Data.Metrics.Snapshot:quantile from metrics-0.3.0.2

Percentage Accurate: 100.0% → 100.0%

Time: 3.6s

Alternatives: 14

Speedup: 1.0×

• 35.2% of points produce a very large (infinite) output. You may want to add a precondition.

Specification

Math

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

FPCore

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

C

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

Fortran

module fmin_fmax_functions
    implicit none
    private
    public fmax
    public fmin

    interface fmax
        module procedure fmax88
        module procedure fmax44
        module procedure fmax84
        module procedure fmax48
    end interface
    interface fmin
        module procedure fmin88
        module procedure fmin44
        module procedure fmin84
        module procedure fmin48
    end interface
contains
    real(8) function fmax88(x, y) result (res)
        real(8), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(x, max(x, y), y /= y), x /= x)
    end function
    real(4) function fmax44(x, y) result (res)
        real(4), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(y, merge(x, max(x, y), y /= y), x /= x)
    end function
    real(8) function fmax84(x, y) result(res)
        real(8), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(dble(y), merge(x, max(x, dble(y)), y /= y), x /= x)
    end function
    real(8) function fmax48(x, y) result(res)
        real(4), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(dble(x), max(dble(x), y), y /= y), x /= x)
    end function
    real(8) function fmin88(x, y) result (res)
        real(8), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(x, min(x, y), y /= y), x /= x)
    end function
    real(4) function fmin44(x, y) result (res)
        real(4), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(y, merge(x, min(x, y), y /= y), x /= x)
    end function
    real(8) function fmin84(x, y) result(res)
        real(8), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(dble(y), merge(x, min(x, dble(y)), y /= y), x /= x)
    end function
    real(8) function fmin48(x, y) result(res)
        real(4), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(dble(x), min(dble(x), y), y /= y), x /= x)
    end function
end module

real(8) function code(x, y, z, t)
use fmin_fmax_functions
    real(8), intent (in) :: x
    real(8), intent (in) :: y
    real(8), intent (in) :: z
    real(8), intent (in) :: t
    code = x + ((y - z) * (t - x))
end function

Java

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

Python

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

Julia

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

MATLAB

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

Wolfram

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

Initial Program: 100.0% accurate, 1.0× speedup

Math

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

FPCore

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

C

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

Fortran

module fmin_fmax_functions
    implicit none
    private
    public fmax
    public fmin

    interface fmax
        module procedure fmax88
        module procedure fmax44
        module procedure fmax84
        module procedure fmax48
    end interface
    interface fmin
        module procedure fmin88
        module procedure fmin44
        module procedure fmin84
        module procedure fmin48
    end interface
contains
    real(8) function fmax88(x, y) result (res)
        real(8), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(x, max(x, y), y /= y), x /= x)
    end function
    real(4) function fmax44(x, y) result (res)
        real(4), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(y, merge(x, max(x, y), y /= y), x /= x)
    end function
    real(8) function fmax84(x, y) result(res)
        real(8), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(dble(y), merge(x, max(x, dble(y)), y /= y), x /= x)
    end function
    real(8) function fmax48(x, y) result(res)
        real(4), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(dble(x), max(dble(x), y), y /= y), x /= x)
    end function
    real(8) function fmin88(x, y) result (res)
        real(8), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(x, min(x, y), y /= y), x /= x)
    end function
    real(4) function fmin44(x, y) result (res)
        real(4), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(y, merge(x, min(x, y), y /= y), x /= x)
    end function
    real(8) function fmin84(x, y) result(res)
        real(8), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(dble(y), merge(x, min(x, dble(y)), y /= y), x /= x)
    end function
    real(8) function fmin48(x, y) result(res)
        real(4), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(dble(x), min(dble(x), y), y /= y), x /= x)
    end function
end module

real(8) function code(x, y, z, t)
use fmin_fmax_functions
    real(8), intent (in) :: x
    real(8), intent (in) :: y
    real(8), intent (in) :: z
    real(8), intent (in) :: t
    code = x + ((y - z) * (t - x))
end function

Java

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

Python

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

Julia

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

MATLAB

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

Wolfram

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

Alternative 1: 100.0% accurate, 1.0× speedup

Math

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

FPCore

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

C

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

Fortran

module fmin_fmax_functions
    implicit none
    private
    public fmax
    public fmin

    interface fmax
        module procedure fmax88
        module procedure fmax44
        module procedure fmax84
        module procedure fmax48
    end interface
    interface fmin
        module procedure fmin88
        module procedure fmin44
        module procedure fmin84
        module procedure fmin48
    end interface
contains
    real(8) function fmax88(x, y) result (res)
        real(8), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(x, max(x, y), y /= y), x /= x)
    end function
    real(4) function fmax44(x, y) result (res)
        real(4), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(y, merge(x, max(x, y), y /= y), x /= x)
    end function
    real(8) function fmax84(x, y) result(res)
        real(8), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(dble(y), merge(x, max(x, dble(y)), y /= y), x /= x)
    end function
    real(8) function fmax48(x, y) result(res)
        real(4), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(dble(x), max(dble(x), y), y /= y), x /= x)
    end function
    real(8) function fmin88(x, y) result (res)
        real(8), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(x, min(x, y), y /= y), x /= x)
    end function
    real(4) function fmin44(x, y) result (res)
        real(4), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(y, merge(x, min(x, y), y /= y), x /= x)
    end function
    real(8) function fmin84(x, y) result(res)
        real(8), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(dble(y), merge(x, min(x, dble(y)), y /= y), x /= x)
    end function
    real(8) function fmin48(x, y) result(res)
        real(4), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(dble(x), min(dble(x), y), y /= y), x /= x)
    end function
end module

real(8) function code(x, y, z, t)
use fmin_fmax_functions
    real(8), intent (in) :: x
    real(8), intent (in) :: y
    real(8), intent (in) :: z
    real(8), intent (in) :: t
    code = x + ((y - z) * (t - x))
end function

