * x_m)) - Float64(z_m * z_m)) / Float64(2.0 * y))
	tmp = 0.0
	if (t_1 <= -1e-50)
		tmp = t_0;
	elseif (t_1 <= 1e+148)
		tmp = Float64(0.5 * y);
	elseif (t_1 <= Inf)
		tmp = Float64(Float64(Float64(x_m / y) * x_m) * 0.5);
	else
		tmp = t_0;
	end
	return tmp
end

MATLAB

z_m = abs(z);
x_m = abs(x);
function tmp_2 = code(x_m, y, z_m)
	t_0 = ((z_m / y) * z_m) * -0.5;
	t_1 = (((y * y) + (x_m * x_m)) - (z_m * z_m)) / (2.0 * y);
	tmp = 0.0;
	if (t_1 <= -1e-50)
		tmp = t_0;
	elseif (t_1 <= 1e+148)
		tmp = 0.5 * y;
	elseif (t_1 <= Inf)
		tmp = ((x_m / y) * x_m) * 0.5;
	else
		tmp = t_0;
	end
	tmp_2 = tmp;
end

Wolfram

z_m = N[Abs[z], $MachinePrecision]
x_m = N[Abs[x], $MachinePrecision]
code[x$95$m_, y_, z$95$m_] := Block[{t$95$0 = N[(N[(N[(z$95$m / y), $MachinePrecision] * z$95$m), $MachinePrecision] * -0.5), $MachinePrecision]}, Block[{t$95$1 = N[(N[(N[(N[(y * y), $MachinePrecision] + N[(x$95$m * x$95$m), $MachinePrecision]), $MachinePrecision] - N[(z$95$m * z$95$m), $MachinePrecision]), $MachinePrecision] / N[(2.0 * y), $MachinePrecision]), $MachinePrecision]}, If[LessEqual[t$95$1, -1e-50], t$95$0, If[LessEqual[t$95$1, 1e+148], N[(0.5 * y), $MachinePrecision], If[LessEqual[t$95$1, Infinity], N[(N[(N[(x$95$m / y), $MachinePrecision] * x$95$m), $MachinePrecision] * 0.5), $MachinePrecision], t$95$0]]]]]

Derivation

1. Split input into 3 regimes

2. if (/.f64 (-.f64 (+.f64 (*.f64 x x) (*.f64 y y)) (*.f64 z z)) (*.f64 y #s(literal 2 binary64))) < -1.00000000000000001e-50 or +inf.0 < (/.f64 (-.f64 (+.f64 (*.f64 x x) (*.f64 y y)) (*.f64 z z)) (*.f64 y #s(literal 2 binary64)))

   1. Initial program 66.8%

      \[\frac{\left(x \cdot x + y \cdot y\right) - z \cdot z}{y \cdot 2} \]
   2. Add Preprocessing

   3. Taylor expanded in z around inf

      \[\leadsto \color{blue}{\frac{-1}{2} \cdot \frac{{z}^{2}}{y}} \]
   4. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{\frac{{z}^{2}}{y} \cdot \frac{-1}{2}} \]

      2. lower-*.f64 N/A

         \[\leadsto \color{blue}{\frac{{z}^{2}}{y} \cdot \frac{-1}{2}} \]

      3. lower-/.f64 N/A

         \[\leadsto \color{blue}{\frac{{z}^{2}}{y}} \cdot \frac{-1}{2} \]

      4. unpow2 N/A

         \[\leadsto \frac{\color{blue}{z \cdot z}}{y} \cdot \frac{-1}{2} \]

      5. lower-*.f64 35.6

         \[\leadsto \frac{\color{blue}{z \cdot z}}{y} \cdot -0.5 \]

   5. Applied rewrites 35.6%

      \[\leadsto \color{blue}{\frac{z \cdot z}{y} \cdot -0.5} \]
   6. Step-by-step derivation

   7. Applied rewrites 36.9%

      \[\leadsto \left(\frac{z}{y} \cdot z\right) \cdot \color{blue}{-0.5} \]

   if -1.00000000000000001e-50 < (/.f64 (-.f64 (+.f64 (*.f64 x x) (*.f64 y y)) (*.f64 z z)) (*.f64 y #s(literal 2 binary64))) < 1e148

   1. Initial program 83.1%

      \[\frac{\left(x \cdot x + y \cdot y\right) - z \cdot z}{y \cdot 2} \]
   2. Add Preprocessing

   3. Taylor expanded in y around inf

      \[\leadsto \color{blue}{\frac{1}{2} \cdot y} \]
   4. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{y \cdot \frac{1}{2}} \]

      2. lower-*.f64 66.2

         \[\leadsto \color{blue}{y \cdot 0.5} \]

   5. Applied rewrites 66.2%

      \[\leadsto \color{blue}{y \cdot 0.5} \]

   if 1e148 < (/.f64 (-.f64 (+.f64 (*.f64 x x) (*.f64 y y)) (*.f64 z z)) (*.f64 y #s(literal 2 binary64))) < +inf.0

   1. Initial program 74.0%

      \[\frac{\left(x \cdot x + y \cdot y\right) - z \cdot z}{y \cdot 2} \]
   2. Add Preprocessing

   3. Taylor expanded in y around inf

      \[\leadsto \color{blue}{\frac{1}{2} \cdot y} \]
   4. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{y \cdot \frac{1}{2}} \]

      2. lower-*.f64 27.6

         \[\leadsto \color{blue}{y \cdot 0.5} \]

   5. Applied rewrites 27.6%

      \[\leadsto \color{blue}{y \cdot 0.5} \]

   6. Taylor expanded in x around inf

      \[\leadsto \color{blue}{\frac{1}{2} \cdot \frac{{x}^{2}}{y}} \]
   7. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{\frac{{x}^{2}}{y} \cdot \frac{1}{2}} \]

