Kahan p9 Example

Percentage Accurate: 67.4% → 92.1%

Time: 8.8s

Alternatives: 6

Speedup: 0.8×

Specification

\[\left(0 < x \land x < 1\right) \land y < 1\]

Math

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

FPCore

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

C

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

Fortran

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

Java

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

Python

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

Julia

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

MATLAB

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

Wolfram

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

Initial Program: 67.4% accurate, 1.0× speedup

Math

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

FPCore

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

C

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

Fortran

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

Java

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

Python

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

Julia

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

MATLAB

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

Wolfram

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

Alternative 1: 92.1% accurate, 0.8× speedup

Math

\[\begin{array}{l} y_m = \left|y\right| \\ \begin{array}{l} \mathbf{if}\;y\_m \leq 2.6 \cdot 10^{-168}:\\ \;\;\;\;\mathsf{fma}\left(\frac{-2 \cdot y\_m}{x}, \frac{y\_m}{x}, 1\right)\\ \mathbf{elif}\;y\_m \leq 2 \cdot 10^{-39}:\\ \;\;\;\;\frac{\left(x + y\_m\right) \cdot \left(x - y\_m\right)}{\mathsf{fma}\left(y\_m, y\_m, x \cdot x\right)}\\ \mathbf{else}:\\ \;\;\;\;-1\\ \end{array} \end{array} \]

FPCore

y_m = (fabs.f64 y)
(FPCore (x y_m)
 :precision binary64
 (if (<= y_m 2.6e-168)
   (fma (/ (* -2.0 y_m) x) (/ y_m x) 1.0)
   (if (<= y_m 2e-39) (/ (* (+ x y_m) (- x y_m)) (fma y_m y_m (* x x))) -1.0)))

C

y_m = fabs(y);
double code(double x, double y_m) {
	double tmp;
	if (y_m <= 2.6e-168) {
		tmp = fma(((-2.0 * y_m) / x), (y_m / x), 1.0);
	} else if (y_m <= 2e-39) {
		tmp = ((x + y_m) * (x - y_m)) / fma(y_m, y_m, (x * x));
	} else {
		tmp = -1.0;
	}
	return tmp;
}

Julia

y_m = abs(y)
function code(x, y_m)
	tmp = 0.0
	if (y_m <= 2.6e-168)
		tmp = fma(Float64(Float64(-2.0 * y_m) / x), Float64(y_m / x), 1.0);
	elseif (y_m <= 2e-39)
		tmp = Float64(Float64(Float64(x + y_m) * Float64(x - y_m)) / fma(y_m, y_m, Float64(x * x)));
	else
		tmp = -1.0;
	end
	return tmp
end

Wolfram

y_m = N[Abs[y], $MachinePrecision]
code[x_, y$95$m_] := If[LessEqual[y$95$m, 2.6e-168], N[(N[(N[(-2.0 * y$95$m), $MachinePrecision] / x), $MachinePrecision] * N[(y$95$m / x), $MachinePrecision] + 1.0), $MachinePrecision], If[LessEqual[y$95$m, 2e-39], N[(N[(N[(x + y$95$m), $MachinePrecision] * N[(x - y$95$m), $MachinePrecision]), $MachinePrecision] / N[(y$95$m * y$95$m + N[(x * x), $MachinePrecision]), $MachinePrecision]), $MachinePrecision], -1.0]]

Derivation

1. Split input into 3 regimes

2. if y < 2.6000000000000001e-168

   1. Initial program 60.8%

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

   3. Taylor expanded in y around 0

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

      1. +-commutative N/A

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

      2. *-commutative N/A

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

      3. associate-*l/ N/A

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

      4. *-commutative N/A

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

      5. unpow2 N/A

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

      6. associate-*r* N/A

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

      7. associate-/l* N/A

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

      8. lower-fma.f64 N/A

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

      9. lower-*.f64 N/A

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

      10. lower-/.f64 N/A

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

      11. unpow2 N/A

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

      12. lower-*.f64 28.0

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

   5. Applied rewrites 28.0%

      \[\leadsto \color{blue}{\mathsf{fma}\left(-2 \cdot y, \frac{y}{x \cdot x}, 1\right)} \]
   6. Step-by-step derivation

   7. Applied rewrites 39.2%

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

   if 2.6000000000000001e-168 < y < 1.99999999999999986e-39

   1. Initial program 100.0%

      \[\frac{\left(x - y\right) \cdot \left(x + y\right)}{x \cdot x + y \cdot y} \]
   2. Add Preprocessing
   3. Step-by-step derivation

      1. lift-+.f64 N/A

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

      2. +-commutative N/A

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

      3. lift-*.f64 N/A

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

      4. lower-fma.f64 100.0

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

   4. Applied rewrites 100.0%

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

   if 1.99999999999999986e-39 < y

   1. Initial program 100.0%

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

   3. Taylor expanded in y around inf

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

   5. Applied rewrites 100.0%

      \[\leadsto \color{blue}{-1} \]
3. Recombined 3 regimes into one program.