Java

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

Python

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

Julia

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

MATLAB

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

Wolfram

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

Derivation

1. Initial program 100.0%

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

Alternative 2: 84.0% accurate, 0.7× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;z \leq -10000000000000:\\ \;\;\;\;\mathsf{fma}\left(-t, z, \mathsf{fma}\left(x, z, x\right)\right)\\ \mathbf{elif}\;z \leq 1.75 \cdot 10^{+70}:\\ \;\;\;\;\mathsf{fma}\left(t - x, y, x\right)\\ \mathbf{else}:\\ \;\;\;\;\left(-z\right) \cdot \left(t - x\right)\\ \end{array} \end{array} \]

FPCore

(FPCore (x y z t)
 :precision binary64
 (if (<= z -10000000000000.0)
   (fma (- t) z (fma x z x))
   (if (<= z 1.75e+70) (fma (- t x) y x) (* (- z) (- t x)))))

C

double code(double x, double y, double z, double t) {
	double tmp;
	if (z <= -10000000000000.0) {
		tmp = fma(-t, z, fma(x, z, x));
	} else if (z <= 1.75e+70) {
		tmp = fma((t - x), y, x);
	} else {
		tmp = -z * (t - x);
	}
	return tmp;
}

Julia

function code(x, y, z, t)
	tmp = 0.0
	if (z <= -10000000000000.0)
		tmp = fma(Float64(-t), z, fma(x, z, x));
	elseif (z <= 1.75e+70)
		tmp = fma(Float64(t - x), y, x);
	else
		tmp = Float64(Float64(-z) * Float64(t - x));
	end
	return tmp
end

Wolfram

code[x_, y_, z_, t_] := If[LessEqual[z, -10000000000000.0], N[((-t) * z + N[(x * z + x), $MachinePrecision]), $MachinePrecision], If[LessEqual[z, 1.75e+70], N[(N[(t - x), $MachinePrecision] * y + x), $MachinePrecision], N[((-z) * N[(t - x), $MachinePrecision]), $MachinePrecision]]]

Derivation

1. Split input into 3 regimes

2. if z < -1e13

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in x around 0

      \[\leadsto \color{blue}{t \cdot \left(y - z\right) + x \cdot \left(1 + -1 \cdot \left(y - z\right)\right)} \]
   3. Step-by-step derivation

      1. +-commutative N/A

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

      2. *-commutative N/A

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

      3. lower-fma.f64 N/A

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

      4. +-commutative N/A

         \[\leadsto \mathsf{fma}\left(-1 \cdot \left(y - z\right) + 1, x, t \cdot \left(y - z\right)\right) \]

      5. add-flip N/A

         \[\leadsto \mathsf{fma}\left(-1 \cdot \left(y - z\right) - \left(\mathsf{neg}\left(1\right)\right), x, t \cdot \left(y - z\right)\right) \]

      6. metadata-eval N/A

         \[\leadsto \mathsf{fma}\left(-1 \cdot \left(y - z\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      7. lower--.f64 N/A

         \[\leadsto \mathsf{fma}\left(-1 \cdot \left(y - z\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      8. mul-1-neg N/A

         \[\leadsto \mathsf{fma}\left(\left(\mathsf{neg}\left(\left(y - z\right)\right)\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      9. lower-neg.f64 N/A

         \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      10. lift--.f64 N/A

          \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      11. *-commutative N/A

          \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, \left(y - z\right) \cdot t\right) \]

      12. lower-*.f64 N/A

          \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, \left(y - z\right) \cdot t\right) \]

      13. lift--.f64 98.2

          \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, \left(y - z\right) \cdot t\right) \]

   4. Applied rewrites 98.2%

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

   5. Taylor expanded in z around -inf

      \[\leadsto -1 \cdot \color{blue}{\left(z \cdot \left(t + -1 \cdot x\right)\right)} \]
   6. Step-by-step derivation

      1. associate-*r* N/A

         \[\leadsto \left(-1 \cdot z\right) \cdot \left(t + \color{blue}{-1 \cdot x}\right) \]

      2. mul-1-neg N/A

         \[\leadsto \left(\mathsf{neg}\left(z\right)\right) \cdot \left(t + \color{blue}{-1} \cdot x\right) \]

      3. lift-neg.f64 N/A

         \[\leadsto \left(-z\right) \cdot \left(t + \color{blue}{-1} \cdot x\right) \]

      4. mul-1-neg N/A

         \[\leadsto \left(-z\right) \cdot \left(t + \left(\mathsf{neg}\left(x\right)\right)\right) \]

      5. sub-flip N/A

         \[\leadsto \left(-z\right) \cdot \left(t - x\right) \]

      6. lower-*.f64 N/A

         \[\leadsto \left(-z\right) \cdot \left(t - \color{blue}{x}\right) \]

      7. lift--.f64 44.5

         \[\leadsto \left(-z\right) \cdot \left(t - x\right) \]

   7. Applied rewrites 44.5%

      \[\leadsto \left(-z\right) \cdot \color{blue}{\left(t - x\right)} \]

   8. Taylor expanded in y around 0

      \[\leadsto -1 \cdot \left(t \cdot z\right) + \color{blue}{x \cdot \left(1 + z\right)} \]
   9. Step-by-step derivation

      1. associate-*r* N/A

         \[\leadsto \left(-1 \cdot t\right) \cdot z + x \cdot \left(\color{blue}{1} + z\right) \]

      2. lower-fma.f64 N/A

         \[\leadsto \mathsf{fma}\left(-1 \cdot t, z, x \cdot \left(1 + z\right)\right) \]

      3. mul-1-neg N/A

         \[\leadsto \mathsf{fma}\left(\mathsf{neg}\left(t\right), z, x \cdot \left(1 + z\right)\right) \]

      4. lower-neg.f64 N/A

         \[\leadsto \mathsf{fma}\left(-t, z, x \cdot \left(1 + z\right)\right) \]

      5. *-commutative N/A

         \[\leadsto \mathsf{fma}\left(-t, z, \left(1 + z\right) \cdot x\right) \]