      2. lower-*.f64 N/A

         \[\leadsto \color{blue}{\frac{{x}^{2}}{y} \cdot \frac{1}{2}} \]

      3. unpow2 N/A

         \[\leadsto \frac{\color{blue}{x \cdot x}}{y} \cdot \frac{1}{2} \]

      4. associate-*l/ N/A

         \[\leadsto \color{blue}{\left(\frac{x}{y} \cdot x\right)} \cdot \frac{1}{2} \]

      5. lower-*.f64 N/A

         \[\leadsto \color{blue}{\left(\frac{x}{y} \cdot x\right)} \cdot \frac{1}{2} \]

      6. lower-/.f64 49.8

         \[\leadsto \left(\color{blue}{\frac{x}{y}} \cdot x\right) \cdot 0.5 \]

   8. Applied rewrites 49.8%

      \[\leadsto \color{blue}{\left(\frac{x}{y} \cdot x\right) \cdot 0.5} \]
3. Recombined 3 regimes into one program.

4. Final simplification 44.4%

   \[\leadsto \begin{array}{l} \mathbf{if}\;\frac{\left(y \cdot y + x \cdot x\right) - z \cdot z}{2 \cdot y} \leq -1 \cdot 10^{-50}:\\ \;\;\;\;\left(\frac{z}{y} \cdot z\right) \cdot -0.5\\ \mathbf{elif}\;\frac{\left(y \cdot y + x \cdot x\right) - z \cdot z}{2 \cdot y} \leq 10^{+148}:\\ \;\;\;\;0.5 \cdot y\\ \mathbf{elif}\;\frac{\left(y \cdot y + x \cdot x\right) - z \cdot z}{2 \cdot y} \leq \infty:\\ \;\;\;\;\left(\frac{x}{y} \cdot x\right) \cdot 0.5\\ \mathbf{else}:\\ \;\;\;\;\left(\frac{z}{y} \cdot z\right) \cdot -0.5\\ \end{array} \]
5. Add Preprocessing

Alternative 4: 67.9% accurate, 0.3× speedup

Math

\[\begin{array}{l} z_m = \left|z\right| \\ x_m = \left|x\right| \\ \begin{array}{l} t_0 := \mathsf{fma}\left(-z\_m, \frac{z\_m}{y}, y\right) \cdot 0.5\\ t_1 := \frac{\left(y \cdot y + x\_m \cdot x\_m\right) - z\_m \cdot z\_m}{2 \cdot y}\\ \mathbf{if}\;t\_1 \leq 0:\\ \;\;\;\;t\_0\\ \mathbf{elif}\;t\_1 \leq \infty:\\ \;\;\;\;\mathsf{fma}\left(\frac{x\_m}{y}, x\_m, y\right) \cdot 0.5\\ \mathbf{else}:\\ \;\;\;\;t\_0\\ \end{array} \end{array} \]

FPCore

z_m = (fabs.f64 z)
x_m = (fabs.f64 x)
(FPCore (x_m y z_m)
 :precision binary64
 (let* ((t_0 (* (fma (- z_m) (/ z_m y) y) 0.5))
        (t_1 (/ (- (+ (* y y) (* x_m x_m)) (* z_m z_m)) (* 2.0 y))))
   (if (<= t_1 0.0)
     t_0
     (if (<= t_1 INFINITY) (* (fma (/ x_m y) x_m y) 0.5) t_0))))

C

z_m = fabs(z);
x_m = fabs(x);
double code(double x_m, double y, double z_m) {
	double t_0 = fma(-z_m, (z_m / y), y) * 0.5;
	double t_1 = (((y * y) + (x_m * x_m)) - (z_m * z_m)) / (2.0 * y);
	double tmp;
	if (t_1 <= 0.0) {
		tmp = t_0;
	} else if (t_1 <= ((double) INFINITY)) {
		tmp = fma((x_m / y), x_m, y) * 0.5;
	} else {
		tmp = t_0;
	}
	return tmp;
}

Julia

z_m = abs(z)
x_m = abs(x)
function code(x_m, y, z_m)
	t_0 = Float64(fma(Float64(-z_m), Float64(z_m / y), y) * 0.5)
	t_1 = Float64(Float64(Float64(Float64(y * y) + Float64(x_m * x_m)) - Float64(z_m * z_m)) / Float64(2.0 * y))
	tmp = 0.0
	if (t_1 <= 0.0)
		tmp = t_0;
	elseif (t_1 <= Inf)
		tmp = Float64(fma(Float64(x_m / y), x_m, y) * 0.5);
	else
		tmp = t_0;
	end
	return tmp
end

Wolfram

z_m = N[Abs[z], $MachinePrecision]
x_m = N[Abs[x], $MachinePrecision]
code[x$95$m_, y_, z$95$m_] := Block[{t$95$0 = N[(N[((-z$95$m) * N[(z$95$m / y), $MachinePrecision] + y), $MachinePrecision] * 0.5), $MachinePrecision]}, Block[{t$95$1 = N[(N[(N[(N[(y * y), $MachinePrecision] + N[(x$95$m * x$95$m), $MachinePrecision]), $MachinePrecision] - N[(z$95$m * z$95$m), $MachinePrecision]), $MachinePrecision] / N[(2.0 * y), $MachinePrecision]), $MachinePrecision]}, If[LessEqual[t$95$1, 0.0], t$95$0, If[LessEqual[t$95$1, Infinity], N[(N[(N[(x$95$m / y), $MachinePrecision] * x$95$m + y), $MachinePrecision] * 0.5), $MachinePrecision], t$95$0]]]]

Derivation

1. Split input into 2 regimes

2. if (/.f64 (-.f64 (+.f64 (*.f64 x x) (*.f64 y y)) (*.f64 z z)) (*.f64 y #s(literal 2 binary64))) < 0.0 or +inf.0 < (/.f64 (-.f64 (+.f64 (*.f64 x x) (*.f64 y y)) (*.f64 z z)) (*.f64 y #s(literal 2 binary64)))