4. Final simplification 50.6%

   \[\leadsto \begin{array}{l} \mathbf{if}\;y \leq 2.6 \cdot 10^{-168}:\\ \;\;\;\;\mathsf{fma}\left(\frac{-2 \cdot y}{x}, \frac{y}{x}, 1\right)\\ \mathbf{elif}\;y \leq 2 \cdot 10^{-39}:\\ \;\;\;\;\frac{\left(x + y\right) \cdot \left(x - y\right)}{\mathsf{fma}\left(y, y, x \cdot x\right)}\\ \mathbf{else}:\\ \;\;\;\;-1\\ \end{array} \]
5. Add Preprocessing

Alternative 2: 92.1% accurate, 0.3× speedup

Math

\[\begin{array}{l} y_m = \left|y\right| \\ \begin{array}{l} t_0 := \frac{\left(x + y\_m\right) \cdot \left(x - y\_m\right)}{y\_m \cdot y\_m + x \cdot x}\\ \mathbf{if}\;t\_0 \leq -0.5:\\ \;\;\;\;\mathsf{fma}\left(\frac{x}{y\_m \cdot y\_m}, 2 \cdot x, -1\right)\\ \mathbf{elif}\;t\_0 \leq 2:\\ \;\;\;\;\mathsf{fma}\left(-2 \cdot y\_m, \frac{y\_m}{x \cdot x}, 1\right)\\ \mathbf{else}:\\ \;\;\;\;\mathsf{fma}\left(\frac{2 \cdot x}{y\_m}, \frac{x}{y\_m}, -1\right)\\ \end{array} \end{array} \]

FPCore

y_m = (fabs.f64 y)
(FPCore (x y_m)
 :precision binary64
 (let* ((t_0 (/ (* (+ x y_m) (- x y_m)) (+ (* y_m y_m) (* x x)))))
   (if (<= t_0 -0.5)
     (fma (/ x (* y_m y_m)) (* 2.0 x) -1.0)
     (if (<= t_0 2.0)
       (fma (* -2.0 y_m) (/ y_m (* x x)) 1.0)
       (fma (/ (* 2.0 x) y_m) (/ x y_m) -1.0)))))

C

y_m = fabs(y);
double code(double x, double y_m) {
	double t_0 = ((x + y_m) * (x - y_m)) / ((y_m * y_m) + (x * x));
	double tmp;
	if (t_0 <= -0.5) {
		tmp = fma((x / (y_m * y_m)), (2.0 * x), -1.0);
	} else if (t_0 <= 2.0) {
		tmp = fma((-2.0 * y_m), (y_m / (x * x)), 1.0);
	} else {
		tmp = fma(((2.0 * x) / y_m), (x / y_m), -1.0);
	}
	return tmp;
}

Julia

y_m = abs(y)
function code(x, y_m)
	t_0 = Float64(Float64(Float64(x + y_m) * Float64(x - y_m)) / Float64(Float64(y_m * y_m) + Float64(x * x)))
	tmp = 0.0
	if (t_0 <= -0.5)
		tmp = fma(Float64(x / Float64(y_m * y_m)), Float64(2.0 * x), -1.0);
	elseif (t_0 <= 2.0)
		tmp = fma(Float64(-2.0 * y_m), Float64(y_m / Float64(x * x)), 1.0);
	else
		tmp = fma(Float64(Float64(2.0 * x) / y_m), Float64(x / y_m), -1.0);
	end
	return tmp
end