      6. +-commutative N/A

         \[\leadsto \mathsf{fma}\left(-t, z, \left(z + 1\right) \cdot x\right) \]

      7. distribute-lft1-in N/A

         \[\leadsto \mathsf{fma}\left(-t, z, z \cdot x + x\right) \]

      8. *-commutative N/A

         \[\leadsto \mathsf{fma}\left(-t, z, x \cdot z + x\right) \]

      9. lower-fma.f64 58.5

         \[\leadsto \mathsf{fma}\left(-t, z, \mathsf{fma}\left(x, z, x\right)\right) \]

   10. Applied rewrites 58.5%

       \[\leadsto \mathsf{fma}\left(-t, \color{blue}{z}, \mathsf{fma}\left(x, z, x\right)\right) \]

   if -1e13 < z < 1.75000000000000001e70

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in z around 0

      \[\leadsto \color{blue}{x + y \cdot \left(t - x\right)} \]
   3. Step-by-step derivation

      1. +-commutative N/A

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

      2. *-commutative N/A

         \[\leadsto \left(t - x\right) \cdot y + x \]

      3. lower-fma.f64 N/A

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

      4. lift--.f64 60.8

         \[\leadsto \mathsf{fma}\left(t - x, y, x\right) \]

   4. Applied rewrites 60.8%

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

   if 1.75000000000000001e70 < z

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in x around 0

      \[\leadsto \color{blue}{t \cdot \left(y - z\right) + x \cdot \left(1 + -1 \cdot \left(y - z\right)\right)} \]
   3. Step-by-step derivation

      1. +-commutative N/A

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

      2. *-commutative N/A

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

      3. lower-fma.f64 N/A

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

      4. +-commutative N/A

         \[\leadsto \mathsf{fma}\left(-1 \cdot \left(y - z\right) + 1, x, t \cdot \left(y - z\right)\right) \]

      5. add-flip N/A

         \[\leadsto \mathsf{fma}\left(-1 \cdot \left(y - z\right) - \left(\mathsf{neg}\left(1\right)\right), x, t \cdot \left(y - z\right)\right) \]

      6. metadata-eval N/A

         \[\leadsto \mathsf{fma}\left(-1 \cdot \left(y - z\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      7. lower--.f64 N/A

         \[\leadsto \mathsf{fma}\left(-1 \cdot \left(y - z\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      8. mul-1-neg N/A

         \[\leadsto \mathsf{fma}\left(\left(\mathsf{neg}\left(\left(y - z\right)\right)\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      9. lower-neg.f64 N/A

         \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      10. lift--.f64 N/A

          \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      11. *-commutative N/A

          \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, \left(y - z\right) \cdot t\right) \]

      12. lower-*.f64 N/A

          \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, \left(y - z\right) \cdot t\right) \]

      13. lift--.f64 98.2

          \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, \left(y - z\right) \cdot t\right) \]

   4. Applied rewrites 98.2%

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

   5. Taylor expanded in z around -inf

      \[\leadsto -1 \cdot \color{blue}{\left(z \cdot \left(t + -1 \cdot x\right)\right)} \]
   6. Step-by-step derivation

      1. associate-*r* N/A

         \[\leadsto \left(-1 \cdot z\right) \cdot \left(t + \color{blue}{-1 \cdot x}\right) \]

      2. mul-1-neg N/A

         \[\leadsto \left(\mathsf{neg}\left(z\right)\right) \cdot \left(t + \color{blue}{-1} \cdot x\right) \]

      3. lift-neg.f64 N/A

         \[\leadsto \left(-z\right) \cdot \left(t + \color{blue}{-1} \cdot x\right) \]

      4. mul-1-neg N/A

         \[\leadsto \left(-z\right) \cdot \left(t + \left(\mathsf{neg}\left(x\right)\right)\right) \]

      5. sub-flip N/A

         \[\leadsto \left(-z\right) \cdot \left(t - x\right) \]

      6. lower-*.f64 N/A

         \[\leadsto \left(-z\right) \cdot \left(t - \color{blue}{x}\right) \]

      7. lift--.f64 44.5

         \[\leadsto \left(-z\right) \cdot \left(t - x\right) \]

   7. Applied rewrites 44.5%

      \[\leadsto \left(-z\right) \cdot \color{blue}{\left(t - x\right)} \]
3. Recombined 3 regimes into one program.
4. Add Preprocessing

Alternative 3: 83.4% accurate, 0.7× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} t_1 := \left(-z\right) \cdot \left(t - x\right)\\ \mathbf{if}\;z \leq -10000000000000:\\ \;\;\;\;t\_1\\ \mathbf{elif}\;z \leq 1.75 \cdot 10^{+70}:\\ \;\;\;\;\mathsf{fma}\left(t - x, y, x\right)\\ \mathbf{else}:\\ \;\;\;\;t\_1\\ \end{array} \end{array} \]

FPCore

(FPCore (x y z t)
 :precision binary64
 (let* ((t_1 (* (- z) (- t x))))
   (if (<= z -10000000000000.0)
     t_1
     (if (<= z 1.75e+70) (fma (- t x) y x) t_1))))

C

double code(double x, double y, double z, double t) {
	double t_1 = -z * (t - x);
	double tmp;
	if (z <= -10000000000000.0) {
		tmp = t_1;
	} else if (z <= 1.75e+70) {
		tmp = fma((t - x), y, x);
	} else {
		tmp = t_1;
	}
	return tmp;
}

Julia

function code(x, y, z, t)
	t_1 = Float64(Float64(-z) * Float64(t - x))
	tmp = 0.0
	if (z <= -10000000000000.0)
		tmp = t_1;
	elseif (z <= 1.75e+70)
		tmp = fma(Float64(t - x), y, x);
	else
		tmp = t_1;
	end
	return tmp
end

Wolfram

code[x_, y_, z_, t_] := Block[{t$95$1 = N[((-z) * N[(t - x), $MachinePrecision]), $MachinePrecision]}, If[LessEqual[z, -10000000000000.0], t$95$1, If[LessEqual[z, 1.75e+70], N[(N[(t - x), $MachinePrecision] * y + x), $MachinePrecision], t$95$1]]]