   1. Initial program 64.9%

      \[\frac{\left(x \cdot x + y \cdot y\right) - z \cdot z}{y \cdot 2} \]
   2. Add Preprocessing

   3. Taylor expanded in y around inf

      \[\leadsto \color{blue}{\frac{1}{2} \cdot y} \]
   4. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{y \cdot \frac{1}{2}} \]

      2. lower-*.f64 37.9

         \[\leadsto \color{blue}{y \cdot 0.5} \]

   5. Applied rewrites 37.9%

      \[\leadsto \color{blue}{y \cdot 0.5} \]

   6. Taylor expanded in x around 0

      \[\leadsto \color{blue}{\frac{1}{2} \cdot \frac{{y}^{2} - {z}^{2}}{y}} \]
   7. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{\frac{{y}^{2} - {z}^{2}}{y} \cdot \frac{1}{2}} \]

      2. unpow2 N/A

         \[\leadsto \frac{\color{blue}{y \cdot y} - {z}^{2}}{y} \cdot \frac{1}{2} \]

      3. unpow2 N/A

         \[\leadsto \frac{y \cdot y - \color{blue}{z \cdot z}}{y} \cdot \frac{1}{2} \]

      4. difference-of-squares N/A

         \[\leadsto \frac{\color{blue}{\left(y + z\right) \cdot \left(y - z\right)}}{y} \cdot \frac{1}{2} \]

      5. lower-*.f64 N/A

         \[\leadsto \color{blue}{\frac{\left(y + z\right) \cdot \left(y - z\right)}{y} \cdot \frac{1}{2}} \]

   8. Applied rewrites 73.1%

      \[\leadsto \color{blue}{\mathsf{fma}\left(-z, \frac{z}{y}, y\right) \cdot 0.5} \]

   if 0.0 < (/.f64 (-.f64 (+.f64 (*.f64 x x) (*.f64 y y)) (*.f64 z z)) (*.f64 y #s(literal 2 binary64))) < +inf.0

   1. Initial program 79.3%

      \[\frac{\left(x \cdot x + y \cdot y\right) - z \cdot z}{y \cdot 2} \]
   2. Add Preprocessing

   3. Taylor expanded in y around inf

      \[\leadsto \color{blue}{\frac{1}{2} \cdot y} \]
   4. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{y \cdot \frac{1}{2}} \]

      2. lower-*.f64 36.2

         \[\leadsto \color{blue}{y \cdot 0.5} \]

   5. Applied rewrites 36.2%

      \[\leadsto \color{blue}{y \cdot 0.5} \]

   6. Taylor expanded in x around 0

      \[\leadsto \color{blue}{\frac{1}{2} \cdot \frac{{y}^{2} - {z}^{2}}{y}} \]
   7. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{\frac{{y}^{2} - {z}^{2}}{y} \cdot \frac{1}{2}} \]

      2. unpow2 N/A

         \[\leadsto \frac{\color{blue}{y \cdot y} - {z}^{2}}{y} \cdot \frac{1}{2} \]

      3. unpow2 N/A

         \[\leadsto \frac{y \cdot y - \color{blue}{z \cdot z}}{y} \cdot \frac{1}{2} \]

      4. difference-of-squares N/A

         \[\leadsto \frac{\color{blue}{\left(y + z\right) \cdot \left(y - z\right)}}{y} \cdot \frac{1}{2} \]

      5. lower-*.f64 N/A

         \[\leadsto \color{blue}{\frac{\left(y + z\right) \cdot \left(y - z\right)}{y} \cdot \frac{1}{2}} \]

   8. Applied rewrites 60.4%

      \[\leadsto \color{blue}{\mathsf{fma}\left(-z, \frac{z}{y}, y\right) \cdot 0.5} \]

   9. Taylor expanded in x around 0

      \[\leadsto \color{blue}{\frac{1}{2} \cdot \frac{{y}^{2} - {z}^{2}}{y} + \frac{1}{2} \cdot \frac{{x}^{2}}{y}} \]
   10. Step-by-step derivation

       1. distribute-lft-out N/A

          \[\leadsto \color{blue}{\frac{1}{2} \cdot \left(\frac{{y}^{2} - {z}^{2}}{y} + \frac{{x}^{2}}{y}\right)} \]

       2. *-commutative N/A

          \[\leadsto \color{blue}{\left(\frac{{y}^{2} - {z}^{2}}{y} + \frac{{x}^{2}}{y}\right) \cdot \frac{1}{2}} \]

       3. lower-*.f64 N/A

          \[\leadsto \color{blue}{\left(\frac{{y}^{2} - {z}^{2}}{y} + \frac{{x}^{2}}{y}\right) \cdot \frac{1}{2}} \]

   11. Applied rewrites 99.9%

       \[\leadsto \color{blue}{\left(y + \frac{x - z}{y} \cdot \left(z + x\right)\right) \cdot 0.5} \]

   12. Taylor expanded in z around 0

       \[\leadsto \left(y + \frac{{x}^{2}}{y}\right) \cdot \frac{1}{2} \]
   13. Step-by-step derivation

   14. Applied rewrites 75.2%

       \[\leadsto \mathsf{fma}\left(\frac{x}{y}, x, y\right) \cdot 0.5 \]
3. Recombined 2 regimes into one program.