Wolfram

y_m = N[Abs[y], $MachinePrecision]
code[x_, y$95$m_] := Block[{t$95$0 = N[(N[(N[(x + y$95$m), $MachinePrecision] * N[(x - y$95$m), $MachinePrecision]), $MachinePrecision] / N[(N[(y$95$m * y$95$m), $MachinePrecision] + N[(x * x), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]}, If[LessEqual[t$95$0, -0.5], N[(N[(x / N[(y$95$m * y$95$m), $MachinePrecision]), $MachinePrecision] * N[(2.0 * x), $MachinePrecision] + -1.0), $MachinePrecision], If[LessEqual[t$95$0, 2.0], N[(N[(-2.0 * y$95$m), $MachinePrecision] * N[(y$95$m / N[(x * x), $MachinePrecision]), $MachinePrecision] + 1.0), $MachinePrecision], N[(N[(N[(2.0 * x), $MachinePrecision] / y$95$m), $MachinePrecision] * N[(x / y$95$m), $MachinePrecision] + -1.0), $MachinePrecision]]]]

Derivation

1. Split input into 3 regimes

2. if (/.f64 (*.f64 (-.f64 x y) (+.f64 x y)) (+.f64 (*.f64 x x) (*.f64 y y))) < -0.5

   1. Initial program 99.6%

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

   3. Taylor expanded in y around 0

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

      1. +-commutative N/A

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

      2. *-commutative N/A

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

      3. associate-*l/ N/A

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

      4. *-commutative N/A

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

      5. unpow2 N/A

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

      6. associate-*r* N/A

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

      7. associate-/l* N/A

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

      8. lower-fma.f64 N/A

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

      9. lower-*.f64 N/A

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

      10. lower-/.f64 N/A

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

      11. unpow2 N/A

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

      12. lower-*.f64 3.7

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

   5. Applied rewrites 3.7%

      \[\leadsto \color{blue}{\mathsf{fma}\left(-2 \cdot y, \frac{y}{x \cdot x}, 1\right)} \]
   6. Step-by-step derivation

   7. Applied rewrites 4.0%

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

   8. Taylor expanded in y around inf

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

      1. associate-+r+ N/A

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

      2. distribute-lft1-in N/A

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

      3. metadata-eval N/A

         \[\leadsto \left(\color{blue}{0} \cdot \frac{x}{y} + \frac{{x}^{2}}{{y}^{2}}\right) - \left(1 + -1 \cdot \frac{{x}^{2}}{{y}^{2}}\right) \]

      4. mul0-lft N/A

         \[\leadsto \left(\color{blue}{0} + \frac{{x}^{2}}{{y}^{2}}\right) - \left(1 + -1 \cdot \frac{{x}^{2}}{{y}^{2}}\right) \]

      5. +-lft-identity N/A

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

      6. +-commutative N/A

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

      7. associate--r+ N/A

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

      8. associate-*r/ N/A

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

      9. div-sub N/A

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

      10. cancel-sign-sub-inv N/A

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

      11. metadata-eval N/A

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

      12. distribute-rgt1-in N/A

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

      13. metadata-eval N/A

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

      14. associate-*r/ N/A

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

   10. Applied rewrites 100.0%

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

   if -0.5 < (/.f64 (*.f64 (-.f64 x y) (+.f64 x y)) (+.f64 (*.f64 x x) (*.f64 y y))) < 2

   1. Initial program 100.0%

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

   3. Taylor expanded in y around 0

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

      1. +-commutative N/A

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

      2. *-commutative N/A

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

      3. associate-*l/ N/A

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

      4. *-commutative N/A

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

      5. unpow2 N/A

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

      6. associate-*r* N/A

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

      7. associate-/l* N/A

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

      8. lower-fma.f64 N/A

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

      9. lower-*.f64 N/A

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

      10. lower-/.f64 N/A

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

      11. unpow2 N/A

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

      12. lower-*.f64 99.3

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

   5. Applied rewrites 99.3%

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

   if 2 < (/.f64 (*.f64 (-.f64 x y) (+.f64 x y)) (+.f64 (*.f64 x x) (*.f64 y y)))

   1. Initial program 0.0%

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

   3. Taylor expanded in y around inf

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

      1. associate-+r+ N/A

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

      2. distribute-lft1-in N/A

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

      3. metadata-eval N/A

         \[\leadsto \left(\color{blue}{0} \cdot \frac{x}{y} + \frac{{x}^{2}}{{y}^{2}}\right) - \left(1 + -1 \cdot \frac{{x}^{2}}{{y}^{2}}\right) \]

      4. mul0-lft N/A

         \[\leadsto \left(\color{blue}{0} + \frac{{x}^{2}}{{y}^{2}}\right) - \left(1 + -1 \cdot \frac{{x}^{2}}{{y}^{2}}\right) \]

      5. associate--l+ N/A

         \[\leadsto \color{blue}{0 + \left(\frac{{x}^{2}}{{y}^{2}} - \left(1 + -1 \cdot \frac{{x}^{2}}{{y}^{2}}\right)\right)} \]