Derivation

1. Split input into 2 regimes

2. if z < -1e13 or 1.75000000000000001e70 < z

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in x around 0

      \[\leadsto \color{blue}{t \cdot \left(y - z\right) + x \cdot \left(1 + -1 \cdot \left(y - z\right)\right)} \]
   3. Step-by-step derivation

      1. +-commutative N/A

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

      2. *-commutative N/A

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

      3. lower-fma.f64 N/A

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

      4. +-commutative N/A

         \[\leadsto \mathsf{fma}\left(-1 \cdot \left(y - z\right) + 1, x, t \cdot \left(y - z\right)\right) \]

      5. add-flip N/A

         \[\leadsto \mathsf{fma}\left(-1 \cdot \left(y - z\right) - \left(\mathsf{neg}\left(1\right)\right), x, t \cdot \left(y - z\right)\right) \]

      6. metadata-eval N/A

         \[\leadsto \mathsf{fma}\left(-1 \cdot \left(y - z\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      7. lower--.f64 N/A

         \[\leadsto \mathsf{fma}\left(-1 \cdot \left(y - z\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      8. mul-1-neg N/A

         \[\leadsto \mathsf{fma}\left(\left(\mathsf{neg}\left(\left(y - z\right)\right)\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      9. lower-neg.f64 N/A

         \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      10. lift--.f64 N/A

          \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      11. *-commutative N/A

          \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, \left(y - z\right) \cdot t\right) \]

      12. lower-*.f64 N/A

          \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, \left(y - z\right) \cdot t\right) \]

      13. lift--.f64 98.2

          \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, \left(y - z\right) \cdot t\right) \]

   4. Applied rewrites 98.2%

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

   5. Taylor expanded in z around -inf

      \[\leadsto -1 \cdot \color{blue}{\left(z \cdot \left(t + -1 \cdot x\right)\right)} \]
   6. Step-by-step derivation

      1. associate-*r* N/A

         \[\leadsto \left(-1 \cdot z\right) \cdot \left(t + \color{blue}{-1 \cdot x}\right) \]

      2. mul-1-neg N/A

         \[\leadsto \left(\mathsf{neg}\left(z\right)\right) \cdot \left(t + \color{blue}{-1} \cdot x\right) \]

      3. lift-neg.f64 N/A

         \[\leadsto \left(-z\right) \cdot \left(t + \color{blue}{-1} \cdot x\right) \]

      4. mul-1-neg N/A

         \[\leadsto \left(-z\right) \cdot \left(t + \left(\mathsf{neg}\left(x\right)\right)\right) \]

      5. sub-flip N/A

         \[\leadsto \left(-z\right) \cdot \left(t - x\right) \]

      6. lower-*.f64 N/A

         \[\leadsto \left(-z\right) \cdot \left(t - \color{blue}{x}\right) \]

      7. lift--.f64 44.5

         \[\leadsto \left(-z\right) \cdot \left(t - x\right) \]

   7. Applied rewrites 44.5%

      \[\leadsto \left(-z\right) \cdot \color{blue}{\left(t - x\right)} \]

   if -1e13 < z < 1.75000000000000001e70

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in z around 0

      \[\leadsto \color{blue}{x + y \cdot \left(t - x\right)} \]
   3. Step-by-step derivation

      1. +-commutative N/A

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

      2. *-commutative N/A

         \[\leadsto \left(t - x\right) \cdot y + x \]

      3. lower-fma.f64 N/A

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

      4. lift--.f64 60.8

         \[\leadsto \mathsf{fma}\left(t - x, y, x\right) \]

   4. Applied rewrites 60.8%

      \[\leadsto \color{blue}{\mathsf{fma}\left(t - x, y, x\right)} \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 4: 72.5% accurate, 0.6× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} t_1 := \left(t - x\right) \cdot y\\ \mathbf{if}\;y \leq -3.9 \cdot 10^{-8}:\\ \;\;\;\;t\_1\\ \mathbf{elif}\;y \leq 1.7 \cdot 10^{-64}:\\ \;\;\;\;\mathsf{fma}\left(-z, t, x\right)\\ \mathbf{elif}\;y \leq 2.7 \cdot 10^{+29}:\\ \;\;\;\;\left(-z\right) \cdot \left(t - x\right)\\ \mathbf{else}:\\ \;\;\;\;t\_1\\ \end{array} \end{array} \]

FPCore

(FPCore (x y z t)
 :precision binary64
 (let* ((t_1 (* (- t x) y)))
   (if (<= y -3.9e-8)
     t_1
     (if (<= y 1.7e-64)
       (fma (- z) t x)
       (if (<= y 2.7e+29) (* (- z) (- t x)) t_1)))))

C

double code(double x, double y, double z, double t) {
	double t_1 = (t - x) * y;
	double tmp;
	if (y <= -3.9e-8) {
		tmp = t_1;
	} else if (y <= 1.7e-64) {
		tmp = fma(-z, t, x);
	} else if (y <= 2.7e+29) {
		tmp = -z * (t - x);
	} else {
		tmp = t_1;
	}
	return tmp;
}

Julia

function code(x, y, z, t)
	t_1 = Float64(Float64(t - x) * y)
	tmp = 0.0
	if (y <= -3.9e-8)
		tmp = t_1;
	elseif (y <= 1.7e-64)
		tmp = fma(Float64(-z), t, x);
	elseif (y <= 2.7e+29)
		tmp = Float64(Float64(-z) * Float64(t - x));
	else
		tmp = t_1;
	end
	return tmp
end

Wolfram

code[x_, y_, z_, t_] := Block[{t$95$1 = N[(N[(t - x), $MachinePrecision] * y), $MachinePrecision]}, If[LessEqual[y, -3.9e-8], t$95$1, If[LessEqual[y, 1.7e-64], N[((-z) * t + x), $MachinePrecision], If[LessEqual[y, 2.7e+29], N[((-z) * N[(t - x), $MachinePrecision]), $MachinePrecision], t$95$1]]]]