4. Final simplification 74.0%

   \[\leadsto \begin{array}{l} \mathbf{if}\;\frac{\left(y \cdot y + x \cdot x\right) - z \cdot z}{2 \cdot y} \leq 0:\\ \;\;\;\;\mathsf{fma}\left(-z, \frac{z}{y}, y\right) \cdot 0.5\\ \mathbf{elif}\;\frac{\left(y \cdot y + x \cdot x\right) - z \cdot z}{2 \cdot y} \leq \infty:\\ \;\;\;\;\mathsf{fma}\left(\frac{x}{y}, x, y\right) \cdot 0.5\\ \mathbf{else}:\\ \;\;\;\;\mathsf{fma}\left(-z, \frac{z}{y}, y\right) \cdot 0.5\\ \end{array} \]
5. Add Preprocessing

Alternative 5: 64.5% accurate, 0.6× speedup

Math

\[\begin{array}{l} z_m = \left|z\right| \\ x_m = \left|x\right| \\ \begin{array}{l} \mathbf{if}\;\frac{\left(y \cdot y + x\_m \cdot x\_m\right) - z\_m \cdot z\_m}{2 \cdot y} \leq -1 \cdot 10^{-50}:\\ \;\;\;\;\left(y - \frac{z\_m \cdot z\_m}{y}\right) \cdot 0.5\\ \mathbf{else}:\\ \;\;\;\;\mathsf{fma}\left(\frac{x\_m}{y}, x\_m, y\right) \cdot 0.5\\ \end{array} \end{array} \]

FPCore

z_m = (fabs.f64 z)
x_m = (fabs.f64 x)
(FPCore (x_m y z_m)
 :precision binary64
 (if (<= (/ (- (+ (* y y) (* x_m x_m)) (* z_m z_m)) (* 2.0 y)) -1e-50)
   (* (- y (/ (* z_m z_m) y)) 0.5)
   (* (fma (/ x_m y) x_m y) 0.5)))

C

z_m = fabs(z);
x_m = fabs(x);
double code(double x_m, double y, double z_m) {
	double tmp;
	if (((((y * y) + (x_m * x_m)) - (z_m * z_m)) / (2.0 * y)) <= -1e-50) {
		tmp = (y - ((z_m * z_m) / y)) * 0.5;
	} else {
		tmp = fma((x_m / y), x_m, y) * 0.5;
	}
	return tmp;
}

Julia

z_m = abs(z)
x_m = abs(x)
function code(x_m, y, z_m)
	tmp = 0.0
	if (Float64(Float64(Float64(Float64(y * y) + Float64(x_m * x_m)) - Float64(z_m * z_m)) / Float64(2.0 * y)) <= -1e-50)
		tmp = Float64(Float64(y - Float64(Float64(z_m * z_m) / y)) * 0.5);
	else
		tmp = Float64(fma(Float64(x_m / y), x_m, y) * 0.5);
	end
	return tmp
end

Wolfram

z_m = N[Abs[z], $MachinePrecision]
x_m = N[Abs[x], $MachinePrecision]
code[x$95$m_, y_, z$95$m_] := If[LessEqual[N[(N[(N[(N[(y * y), $MachinePrecision] + N[(x$95$m * x$95$m), $MachinePrecision]), $MachinePrecision] - N[(z$95$m * z$95$m), $MachinePrecision]), $MachinePrecision] / N[(2.0 * y), $MachinePrecision]), $MachinePrecision], -1e-50], N[(N[(y - N[(N[(z$95$m * z$95$m), $MachinePrecision] / y), $MachinePrecision]), $MachinePrecision] * 0.5), $MachinePrecision], N[(N[(N[(x$95$m / y), $MachinePrecision] * x$95$m + y), $MachinePrecision] * 0.5), $MachinePrecision]]

Derivation

1. Split input into 2 regimes

2. if (/.f64 (-.f64 (+.f64 (*.f64 x x) (*.f64 y y)) (*.f64 z z)) (*.f64 y #s(literal 2 binary64))) < -1.00000000000000001e-50

   1. Initial program 77.7%

      \[\frac{\left(x \cdot x + y \cdot y\right) - z \cdot z}{y \cdot 2} \]
   2. Add Preprocessing

   3. Taylor expanded in x around 0

      \[\leadsto \color{blue}{\frac{1}{2} \cdot \frac{{y}^{2} - {z}^{2}}{y}} \]
   4. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{\frac{{y}^{2} - {z}^{2}}{y} \cdot \frac{1}{2}} \]

      2. div-sub N/A

         \[\leadsto \color{blue}{\left(\frac{{y}^{2}}{y} - \frac{{z}^{2}}{y}\right)} \cdot \frac{1}{2} \]

      3. unpow2 N/A

         \[\leadsto \left(\frac{\color{blue}{y \cdot y}}{y} - \frac{{z}^{2}}{y}\right) \cdot \frac{1}{2} \]

      4. associate-/l* N/A

         \[\leadsto \left(\color{blue}{y \cdot \frac{y}{y}} - \frac{{z}^{2}}{y}\right) \cdot \frac{1}{2} \]

      5. *-inverses N/A

         \[\leadsto \left(y \cdot \color{blue}{1} - \frac{{z}^{2}}{y}\right) \cdot \frac{1}{2} \]

      6. *-rgt-identity N/A

         \[\leadsto \left(\color{blue}{y} - \frac{{z}^{2}}{y}\right) \cdot \frac{1}{2} \]

      7. lower-*.f64 N/A

         \[\leadsto \color{blue}{\left(y - \frac{{z}^{2}}{y}\right) \cdot \frac{1}{2}} \]

      8. lower--.f64 N/A

         \[\leadsto \color{blue}{\left(y - \frac{{z}^{2}}{y}\right)} \cdot \frac{1}{2} \]

      9. lower-/.f64 N/A

         \[\leadsto \left(y - \color{blue}{\frac{{z}^{2}}{y}}\right) \cdot \frac{1}{2} \]

      10. unpow2 N/A

          \[\leadsto \left(y - \frac{\color{blue}{z \cdot z}}{y}\right) \cdot \frac{1}{2} \]