      6. +-lft-identity N/A

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

      7. +-commutative N/A

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

      8. associate--r+ N/A

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

   5. Applied rewrites 56.8%

      \[\leadsto \color{blue}{\mathsf{fma}\left(\frac{2}{y \cdot y}, x \cdot x, -1\right)} \]
   6. Step-by-step derivation

   7. Applied rewrites 73.3%

      \[\leadsto \mathsf{fma}\left(\frac{2 \cdot x}{y}, \color{blue}{\frac{x}{y}}, -1\right) \]
3. Recombined 3 regimes into one program.

4. Final simplification 91.3%

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

Alternative 3: 91.5% accurate, 0.3× speedup

Math

\[\begin{array}{l} y_m = \left|y\right| \\ \begin{array}{l} t_0 := \frac{\left(x + y\_m\right) \cdot \left(x - y\_m\right)}{y\_m \cdot y\_m + x \cdot x}\\ \mathbf{if}\;t\_0 \leq -0.5:\\ \;\;\;\;\mathsf{fma}\left(\frac{x}{y\_m \cdot y\_m}, 2 \cdot x, -1\right)\\ \mathbf{elif}\;t\_0 \leq 2:\\ \;\;\;\;\mathsf{fma}\left(-2 \cdot y\_m, \frac{y\_m}{x \cdot x}, 1\right)\\ \mathbf{else}:\\ \;\;\;\;-1\\ \end{array} \end{array} \]

FPCore

y_m = (fabs.f64 y)
(FPCore (x y_m)
 :precision binary64
 (let* ((t_0 (/ (* (+ x y_m) (- x y_m)) (+ (* y_m y_m) (* x x)))))
   (if (<= t_0 -0.5)
     (fma (/ x (* y_m y_m)) (* 2.0 x) -1.0)
     (if (<= t_0 2.0) (fma (* -2.0 y_m) (/ y_m (* x x)) 1.0) -1.0))))

C

y_m = fabs(y);
double code(double x, double y_m) {
	double t_0 = ((x + y_m) * (x - y_m)) / ((y_m * y_m) + (x * x));
	double tmp;
	if (t_0 <= -0.5) {
		tmp = fma((x / (y_m * y_m)), (2.0 * x), -1.0);
	} else if (t_0 <= 2.0) {
		tmp = fma((-2.0 * y_m), (y_m / (x * x)), 1.0);
	} else {
		tmp = -1.0;
	}
	return tmp;
}

Julia

y_m = abs(y)
function code(x, y_m)
	t_0 = Float64(Float64(Float64(x + y_m) * Float64(x - y_m)) / Float64(Float64(y_m * y_m) + Float64(x * x)))
	tmp = 0.0
	if (t_0 <= -0.5)
		tmp = fma(Float64(x / Float64(y_m * y_m)), Float64(2.0 * x), -1.0);
	elseif (t_0 <= 2.0)
		tmp = fma(Float64(-2.0 * y_m), Float64(y_m / Float64(x * x)), 1.0);
	else
		tmp = -1.0;
	end
	return tmp
end

Wolfram

y_m = N[Abs[y], $MachinePrecision]
code[x_, y$95$m_] := Block[{t$95$0 = N[(N[(N[(x + y$95$m), $MachinePrecision] * N[(x - y$95$m), $MachinePrecision]), $MachinePrecision] / N[(N[(y$95$m * y$95$m), $MachinePrecision] + N[(x * x), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]}, If[LessEqual[t$95$0, -0.5], N[(N[(x / N[(y$95$m * y$95$m), $MachinePrecision]), $MachinePrecision] * N[(2.0 * x), $MachinePrecision] + -1.0), $MachinePrecision], If[LessEqual[t$95$0, 2.0], N[(N[(-2.0 * y$95$m), $MachinePrecision] * N[(y$95$m / N[(x * x), $MachinePrecision]), $MachinePrecision] + 1.0), $MachinePrecision], -1.0]]]

Derivation

1. Split input into 3 regimes

2. if (/.f64 (*.f64 (-.f64 x y) (+.f64 x y)) (+.f64 (*.f64 x x) (*.f64 y y))) < -0.5