Derivation

1. Split input into 3 regimes

2. if y < -3.89999999999999985e-8 or 2.7e29 < y

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in y around inf

      \[\leadsto \color{blue}{y \cdot \left(t - x\right)} \]
   3. Step-by-step derivation

      1. *-commutative N/A

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

      2. lower-*.f64 N/A

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

      3. lift--.f64 45.7

         \[\leadsto \left(t - x\right) \cdot y \]

   4. Applied rewrites 45.7%

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

   if -3.89999999999999985e-8 < y < 1.70000000000000006e-64

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in x around 0

      \[\leadsto x + \left(y - z\right) \cdot \color{blue}{t} \]
   3. Step-by-step derivation

   4. Applied rewrites 64.7%

      \[\leadsto x + \left(y - z\right) \cdot \color{blue}{t} \]
   5. Step-by-step derivation

      1. lift-+.f64 N/A

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

      2. +-commutative N/A

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

      3. lift-*.f64 N/A

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

      4. lift--.f64 N/A

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

      5. lower-fma.f64 N/A

         \[\leadsto \color{blue}{\mathsf{fma}\left(y - z, t, x\right)} \]

      6. lift--.f64 64.7

         \[\leadsto \mathsf{fma}\left(\color{blue}{y - z}, t, x\right) \]

   6. Applied rewrites 64.7%

      \[\leadsto \color{blue}{\mathsf{fma}\left(y - z, t, x\right)} \]

   7. Taylor expanded in y around 0

      \[\leadsto \mathsf{fma}\left(\color{blue}{-1 \cdot z}, t, x\right) \]
   8. Step-by-step derivation

      1. mul-1-neg N/A

         \[\leadsto \mathsf{fma}\left(\mathsf{neg}\left(z\right), t, x\right) \]

      2. lift-neg.f64 41.8

         \[\leadsto \mathsf{fma}\left(-z, t, x\right) \]

   9. Applied rewrites 41.8%

      \[\leadsto \mathsf{fma}\left(\color{blue}{-z}, t, x\right) \]

   if 1.70000000000000006e-64 < y < 2.7e29

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in x around 0

      \[\leadsto \color{blue}{t \cdot \left(y - z\right) + x \cdot \left(1 + -1 \cdot \left(y - z\right)\right)} \]
   3. Step-by-step derivation

      1. +-commutative N/A

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

      2. *-commutative N/A

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

      3. lower-fma.f64 N/A

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

      4. +-commutative N/A

         \[\leadsto \mathsf{fma}\left(-1 \cdot \left(y - z\right) + 1, x, t \cdot \left(y - z\right)\right) \]

      5. add-flip N/A

         \[\leadsto \mathsf{fma}\left(-1 \cdot \left(y - z\right) - \left(\mathsf{neg}\left(1\right)\right), x, t \cdot \left(y - z\right)\right) \]

      6. metadata-eval N/A

         \[\leadsto \mathsf{fma}\left(-1 \cdot \left(y - z\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      7. lower--.f64 N/A

         \[\leadsto \mathsf{fma}\left(-1 \cdot \left(y - z\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      8. mul-1-neg N/A

         \[\leadsto \mathsf{fma}\left(\left(\mathsf{neg}\left(\left(y - z\right)\right)\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      9. lower-neg.f64 N/A

         \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      10. lift--.f64 N/A

          \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, t \cdot \left(y - z\right)\right) \]

      11. *-commutative N/A

          \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, \left(y - z\right) \cdot t\right) \]

      12. lower-*.f64 N/A

          \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, \left(y - z\right) \cdot t\right) \]

      13. lift--.f64 98.2

          \[\leadsto \mathsf{fma}\left(\left(-\left(y - z\right)\right) - -1, x, \left(y - z\right) \cdot t\right) \]

   4. Applied rewrites 98.2%

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

   5. Taylor expanded in z around -inf

      \[\leadsto -1 \cdot \color{blue}{\left(z \cdot \left(t + -1 \cdot x\right)\right)} \]
   6. Step-by-step derivation

      1. associate-*r* N/A

         \[\leadsto \left(-1 \cdot z\right) \cdot \left(t + \color{blue}{-1 \cdot x}\right) \]

      2. mul-1-neg N/A

         \[\leadsto \left(\mathsf{neg}\left(z\right)\right) \cdot \left(t + \color{blue}{-1} \cdot x\right) \]

      3. lift-neg.f64 N/A

         \[\leadsto \left(-z\right) \cdot \left(t + \color{blue}{-1} \cdot x\right) \]

      4. mul-1-neg N/A

         \[\leadsto \left(-z\right) \cdot \left(t + \left(\mathsf{neg}\left(x\right)\right)\right) \]

      5. sub-flip N/A

         \[\leadsto \left(-z\right) \cdot \left(t - x\right) \]

      6. lower-*.f64 N/A

         \[\leadsto \left(-z\right) \cdot \left(t - \color{blue}{x}\right) \]

      7. lift--.f64 44.5

         \[\leadsto \left(-z\right) \cdot \left(t - x\right) \]

   7. Applied rewrites 44.5%

      \[\leadsto \left(-z\right) \cdot \color{blue}{\left(t - x\right)} \]
3. Recombined 3 regimes into one program.
4. Add Preprocessing

Alternative 5: 72.4% accurate, 0.7× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} t_1 := \left(t - x\right) \cdot y\\ \mathbf{if}\;y \leq -3.9 \cdot 10^{-8}:\\ \;\;\;\;t\_1\\ \mathbf{elif}\;y \leq 1.36 \cdot 10^{-46}:\\ \;\;\;\;\mathsf{fma}\left(-z, t, x\right)\\ \mathbf{elif}\;y \leq 4.6 \cdot 10^{+25}:\\ \;\;\;\;\left(y - z\right) \cdot t\\ \mathbf{else}:\\ \;\;\;\;t\_1\\ \end{array} \end{array} \]