      11. lower-*.f64 71.4

          \[\leadsto \left(y - \frac{\color{blue}{z \cdot z}}{y}\right) \cdot 0.5 \]

   5. Applied rewrites 71.4%

      \[\leadsto \color{blue}{\left(y - \frac{z \cdot z}{y}\right) \cdot 0.5} \]

   if -1.00000000000000001e-50 < (/.f64 (-.f64 (+.f64 (*.f64 x x) (*.f64 y y)) (*.f64 z z)) (*.f64 y #s(literal 2 binary64)))

   1. Initial program 64.9%

      \[\frac{\left(x \cdot x + y \cdot y\right) - z \cdot z}{y \cdot 2} \]
   2. Add Preprocessing

   3. Taylor expanded in y around inf

      \[\leadsto \color{blue}{\frac{1}{2} \cdot y} \]
   4. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{y \cdot \frac{1}{2}} \]

      2. lower-*.f64 36.1

         \[\leadsto \color{blue}{y \cdot 0.5} \]

   5. Applied rewrites 36.1%

      \[\leadsto \color{blue}{y \cdot 0.5} \]

   6. Taylor expanded in x around 0

      \[\leadsto \color{blue}{\frac{1}{2} \cdot \frac{{y}^{2} - {z}^{2}}{y}} \]
   7. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{\frac{{y}^{2} - {z}^{2}}{y} \cdot \frac{1}{2}} \]

      2. unpow2 N/A

         \[\leadsto \frac{\color{blue}{y \cdot y} - {z}^{2}}{y} \cdot \frac{1}{2} \]

      3. unpow2 N/A

         \[\leadsto \frac{y \cdot y - \color{blue}{z \cdot z}}{y} \cdot \frac{1}{2} \]

      4. difference-of-squares N/A

         \[\leadsto \frac{\color{blue}{\left(y + z\right) \cdot \left(y - z\right)}}{y} \cdot \frac{1}{2} \]

      5. lower-*.f64 N/A

         \[\leadsto \color{blue}{\frac{\left(y + z\right) \cdot \left(y - z\right)}{y} \cdot \frac{1}{2}} \]

   8. Applied rewrites 64.5%

      \[\leadsto \color{blue}{\mathsf{fma}\left(-z, \frac{z}{y}, y\right) \cdot 0.5} \]

   9. Taylor expanded in x around 0

      \[\leadsto \color{blue}{\frac{1}{2} \cdot \frac{{y}^{2} - {z}^{2}}{y} + \frac{1}{2} \cdot \frac{{x}^{2}}{y}} \]
   10. Step-by-step derivation

       1. distribute-lft-out N/A

          \[\leadsto \color{blue}{\frac{1}{2} \cdot \left(\frac{{y}^{2} - {z}^{2}}{y} + \frac{{x}^{2}}{y}\right)} \]

       2. *-commutative N/A

          \[\leadsto \color{blue}{\left(\frac{{y}^{2} - {z}^{2}}{y} + \frac{{x}^{2}}{y}\right) \cdot \frac{1}{2}} \]

       3. lower-*.f64 N/A

          \[\leadsto \color{blue}{\left(\frac{{y}^{2} - {z}^{2}}{y} + \frac{{x}^{2}}{y}\right) \cdot \frac{1}{2}} \]

   11. Applied rewrites 100.0%

       \[\leadsto \color{blue}{\left(y + \frac{x - z}{y} \cdot \left(z + x\right)\right) \cdot 0.5} \]

   12. Taylor expanded in z around 0

       \[\leadsto \left(y + \frac{{x}^{2}}{y}\right) \cdot \frac{1}{2} \]
   13. Step-by-step derivation

   14. Applied rewrites 71.1%

       \[\leadsto \mathsf{fma}\left(\frac{x}{y}, x, y\right) \cdot 0.5 \]
3. Recombined 2 regimes into one program.

4. Final simplification 71.2%

   \[\leadsto \begin{array}{l} \mathbf{if}\;\frac{\left(y \cdot y + x \cdot x\right) - z \cdot z}{2 \cdot y} \leq -1 \cdot 10^{-50}:\\ \;\;\;\;\left(y - \frac{z \cdot z}{y}\right) \cdot 0.5\\ \mathbf{else}:\\ \;\;\;\;\mathsf{fma}\left(\frac{x}{y}, x, y\right) \cdot 0.5\\ \end{array} \]
5. Add Preprocessing

Alternative 6: 50.4% accurate, 0.6× speedup

Math

\[\begin{array}{l} z_m = \left|z\right| \\ x_m = \left|x\right| \\ \begin{array}{l} \mathbf{if}\;\frac{\left(y \cdot y + x\_m \cdot x\_m\right) - z\_m \cdot z\_m}{2 \cdot y} \leq -1 \cdot 10^{-50}:\\ \;\;\;\;-0.5 \cdot \frac{z\_m \cdot z\_m}{y}\\ \mathbf{else}:\\ \;\;\;\;\mathsf{fma}\left(\frac{x\_m}{y}, x\_m, y\right) \cdot 0.5\\ \end{array} \end{array} \]

FPCore

z_m = (fabs.f64 z)
x_m = (fabs.f64 x)
(FPCore (x_m y z_m)
 :precision binary64
 (if (<= (/ (- (+ (* y y) (* x_m x_m)) (* z_m z_m)) (* 2.0 y)) -1e-50)
   (* -0.5 (/ (* z_m z_m) y))
   (* (fma (/ x_m y) x_m y) 0.5)))

C

z_m = fabs(z);
x_m = fabs(x);
double code(double x_m, double y, double z_m) {
	double tmp;
	if (((((y * y) + (x_m * x_m)) - (z_m * z_m)) / (2.0 * y)) <= -1e-50) {
		tmp = -0.5 * ((z_m * z_m) / y);
	} else {
		tmp = fma((x_m / y), x_m, y) * 0.5;
	}
	return tmp;
}