   1. Initial program 99.6%

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

   3. Taylor expanded in y around 0

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

      1. +-commutative N/A

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

      2. *-commutative N/A

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

      3. associate-*l/ N/A

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

      4. *-commutative N/A

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

      5. unpow2 N/A

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

      6. associate-*r* N/A

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

      7. associate-/l* N/A

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

      8. lower-fma.f64 N/A

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

      9. lower-*.f64 N/A

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

      10. lower-/.f64 N/A

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

      11. unpow2 N/A

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

      12. lower-*.f64 3.7

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

   5. Applied rewrites 3.7%

      \[\leadsto \color{blue}{\mathsf{fma}\left(-2 \cdot y, \frac{y}{x \cdot x}, 1\right)} \]
   6. Step-by-step derivation

   7. Applied rewrites 4.0%

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

   8. Taylor expanded in y around inf

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

      1. associate-+r+ N/A

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

      2. distribute-lft1-in N/A

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

      3. metadata-eval N/A

         \[\leadsto \left(\color{blue}{0} \cdot \frac{x}{y} + \frac{{x}^{2}}{{y}^{2}}\right) - \left(1 + -1 \cdot \frac{{x}^{2}}{{y}^{2}}\right) \]

      4. mul0-lft N/A

         \[\leadsto \left(\color{blue}{0} + \frac{{x}^{2}}{{y}^{2}}\right) - \left(1 + -1 \cdot \frac{{x}^{2}}{{y}^{2}}\right) \]

      5. +-lft-identity N/A

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

      6. +-commutative N/A

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

      7. associate--r+ N/A

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

      8. associate-*r/ N/A

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

      9. div-sub N/A

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

      10. cancel-sign-sub-inv N/A

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

      11. metadata-eval N/A

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

      12. distribute-rgt1-in N/A

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

      13. metadata-eval N/A

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

      14. associate-*r/ N/A

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

   10. Applied rewrites 100.0%

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

   if -0.5 < (/.f64 (*.f64 (-.f64 x y) (+.f64 x y)) (+.f64 (*.f64 x x) (*.f64 y y))) < 2

   1. Initial program 100.0%

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

   3. Taylor expanded in y around 0

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

      1. +-commutative N/A

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

      2. *-commutative N/A

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

      3. associate-*l/ N/A

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

      4. *-commutative N/A

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

      5. unpow2 N/A

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

      6. associate-*r* N/A

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

      7. associate-/l* N/A

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

      8. lower-fma.f64 N/A

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

      9. lower-*.f64 N/A

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

      10. lower-/.f64 N/A

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

      11. unpow2 N/A

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

      12. lower-*.f64 99.3

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

   5. Applied rewrites 99.3%

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

   if 2 < (/.f64 (*.f64 (-.f64 x y) (+.f64 x y)) (+.f64 (*.f64 x x) (*.f64 y y)))

   1. Initial program 0.0%

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

   3. Taylor expanded in y around inf

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

   5. Applied rewrites 72.0%

      \[\leadsto \color{blue}{-1} \]
3. Recombined 3 regimes into one program.

4. Final simplification 91.0%

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

Alternative 4: 91.3% accurate, 0.3× speedup

Math

\[\begin{array}{l} y_m = \left|y\right| \\ \begin{array}{l} t_0 := \frac{\left(x + y\_m\right) \cdot \left(x - y\_m\right)}{y\_m \cdot y\_m + x \cdot x}\\ \mathbf{if}\;t\_0 \leq -0.5:\\ \;\;\;\;-1\\ \mathbf{elif}\;t\_0 \leq 2:\\ \;\;\;\;\mathsf{fma}\left(-2 \cdot y\_m, \frac{y\_m}{x \cdot x}, 1\right)\\ \mathbf{else}:\\ \;\;\;\;-1\\ \end{array} \end{array} \]

FPCore

y_m = (fabs.f64 y)
(FPCore (x y_m)
 :precision binary64
 (let* ((t_0 (/ (* (+ x y_m) (- x y_m)) (+ (* y_m y_m) (* x x)))))
   (if (<= t_0 -0.5)
     -1.0
     (if (<= t_0 2.0) (fma (* -2.0 y_m) (/ y_m (* x x)) 1.0) -1.0))))

C

y_m = fabs(y);
double code(double x, double y_m) {
	double t_0 = ((x + y_m) * (x - y_m)) / ((y_m * y_m) + (x * x));
	double tmp;
	if (t_0 <= -0.5) {
		tmp = -1.0;
	} else if (t_0 <= 2.0) {
		tmp = fma((-2.0 * y_m), (y_m / (x * x)), 1.0);
	} else {
		tmp = -1.0;
	}
	return tmp;
}