FPCore

(FPCore (x y z t)
 :precision binary64
 (let* ((t_1 (* (- t x) y)))
   (if (<= y -3.9e-8)
     t_1
     (if (<= y 1.36e-46)
       (fma (- z) t x)
       (if (<= y 4.6e+25) (* (- y z) t) t_1)))))

C

double code(double x, double y, double z, double t) {
	double t_1 = (t - x) * y;
	double tmp;
	if (y <= -3.9e-8) {
		tmp = t_1;
	} else if (y <= 1.36e-46) {
		tmp = fma(-z, t, x);
	} else if (y <= 4.6e+25) {
		tmp = (y - z) * t;
	} else {
		tmp = t_1;
	}
	return tmp;
}

Julia

function code(x, y, z, t)
	t_1 = Float64(Float64(t - x) * y)
	tmp = 0.0
	if (y <= -3.9e-8)
		tmp = t_1;
	elseif (y <= 1.36e-46)
		tmp = fma(Float64(-z), t, x);
	elseif (y <= 4.6e+25)
		tmp = Float64(Float64(y - z) * t);
	else
		tmp = t_1;
	end
	return tmp
end

Wolfram

code[x_, y_, z_, t_] := Block[{t$95$1 = N[(N[(t - x), $MachinePrecision] * y), $MachinePrecision]}, If[LessEqual[y, -3.9e-8], t$95$1, If[LessEqual[y, 1.36e-46], N[((-z) * t + x), $MachinePrecision], If[LessEqual[y, 4.6e+25], N[(N[(y - z), $MachinePrecision] * t), $MachinePrecision], t$95$1]]]]

Derivation

1. Split input into 3 regimes

2. if y < -3.89999999999999985e-8 or 4.5999999999999996e25 < y

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in y around inf

      \[\leadsto \color{blue}{y \cdot \left(t - x\right)} \]
   3. Step-by-step derivation

      1. *-commutative N/A

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

      2. lower-*.f64 N/A

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

      3. lift--.f64 45.7

         \[\leadsto \left(t - x\right) \cdot y \]

   4. Applied rewrites 45.7%

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

   if -3.89999999999999985e-8 < y < 1.3600000000000001e-46

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in x around 0

      \[\leadsto x + \left(y - z\right) \cdot \color{blue}{t} \]
   3. Step-by-step derivation

   4. Applied rewrites 64.7%

      \[\leadsto x + \left(y - z\right) \cdot \color{blue}{t} \]
   5. Step-by-step derivation

      1. lift-+.f64 N/A

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

      2. +-commutative N/A

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

      3. lift-*.f64 N/A

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

      4. lift--.f64 N/A

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

      5. lower-fma.f64 N/A

         \[\leadsto \color{blue}{\mathsf{fma}\left(y - z, t, x\right)} \]

      6. lift--.f64 64.7

         \[\leadsto \mathsf{fma}\left(\color{blue}{y - z}, t, x\right) \]

   6. Applied rewrites 64.7%

      \[\leadsto \color{blue}{\mathsf{fma}\left(y - z, t, x\right)} \]

   7. Taylor expanded in y around 0

      \[\leadsto \mathsf{fma}\left(\color{blue}{-1 \cdot z}, t, x\right) \]
   8. Step-by-step derivation

      1. mul-1-neg N/A

         \[\leadsto \mathsf{fma}\left(\mathsf{neg}\left(z\right), t, x\right) \]

      2. lift-neg.f64 41.8

         \[\leadsto \mathsf{fma}\left(-z, t, x\right) \]

   9. Applied rewrites 41.8%

      \[\leadsto \mathsf{fma}\left(\color{blue}{-z}, t, x\right) \]

   if 1.3600000000000001e-46 < y < 4.5999999999999996e25

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in x around 0

      \[\leadsto \color{blue}{t \cdot \left(y - z\right)} \]
   3. Step-by-step derivation

      1. *-commutative N/A

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

      2. lower-*.f64 N/A

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

      3. lift--.f64 49.6

         \[\leadsto \left(y - z\right) \cdot t \]

   4. Applied rewrites 49.6%

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

Alternative 6: 66.7% accurate, 0.7× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} t_1 := \left(t - x\right) \cdot y\\ \mathbf{if}\;y \leq -1.5 \cdot 10^{+50}:\\ \;\;\;\;t\_1\\ \mathbf{elif}\;y \leq 3 \cdot 10^{-54}:\\ \;\;\;\;\mathsf{fma}\left(z, x, x\right)\\ \mathbf{elif}\;y \leq 4.6 \cdot 10^{+25}:\\ \;\;\;\;\left(y - z\right) \cdot t\\ \mathbf{else}:\\ \;\;\;\;t\_1\\ \end{array} \end{array} \]

FPCore

(FPCore (x y z t)
 :precision binary64
 (let* ((t_1 (* (- t x) y)))
   (if (<= y -1.5e+50)
     t_1
     (if (<= y 3e-54) (fma z x x) (if (<= y 4.6e+25) (* (- y z) t) t_1)))))

C

double code(double x, double y, double z, double t) {
	double t_1 = (t - x) * y;
	double tmp;
	if (y <= -1.5e+50) {
		tmp = t_1;
	} else if (y <= 3e-54) {
		tmp = fma(z, x, x);
	} else if (y <= 4.6e+25) {
		tmp = (y - z) * t;
	} else {
		tmp = t_1;
	}
	return tmp;
}

Julia

function code(x, y, z, t)
	t_1 = Float64(Float64(t - x) * y)
	tmp = 0.0
	if (y <= -1.5e+50)
		tmp = t_1;
	elseif (y <= 3e-54)
		tmp = fma(z, x, x);
	elseif (y <= 4.6e+25)
		tmp = Float64(Float64(y - z) * t);
	else
		tmp = t_1;
	end
	return tmp
end