Julia

z_m = abs(z)
x_m = abs(x)
function code(x_m, y, z_m)
	tmp = 0.0
	if (Float64(Float64(Float64(Float64(y * y) + Float64(x_m * x_m)) - Float64(z_m * z_m)) / Float64(2.0 * y)) <= -1e-50)
		tmp = Float64(-0.5 * Float64(Float64(z_m * z_m) / y));
	else
		tmp = Float64(fma(Float64(x_m / y), x_m, y) * 0.5);
	end
	return tmp
end

Wolfram

z_m = N[Abs[z], $MachinePrecision]
x_m = N[Abs[x], $MachinePrecision]
code[x$95$m_, y_, z$95$m_] := If[LessEqual[N[(N[(N[(N[(y * y), $MachinePrecision] + N[(x$95$m * x$95$m), $MachinePrecision]), $MachinePrecision] - N[(z$95$m * z$95$m), $MachinePrecision]), $MachinePrecision] / N[(2.0 * y), $MachinePrecision]), $MachinePrecision], -1e-50], N[(-0.5 * N[(N[(z$95$m * z$95$m), $MachinePrecision] / y), $MachinePrecision]), $MachinePrecision], N[(N[(N[(x$95$m / y), $MachinePrecision] * x$95$m + y), $MachinePrecision] * 0.5), $MachinePrecision]]

Derivation

1. Split input into 2 regimes

2. if (/.f64 (-.f64 (+.f64 (*.f64 x x) (*.f64 y y)) (*.f64 z z)) (*.f64 y #s(literal 2 binary64))) < -1.00000000000000001e-50

   1. Initial program 77.7%

      \[\frac{\left(x \cdot x + y \cdot y\right) - z \cdot z}{y \cdot 2} \]
   2. Add Preprocessing

   3. Taylor expanded in z around inf

      \[\leadsto \color{blue}{\frac{-1}{2} \cdot \frac{{z}^{2}}{y}} \]
   4. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{\frac{{z}^{2}}{y} \cdot \frac{-1}{2}} \]

      2. lower-*.f64 N/A

         \[\leadsto \color{blue}{\frac{{z}^{2}}{y} \cdot \frac{-1}{2}} \]

      3. lower-/.f64 N/A

         \[\leadsto \color{blue}{\frac{{z}^{2}}{y}} \cdot \frac{-1}{2} \]

      4. unpow2 N/A

         \[\leadsto \frac{\color{blue}{z \cdot z}}{y} \cdot \frac{-1}{2} \]

      5. lower-*.f64 34.0

         \[\leadsto \frac{\color{blue}{z \cdot z}}{y} \cdot -0.5 \]

   5. Applied rewrites 34.0%

      \[\leadsto \color{blue}{\frac{z \cdot z}{y} \cdot -0.5} \]

   if -1.00000000000000001e-50 < (/.f64 (-.f64 (+.f64 (*.f64 x x) (*.f64 y y)) (*.f64 z z)) (*.f64 y #s(literal 2 binary64)))

   1. Initial program 64.9%

      \[\frac{\left(x \cdot x + y \cdot y\right) - z \cdot z}{y \cdot 2} \]
   2. Add Preprocessing

   3. Taylor expanded in y around inf

      \[\leadsto \color{blue}{\frac{1}{2} \cdot y} \]
   4. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{y \cdot \frac{1}{2}} \]

      2. lower-*.f64 36.1

         \[\leadsto \color{blue}{y \cdot 0.5} \]

   5. Applied rewrites 36.1%

      \[\leadsto \color{blue}{y \cdot 0.5} \]

   6. Taylor expanded in x around 0

      \[\leadsto \color{blue}{\frac{1}{2} \cdot \frac{{y}^{2} - {z}^{2}}{y}} \]
   7. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{\frac{{y}^{2} - {z}^{2}}{y} \cdot \frac{1}{2}} \]

      2. unpow2 N/A

         \[\leadsto \frac{\color{blue}{y \cdot y} - {z}^{2}}{y} \cdot \frac{1}{2} \]

      3. unpow2 N/A

         \[\leadsto \frac{y \cdot y - \color{blue}{z \cdot z}}{y} \cdot \frac{1}{2} \]

      4. difference-of-squares N/A

         \[\leadsto \frac{\color{blue}{\left(y + z\right) \cdot \left(y - z\right)}}{y} \cdot \frac{1}{2} \]

      5. lower-*.f64 N/A

         \[\leadsto \color{blue}{\frac{\left(y + z\right) \cdot \left(y - z\right)}{y} \cdot \frac{1}{2}} \]

   8. Applied rewrites 64.5%

      \[\leadsto \color{blue}{\mathsf{fma}\left(-z, \frac{z}{y}, y\right) \cdot 0.5} \]

   9. Taylor expanded in x around 0

      \[\leadsto \color{blue}{\frac{1}{2} \cdot \frac{{y}^{2} - {z}^{2}}{y} + \frac{1}{2} \cdot \frac{{x}^{2}}{y}} \]
   10. Step-by-step derivation

       1. distribute-lft-out N/A

          \[\leadsto \color{blue}{\frac{1}{2} \cdot \left(\frac{{y}^{2} - {z}^{2}}{y} + \frac{{x}^{2}}{y}\right)} \]

       2. *-commutative N/A

          \[\leadsto \color{blue}{\left(\frac{{y}^{2} - {z}^{2}}{y} + \frac{{x}^{2}}{y}\right) \cdot \frac{1}{2}} \]

       3. lower-*.f64 N/A

          \[\leadsto \color{blue}{\left(\frac{{y}^{2} - {z}^{2}}{y} + \frac{{x}^{2}}{y}\right) \cdot \frac{1}{2}} \]

   11. Applied rewrites 100.0%

       \[\leadsto \color{blue}{\left(y + \frac{x - z}{y} \cdot \left(z + x\right)\right) \cdot 0.5} \]

   12. Taylor expanded in z around 0

       \[\leadsto \left(y + \frac{{x}^{2}}{y}\right) \cdot \frac{1}{2} \]
   13. Step-by-step derivation

   14. Applied rewrites 71.1%

       \[\leadsto \mathsf{fma}\left(\frac{x}{y}, x, y\right) \cdot 0.5 \]
3. Recombined 2 regimes into one program.