Julia

y_m = abs(y)
function code(x, y_m)
	t_0 = Float64(Float64(Float64(x + y_m) * Float64(x - y_m)) / Float64(Float64(y_m * y_m) + Float64(x * x)))
	tmp = 0.0
	if (t_0 <= -0.5)
		tmp = -1.0;
	elseif (t_0 <= 2.0)
		tmp = fma(Float64(-2.0 * y_m), Float64(y_m / Float64(x * x)), 1.0);
	else
		tmp = -1.0;
	end
	return tmp
end

Wolfram

y_m = N[Abs[y], $MachinePrecision]
code[x_, y$95$m_] := Block[{t$95$0 = N[(N[(N[(x + y$95$m), $MachinePrecision] * N[(x - y$95$m), $MachinePrecision]), $MachinePrecision] / N[(N[(y$95$m * y$95$m), $MachinePrecision] + N[(x * x), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]}, If[LessEqual[t$95$0, -0.5], -1.0, If[LessEqual[t$95$0, 2.0], N[(N[(-2.0 * y$95$m), $MachinePrecision] * N[(y$95$m / N[(x * x), $MachinePrecision]), $MachinePrecision] + 1.0), $MachinePrecision], -1.0]]]

Derivation

1. Split input into 2 regimes

2. if (/.f64 (*.f64 (-.f64 x y) (+.f64 x y)) (+.f64 (*.f64 x x) (*.f64 y y))) < -0.5 or 2 < (/.f64 (*.f64 (-.f64 x y) (+.f64 x y)) (+.f64 (*.f64 x x) (*.f64 y y)))

   1. Initial program 56.7%

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

   3. Taylor expanded in y around inf

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

   5. Applied rewrites 87.6%

      \[\leadsto \color{blue}{-1} \]

   if -0.5 < (/.f64 (*.f64 (-.f64 x y) (+.f64 x y)) (+.f64 (*.f64 x x) (*.f64 y y))) < 2

   1. Initial program 100.0%

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

   3. Taylor expanded in y around 0

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

      1. +-commutative N/A

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

      2. *-commutative N/A

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

      3. associate-*l/ N/A

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

      4. *-commutative N/A

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

      5. unpow2 N/A

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

      6. associate-*r* N/A

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

      7. associate-/l* N/A

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

      8. lower-fma.f64 N/A

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

      9. lower-*.f64 N/A

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

      10. lower-/.f64 N/A

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

      11. unpow2 N/A

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

      12. lower-*.f64 99.3

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

   5. Applied rewrites 99.3%

      \[\leadsto \color{blue}{\mathsf{fma}\left(-2 \cdot y, \frac{y}{x \cdot x}, 1\right)} \]
3. Recombined 2 regimes into one program.

4. Final simplification 90.7%

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

Alternative 5: 91.1% accurate, 0.4× speedup

Math

\[\begin{array}{l} y_m = \left|y\right| \\ \begin{array}{l} t_0 := \frac{\left(x + y\_m\right) \cdot \left(x - y\_m\right)}{y\_m \cdot y\_m + x \cdot x}\\ \mathbf{if}\;t\_0 \leq -5 \cdot 10^{-312}:\\ \;\;\;\;-1\\ \mathbf{elif}\;t\_0 \leq \infty:\\ \;\;\;\;1\\ \mathbf{else}:\\ \;\;\;\;-1\\ \end{array} \end{array} \]

FPCore

y_m = (fabs.f64 y)
(FPCore (x y_m)
 :precision binary64
 (let* ((t_0 (/ (* (+ x y_m) (- x y_m)) (+ (* y_m y_m) (* x x)))))
   (if (<= t_0 -5e-312) -1.0 (if (<= t_0 INFINITY) 1.0 -1.0))))

C

y_m = fabs(y);
double code(double x, double y_m) {
	double t_0 = ((x + y_m) * (x - y_m)) / ((y_m * y_m) + (x * x));
	double tmp;
	if (t_0 <= -5e-312) {
		tmp = -1.0;
	} else if (t_0 <= ((double) INFINITY)) {
		tmp = 1.0;
	} else {
		tmp = -1.0;
	}
	return tmp;
}

Java

y_m = Math.abs(y);
public static double code(double x, double y_m) {
	double t_0 = ((x + y_m) * (x - y_m)) / ((y_m * y_m) + (x * x));
	double tmp;
	if (t_0 <= -5e-312) {
		tmp = -1.0;
	} else if (t_0 <= Double.POSITIVE_INFINITY) {
		tmp = 1.0;
	} else {
		tmp = -1.0;
	}
	return tmp;
}