Wolfram

code[x_, y_, z_, t_] := Block[{t$95$1 = N[(N[(t - x), $MachinePrecision] * y), $MachinePrecision]}, If[LessEqual[y, -1.5e+50], t$95$1, If[LessEqual[y, 3e-54], N[(z * x + x), $MachinePrecision], If[LessEqual[y, 4.6e+25], N[(N[(y - z), $MachinePrecision] * t), $MachinePrecision], t$95$1]]]]

Derivation

1. Split input into 3 regimes

2. if y < -1.4999999999999999e50 or 4.5999999999999996e25 < y

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in y around inf

      \[\leadsto \color{blue}{y \cdot \left(t - x\right)} \]
   3. Step-by-step derivation

      1. *-commutative N/A

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

      2. lower-*.f64 N/A

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

      3. lift--.f64 45.7

         \[\leadsto \left(t - x\right) \cdot y \]

   4. Applied rewrites 45.7%

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

   if -1.4999999999999999e50 < y < 3.00000000000000009e-54

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in t around 0

      \[\leadsto \color{blue}{x + -1 \cdot \left(x \cdot \left(y - z\right)\right)} \]
   3. Step-by-step derivation

      1. +-commutative N/A

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

      2. associate-*r* N/A

         \[\leadsto \left(-1 \cdot x\right) \cdot \left(y - z\right) + x \]

      3. mul-1-neg N/A

         \[\leadsto \left(\mathsf{neg}\left(x\right)\right) \cdot \left(y - z\right) + x \]

      4. lower-fma.f64 N/A

         \[\leadsto \mathsf{fma}\left(\mathsf{neg}\left(x\right), \color{blue}{y - z}, x\right) \]

      5. lower-neg.f64 N/A

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

      6. lift--.f64 55.7

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

   4. Applied rewrites 55.7%

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

   5. Taylor expanded in y around 0

      \[\leadsto x + \color{blue}{x \cdot z} \]
   6. Step-by-step derivation

      1. +-commutative N/A

         \[\leadsto x \cdot z + x \]

      2. *-commutative N/A

         \[\leadsto z \cdot x + x \]

      3. lower-fma.f64 36.8

         \[\leadsto \mathsf{fma}\left(z, x, x\right) \]

   7. Applied rewrites 36.8%

      \[\leadsto \mathsf{fma}\left(z, \color{blue}{x}, x\right) \]

   if 3.00000000000000009e-54 < y < 4.5999999999999996e25

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in x around 0

      \[\leadsto \color{blue}{t \cdot \left(y - z\right)} \]
   3. Step-by-step derivation

      1. *-commutative N/A

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

      2. lower-*.f64 N/A

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

      3. lift--.f64 49.6

         \[\leadsto \left(y - z\right) \cdot t \]

   4. Applied rewrites 49.6%

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

Alternative 7: 66.5% accurate, 0.8× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} t_1 := \left(t - x\right) \cdot y\\ \mathbf{if}\;y \leq -1.5 \cdot 10^{+50}:\\ \;\;\;\;t\_1\\ \mathbf{elif}\;y \leq 2.4 \cdot 10^{+29}:\\ \;\;\;\;\mathsf{fma}\left(z, x, x\right)\\ \mathbf{else}:\\ \;\;\;\;t\_1\\ \end{array} \end{array} \]

FPCore

(FPCore (x y z t)
 :precision binary64
 (let* ((t_1 (* (- t x) y)))
   (if (<= y -1.5e+50) t_1 (if (<= y 2.4e+29) (fma z x x) t_1))))

C

double code(double x, double y, double z, double t) {
	double t_1 = (t - x) * y;
	double tmp;
	if (y <= -1.5e+50) {
		tmp = t_1;
	} else if (y <= 2.4e+29) {
		tmp = fma(z, x, x);
	} else {
		tmp = t_1;
	}
	return tmp;
}

Julia

function code(x, y, z, t)
	t_1 = Float64(Float64(t - x) * y)
	tmp = 0.0
	if (y <= -1.5e+50)
		tmp = t_1;
	elseif (y <= 2.4e+29)
		tmp = fma(z, x, x);
	else
		tmp = t_1;
	end
	return tmp
end

Wolfram

code[x_, y_, z_, t_] := Block[{t$95$1 = N[(N[(t - x), $MachinePrecision] * y), $MachinePrecision]}, If[LessEqual[y, -1.5e+50], t$95$1, If[LessEqual[y, 2.4e+29], N[(z * x + x), $MachinePrecision], t$95$1]]]

Derivation

1. Split input into 2 regimes

2. if y < -1.4999999999999999e50 or 2.4000000000000001e29 < y

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in y around inf

      \[\leadsto \color{blue}{y \cdot \left(t - x\right)} \]
   3. Step-by-step derivation

      1. *-commutative N/A

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

      2. lower-*.f64 N/A

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

      3. lift--.f64 45.7

         \[\leadsto \left(t - x\right) \cdot y \]

   4. Applied rewrites 45.7%

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

   if -1.4999999999999999e50 < y < 2.4000000000000001e29

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in t around 0

      \[\leadsto \color{blue}{x + -1 \cdot \left(x \cdot \left(y - z\right)\right)} \]
   3. Step-by-step derivation

      1. +-commutative N/A

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

      2. associate-*r* N/A

         \[\leadsto \left(-1 \cdot x\right) \cdot \left(y - z\right) + x \]

      3. mul-1-neg N/A

         \[\leadsto \left(\mathsf{neg}\left(x\right)\right) \cdot \left(y - z\right) + x \]

      4. lower-fma.f64 N/A

         \[\leadsto \mathsf{fma}\left(\mathsf{neg}\left(x\right), \color{blue}{y - z}, x\right) \]

      5. lower-neg.f64 N/A

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

      6. lift--.f64 55.7

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

   4. Applied rewrites 55.7%

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

   5. Taylor expanded in y around 0

      \[\leadsto x + \color{blue}{x \cdot z} \]
   6. Step-by-step derivation

      1. +-commutative N/A

         \[\leadsto x \cdot z + x \]

      2. *-commutative N/A

         \[\leadsto z \cdot x + x \]