4. Final simplification 53.4%

   \[\leadsto \begin{array}{l} \mathbf{if}\;\frac{\left(y \cdot y + x \cdot x\right) - z \cdot z}{2 \cdot y} \leq -1 \cdot 10^{-50}:\\ \;\;\;\;-0.5 \cdot \frac{z \cdot z}{y}\\ \mathbf{else}:\\ \;\;\;\;\mathsf{fma}\left(\frac{x}{y}, x, y\right) \cdot 0.5\\ \end{array} \]
5. Add Preprocessing

Alternative 7: 52.3% accurate, 1.4× speedup

Math

\[\begin{array}{l} z_m = \left|z\right| \\ x_m = \left|x\right| \\ \begin{array}{l} \mathbf{if}\;x\_m \leq 6.8 \cdot 10^{+108}:\\ \;\;\;\;0.5 \cdot y\\ \mathbf{else}:\\ \;\;\;\;\left(\frac{x\_m}{y} \cdot x\_m\right) \cdot 0.5\\ \end{array} \end{array} \]

FPCore

z_m = (fabs.f64 z)
x_m = (fabs.f64 x)
(FPCore (x_m y z_m)
 :precision binary64
 (if (<= x_m 6.8e+108) (* 0.5 y) (* (* (/ x_m y) x_m) 0.5)))

C

z_m = fabs(z);
x_m = fabs(x);
double code(double x_m, double y, double z_m) {
	double tmp;
	if (x_m <= 6.8e+108) {
		tmp = 0.5 * y;
	} else {
		tmp = ((x_m / y) * x_m) * 0.5;
	}
	return tmp;
}

Fortran

z_m = abs(z)
x_m = abs(x)
real(8) function code(x_m, y, z_m)
    real(8), intent (in) :: x_m
    real(8), intent (in) :: y
    real(8), intent (in) :: z_m
    real(8) :: tmp
    if (x_m <= 6.8d+108) then
        tmp = 0.5d0 * y
    else
        tmp = ((x_m / y) * x_m) * 0.5d0
    end if
    code = tmp
end function

Java

z_m = Math.abs(z);
x_m = Math.abs(x);
public static double code(double x_m, double y, double z_m) {
	double tmp;
	if (x_m <= 6.8e+108) {
		tmp = 0.5 * y;
	} else {
		tmp = ((x_m / y) * x_m) * 0.5;
	}
	return tmp;
}

Python

z_m = math.fabs(z)
x_m = math.fabs(x)
def code(x_m, y, z_m):
	tmp = 0
	if x_m <= 6.8e+108:
		tmp = 0.5 * y
	else:
		tmp = ((x_m / y) * x_m) * 0.5
	return tmp

Julia

z_m = abs(z)
x_m = abs(x)
function code(x_m, y, z_m)
	tmp = 0.0
	if (x_m <= 6.8e+108)
		tmp = Float64(0.5 * y);
	else
		tmp = Float64(Float64(Float64(x_m / y) * x_m) * 0.5);
	end
	return tmp
end

MATLAB

z_m = abs(z);
x_m = abs(x);
function tmp_2 = code(x_m, y, z_m)
	tmp = 0.0;
	if (x_m <= 6.8e+108)
		tmp = 0.5 * y;
	else
		tmp = ((x_m / y) * x_m) * 0.5;
	end
	tmp_2 = tmp;
end

Wolfram

z_m = N[Abs[z], $MachinePrecision]
x_m = N[Abs[x], $MachinePrecision]
code[x$95$m_, y_, z$95$m_] := If[LessEqual[x$95$m, 6.8e+108], N[(0.5 * y), $MachinePrecision], N[(N[(N[(x$95$m / y), $MachinePrecision] * x$95$m), $MachinePrecision] * 0.5), $MachinePrecision]]

Derivation

1. Split input into 2 regimes

2. if x < 6.79999999999999992e108

   1. Initial program 72.4%

      \[\frac{\left(x \cdot x + y \cdot y\right) - z \cdot z}{y \cdot 2} \]
   2. Add Preprocessing

   3. Taylor expanded in y around inf

      \[\leadsto \color{blue}{\frac{1}{2} \cdot y} \]
   4. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{y \cdot \frac{1}{2}} \]

      2. lower-*.f64 39.1

         \[\leadsto \color{blue}{y \cdot 0.5} \]

   5. Applied rewrites 39.1%

      \[\leadsto \color{blue}{y \cdot 0.5} \]

   if 6.79999999999999992e108 < x

   1. Initial program 62.2%

      \[\frac{\left(x \cdot x + y \cdot y\right) - z \cdot z}{y \cdot 2} \]
   2. Add Preprocessing

   3. Taylor expanded in y around inf

      \[\leadsto \color{blue}{\frac{1}{2} \cdot y} \]
   4. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{y \cdot \frac{1}{2}} \]

      2. lower-*.f64 25.1

         \[\leadsto \color{blue}{y \cdot 0.5} \]

   5. Applied rewrites 25.1%

      \[\leadsto \color{blue}{y \cdot 0.5} \]

   6. Taylor expanded in x around inf

      \[\leadsto \color{blue}{\frac{1}{2} \cdot \frac{{x}^{2}}{y}} \]
   7. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{\frac{{x}^{2}}{y} \cdot \frac{1}{2}} \]