Python

y_m = math.fabs(y)
def code(x, y_m):
	t_0 = ((x + y_m) * (x - y_m)) / ((y_m * y_m) + (x * x))
	tmp = 0
	if t_0 <= -5e-312:
		tmp = -1.0
	elif t_0 <= math.inf:
		tmp = 1.0
	else:
		tmp = -1.0
	return tmp

Julia

y_m = abs(y)
function code(x, y_m)
	t_0 = Float64(Float64(Float64(x + y_m) * Float64(x - y_m)) / Float64(Float64(y_m * y_m) + Float64(x * x)))
	tmp = 0.0
	if (t_0 <= -5e-312)
		tmp = -1.0;
	elseif (t_0 <= Inf)
		tmp = 1.0;
	else
		tmp = -1.0;
	end
	return tmp
end

MATLAB

y_m = abs(y);
function tmp_2 = code(x, y_m)
	t_0 = ((x + y_m) * (x - y_m)) / ((y_m * y_m) + (x * x));
	tmp = 0.0;
	if (t_0 <= -5e-312)
		tmp = -1.0;
	elseif (t_0 <= Inf)
		tmp = 1.0;
	else
		tmp = -1.0;
	end
	tmp_2 = tmp;
end

Wolfram

y_m = N[Abs[y], $MachinePrecision]
code[x_, y$95$m_] := Block[{t$95$0 = N[(N[(N[(x + y$95$m), $MachinePrecision] * N[(x - y$95$m), $MachinePrecision]), $MachinePrecision] / N[(N[(y$95$m * y$95$m), $MachinePrecision] + N[(x * x), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]}, If[LessEqual[t$95$0, -5e-312], -1.0, If[LessEqual[t$95$0, Infinity], 1.0, -1.0]]]

Derivation

1. Split input into 2 regimes

2. if (/.f64 (*.f64 (-.f64 x y) (+.f64 x y)) (+.f64 (*.f64 x x) (*.f64 y y))) < -5.0000000000022e-312 or +inf.0 < (/.f64 (*.f64 (-.f64 x y) (+.f64 x y)) (+.f64 (*.f64 x x) (*.f64 y y)))

   1. Initial program 56.7%

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

   3. Taylor expanded in y around inf

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

   5. Applied rewrites 87.6%

      \[\leadsto \color{blue}{-1} \]

   if -5.0000000000022e-312 < (/.f64 (*.f64 (-.f64 x y) (+.f64 x y)) (+.f64 (*.f64 x x) (*.f64 y y))) < +inf.0

   1. Initial program 100.0%

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

   3. Taylor expanded in y around 0

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

   5. Applied rewrites 98.4%

      \[\leadsto \color{blue}{1} \]
3. Recombined 2 regimes into one program.

4. Final simplification 90.4%

   \[\leadsto \begin{array}{l} \mathbf{if}\;\frac{\left(x + y\right) \cdot \left(x - y\right)}{y \cdot y + x \cdot x} \leq -5 \cdot 10^{-312}:\\ \;\;\;\;-1\\ \mathbf{elif}\;\frac{\left(x + y\right) \cdot \left(x - y\right)}{y \cdot y + x \cdot x} \leq \infty:\\ \;\;\;\;1\\ \mathbf{else}:\\ \;\;\;\;-1\\ \end{array} \]
5. Add Preprocessing

Alternative 6: 66.0% accurate, 36.0× speedup

Math

\[\begin{array}{l} y_m = \left|y\right| \\ -1 \end{array} \]

FPCore

y_m = (fabs.f64 y)
(FPCore (x y_m) :precision binary64 -1.0)

C

y_m = fabs(y);
double code(double x, double y_m) {
	return -1.0;
}

Fortran

y_m = abs(y)
real(8) function code(x, y_m)
    real(8), intent (in) :: x
    real(8), intent (in) :: y_m
    code = -1.0d0
end function

Java

y_m = Math.abs(y);
public static double code(double x, double y_m) {
	return -1.0;
}

Python

y_m = math.fabs(y)
def code(x, y_m):
	return -1.0

Julia

y_m = abs(y)
function code(x, y_m)
	return -1.0
end

MATLAB

y_m = abs(y);
function tmp = code(x, y_m)
	tmp = -1.0;
end

Wolfram

y_m = N[Abs[y], $MachinePrecision]
code[x_, y$95$m_] := -1.0

Derivation

1. Initial program 68.2%

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

3. Taylor expanded in y around inf

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

5. Applied rewrites 64.7%

   \[\leadsto \color{blue}{-1} \]
6. Add Preprocessing

Developer Target 1: 99.9% accurate, 0.4× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} t_0 := \left|\frac{x}{y}\right|\\ \mathbf{if}\;0.5 < t\_0 \land t\_0 < 2:\\ \;\;\;\;\frac{\left(x - y\right) \cdot \left(x + y\right)}{x \cdot x + y \cdot y}\\ \mathbf{else}:\\ \;\;\;\;1 - \frac{2}{1 + \frac{x}{y} \cdot \frac{x}{y}}\\ \end{array} \end{array} \]

FPCore

(FPCore (x y)
 :precision binary64
 (let* ((t_0 (fabs (/ x y))))
   (if (and (< 0.5 t_0) (< t_0 2.0))
     (/ (* (- x y) (+ x y)) (+ (* x x) (* y y)))
     (- 1.0 (/ 2.0 (+ 1.0 (* (/ x y) (/ x y))))))))

C

double code(double x, double y) {
	double t_0 = fabs((x / y));
	double tmp;
	if ((0.5 < t_0) && (t_0 < 2.0)) {
		tmp = ((x - y) * (x + y)) / ((x * x) + (y * y));
	} else {
		tmp = 1.0 - (2.0 / (1.0 + ((x / y) * (x / y))));
	}
	return tmp;
}

Fortran

real(8) function code(x, y)
    real(8), intent (in) :: x
    real(8), intent (in) :: y
    real(8) :: t_0
    real(8) :: tmp
    t_0 = abs((x / y))
    if ((0.5d0 < t_0) .and. (t_0 < 2.0d0)) then
        tmp = ((x - y) * (x + y)) / ((x * x) + (y * y))
    else
        tmp = 1.0d0 - (2.0d0 / (1.0d0 + ((x / y) * (x / y))))
    end if
    code = tmp
end function

Java

public static double code(double x, double y) {
	double t_0 = Math.abs((x / y));
	double tmp;
	if ((0.5 < t_0) && (t_0 < 2.0)) {
		tmp = ((x - y) * (x + y)) / ((x * x) + (y * y));
	} else {
		tmp = 1.0 - (2.0 / (1.0 + ((x / y) * (x / y))));
	}
	return tmp;
}

Python

def code(x, y):
	t_0 = math.fabs((x / y))
	tmp = 0
	if (0.5 < t_0) and (t_0 < 2.0):
		tmp = ((x - y) * (x + y)) / ((x * x) + (y * y))
	else:
		tmp = 1.0 - (2.0 / (1.0 + ((x / y) * (x / y))))
	return tmp

Julia

function code(x, y)
	t_0 = abs(Float64(x / y))
	tmp = 0.0
	if ((0.5 < t_0) && (t_0 < 2.0))
		tmp = Float64(Float64(Float64(x - y) * Float64(x + y)) / Float64(Float64(x * x) + Float64(y * y)));
	else
		tmp = Float64(1.0 - Float64(2.0 / Float64(1.0 + Float64(Float64(x / y) * Float64(x / y)))));
	end
	return tmp
end

MATLAB

function tmp_2 = code(x, y)
	t_0 = abs((x / y));
	tmp = 0.0;
	if ((0.5 < t_0) && (t_0 < 2.0))
		tmp = ((x - y) * (x + y)) / ((x * x) + (y * y));
	else
		tmp = 1.0 - (2.0 / (1.0 + ((x / y) * (x / y))));
	end
	tmp_2 = tmp;
end

Wolfram

code[x_, y_] := Block[{t$95$0 = N[Abs[N[(x / y), $MachinePrecision]], $MachinePrecision]}, If[And[Less[0.5, t$95$0], Less[t$95$0, 2.0]], N[(N[(N[(x - y), $MachinePrecision] * N[(x + y), $MachinePrecision]), $MachinePrecision] / N[(N[(x * x), $MachinePrecision] + N[(y * y), $MachinePrecision]), $MachinePrecision]), $MachinePrecision], N[(1.0 - N[(2.0 / N[(1.0 + N[(N[(x / y), $MachinePrecision] * N[(x / y), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]]]

Reproduce

herbie shell --seed 2024235 
(FPCore (x y)
  :name "Kahan p9 Example"
  :precision binary64
  :pre (and (and (< 0.0 x) (< x 1.0)) (< y 1.0))

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

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