      2. lower-*.f64 N/A

         \[\leadsto \color{blue}{\frac{{x}^{2}}{y} \cdot \frac{1}{2}} \]

      3. unpow2 N/A

         \[\leadsto \frac{\color{blue}{x \cdot x}}{y} \cdot \frac{1}{2} \]

      4. associate-*l/ N/A

         \[\leadsto \color{blue}{\left(\frac{x}{y} \cdot x\right)} \cdot \frac{1}{2} \]

      5. lower-*.f64 N/A

         \[\leadsto \color{blue}{\left(\frac{x}{y} \cdot x\right)} \cdot \frac{1}{2} \]

      6. lower-/.f64 72.7

         \[\leadsto \left(\color{blue}{\frac{x}{y}} \cdot x\right) \cdot 0.5 \]

   8. Applied rewrites 72.7%

      \[\leadsto \color{blue}{\left(\frac{x}{y} \cdot x\right) \cdot 0.5} \]
3. Recombined 2 regimes into one program.

4. Final simplification 43.7%

   \[\leadsto \begin{array}{l} \mathbf{if}\;x \leq 6.8 \cdot 10^{+108}:\\ \;\;\;\;0.5 \cdot y\\ \mathbf{else}:\\ \;\;\;\;\left(\frac{x}{y} \cdot x\right) \cdot 0.5\\ \end{array} \]
5. Add Preprocessing

Alternative 8: 34.4% accurate, 6.3× speedup

Math

\[\begin{array}{l} z_m = \left|z\right| \\ x_m = \left|x\right| \\ 0.5 \cdot y \end{array} \]

FPCore

z_m = (fabs.f64 z)
x_m = (fabs.f64 x)
(FPCore (x_m y z_m) :precision binary64 (* 0.5 y))

C

z_m = fabs(z);
x_m = fabs(x);
double code(double x_m, double y, double z_m) {
	return 0.5 * y;
}

Fortran

z_m = abs(z)
x_m = abs(x)
real(8) function code(x_m, y, z_m)
    real(8), intent (in) :: x_m
    real(8), intent (in) :: y
    real(8), intent (in) :: z_m
    code = 0.5d0 * y
end function

Java

z_m = Math.abs(z);
x_m = Math.abs(x);
public static double code(double x_m, double y, double z_m) {
	return 0.5 * y;
}

Python

z_m = math.fabs(z)
x_m = math.fabs(x)
def code(x_m, y, z_m):
	return 0.5 * y

Julia

z_m = abs(z)
x_m = abs(x)
function code(x_m, y, z_m)
	return Float64(0.5 * y)
end

MATLAB

z_m = abs(z);
x_m = abs(x);
function tmp = code(x_m, y, z_m)
	tmp = 0.5 * y;
end

Wolfram

z_m = N[Abs[z], $MachinePrecision]
x_m = N[Abs[x], $MachinePrecision]
code[x$95$m_, y_, z$95$m_] := N[(0.5 * y), $MachinePrecision]

Derivation

1. Initial program 71.0%

   \[\frac{\left(x \cdot x + y \cdot y\right) - z \cdot z}{y \cdot 2} \]
2. Add Preprocessing

3. Taylor expanded in y around inf

   \[\leadsto \color{blue}{\frac{1}{2} \cdot y} \]
4. Step-by-step derivation

   1. *-commutative N/A

      \[\leadsto \color{blue}{y \cdot \frac{1}{2}} \]

   2. lower-*.f64 37.2

      \[\leadsto \color{blue}{y \cdot 0.5} \]

5. Applied rewrites 37.2%

   \[\leadsto \color{blue}{y \cdot 0.5} \]

6. Final simplification 37.2%

   \[\leadsto 0.5 \cdot y \]
7. Add Preprocessing

Developer Target 1: 99.9% accurate, 1.1× speedup

Math

\[\begin{array}{l} \\ y \cdot 0.5 - \left(\frac{0.5}{y} \cdot \left(z + x\right)\right) \cdot \left(z - x\right) \end{array} \]

FPCore

(FPCore (x y z)
 :precision binary64
 (- (* y 0.5) (* (* (/ 0.5 y) (+ z x)) (- z x))))

C

double code(double x, double y, double z) {
	return (y * 0.5) - (((0.5 / y) * (z + x)) * (z - x));
}

Fortran

real(8) function code(x, y, z)
    real(8), intent (in) :: x
    real(8), intent (in) :: y
    real(8), intent (in) :: z
    code = (y * 0.5d0) - (((0.5d0 / y) * (z + x)) * (z - x))
end function

Java

public static double code(double x, double y, double z) {
	return (y * 0.5) - (((0.5 / y) * (z + x)) * (z - x));
}

Python

def code(x, y, z):
	return (y * 0.5) - (((0.5 / y) * (z + x)) * (z - x))

Julia

function code(x, y, z)
	return Float64(Float64(y * 0.5) - Float64(Float64(Float64(0.5 / y) * Float64(z + x)) * Float64(z - x)))
end

MATLAB

function tmp = code(x, y, z)
	tmp = (y * 0.5) - (((0.5 / y) * (z + x)) * (z - x));
end

Wolfram

code[x_, y_, z_] := N[(N[(y * 0.5), $MachinePrecision] - N[(N[(N[(0.5 / y), $MachinePrecision] * N[(z + x), $MachinePrecision]), $MachinePrecision] * N[(z - x), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]

Reproduce

herbie shell --seed 2024249 
(FPCore (x y z)
  :name "Diagrams.TwoD.Apollonian:initialConfig from diagrams-contrib-1.3.0.5, A"
  :precision binary64

  :alt
  (! :herbie-platform default (- (* y 1/2) (* (* (/ 1/2 y) (+ z x)) (- z x))))

  (/ (- (+ (* x x) (* y y)) (* z z)) (* y 2.0)))