      3. lower-fma.f64 36.8

         \[\leadsto \mathsf{fma}\left(z, x, x\right) \]

   7. Applied rewrites 36.8%

      \[\leadsto \mathsf{fma}\left(z, \color{blue}{x}, x\right) \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 8: 54.4% accurate, 0.5× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} t_1 := \left(-z\right) \cdot t\\ \mathbf{if}\;z \leq -1.6 \cdot 10^{+140}:\\ \;\;\;\;t\_1\\ \mathbf{elif}\;z \leq -14.5:\\ \;\;\;\;\mathsf{fma}\left(z, x, x\right)\\ \mathbf{elif}\;z \leq 7.7 \cdot 10^{-299}:\\ \;\;\;\;\mathsf{fma}\left(t, y, x\right)\\ \mathbf{elif}\;z \leq 7.2 \cdot 10^{+75}:\\ \;\;\;\;\left(1 - y\right) \cdot x\\ \mathbf{elif}\;z \leq 4.05 \cdot 10^{+242}:\\ \;\;\;\;t\_1\\ \mathbf{else}:\\ \;\;\;\;z \cdot x\\ \end{array} \end{array} \]

FPCore

(FPCore (x y z t)
 :precision binary64
 (let* ((t_1 (* (- z) t)))
   (if (<= z -1.6e+140)
     t_1
     (if (<= z -14.5)
       (fma z x x)
       (if (<= z 7.7e-299)
         (fma t y x)
         (if (<= z 7.2e+75)
           (* (- 1.0 y) x)
           (if (<= z 4.05e+242) t_1 (* z x))))))))

C

double code(double x, double y, double z, double t) {
	double t_1 = -z * t;
	double tmp;
	if (z <= -1.6e+140) {
		tmp = t_1;
	} else if (z <= -14.5) {
		tmp = fma(z, x, x);
	} else if (z <= 7.7e-299) {
		tmp = fma(t, y, x);
	} else if (z <= 7.2e+75) {
		tmp = (1.0 - y) * x;
	} else if (z <= 4.05e+242) {
		tmp = t_1;
	} else {
		tmp = z * x;
	}
	return tmp;
}

Julia

function code(x, y, z, t)
	t_1 = Float64(Float64(-z) * t)
	tmp = 0.0
	if (z <= -1.6e+140)
		tmp = t_1;
	elseif (z <= -14.5)
		tmp = fma(z, x, x);
	elseif (z <= 7.7e-299)
		tmp = fma(t, y, x);
	elseif (z <= 7.2e+75)
		tmp = Float64(Float64(1.0 - y) * x);
	elseif (z <= 4.05e+242)
		tmp = t_1;
	else
		tmp = Float64(z * x);
	end
	return tmp
end

Wolfram

code[x_, y_, z_, t_] := Block[{t$95$1 = N[((-z) * t), $MachinePrecision]}, If[LessEqual[z, -1.6e+140], t$95$1, If[LessEqual[z, -14.5], N[(z * x + x), $MachinePrecision], If[LessEqual[z, 7.7e-299], N[(t * y + x), $MachinePrecision], If[LessEqual[z, 7.2e+75], N[(N[(1.0 - y), $MachinePrecision] * x), $MachinePrecision], If[LessEqual[z, 4.05e+242], t$95$1, N[(z * x), $MachinePrecision]]]]]]]

Derivation

1. Split input into 5 regimes

2. if z < -1.60000000000000005e140 or 7.2e75 < z < 4.05e242

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in x around 0

      \[\leadsto \color{blue}{t \cdot \left(y - z\right)} \]
   3. Step-by-step derivation

      1. *-commutative N/A

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

      2. lower-*.f64 N/A

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

      3. lift--.f64 49.6

         \[\leadsto \left(y - z\right) \cdot t \]

   4. Applied rewrites 49.6%

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

   5. Taylor expanded in y around 0

      \[\leadsto \left(-1 \cdot z\right) \cdot t \]
   6. Step-by-step derivation

      1. mul-1-neg N/A

         \[\leadsto \left(\mathsf{neg}\left(z\right)\right) \cdot t \]

      2. lower-neg.f64 26.9

         \[\leadsto \left(-z\right) \cdot t \]

   7. Applied rewrites 26.9%

      \[\leadsto \left(-z\right) \cdot t \]

   if -1.60000000000000005e140 < z < -14.5

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in t around 0

      \[\leadsto \color{blue}{x + -1 \cdot \left(x \cdot \left(y - z\right)\right)} \]
   3. Step-by-step derivation

      1. +-commutative N/A

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

      2. associate-*r* N/A

         \[\leadsto \left(-1 \cdot x\right) \cdot \left(y - z\right) + x \]

      3. mul-1-neg N/A

         \[\leadsto \left(\mathsf{neg}\left(x\right)\right) \cdot \left(y - z\right) + x \]

      4. lower-fma.f64 N/A

         \[\leadsto \mathsf{fma}\left(\mathsf{neg}\left(x\right), \color{blue}{y - z}, x\right) \]

      5. lower-neg.f64 N/A

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

      6. lift--.f64 55.7

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

   4. Applied rewrites 55.7%

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

   5. Taylor expanded in y around 0

      \[\leadsto x + \color{blue}{x \cdot z} \]
   6. Step-by-step derivation

      1. +-commutative N/A

         \[\leadsto x \cdot z + x \]

      2. *-commutative N/A

         \[\leadsto z \cdot x + x \]

      3. lower-fma.f64 36.8

         \[\leadsto \mathsf{fma}\left(z, x, x\right) \]

   7. Applied rewrites 36.8%

      \[\leadsto \mathsf{fma}\left(z, \color{blue}{x}, x\right) \]

   if -14.5 < z < 7.70000000000000011e-299

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in z around 0

      \[\leadsto \color{blue}{x + y \cdot \left(t - x\right)} \]
   3. Step-by-step derivation

      1. +-commutative N/A

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

      2. *-commutative N/A

         \[\leadsto \left(t - x\right) \cdot y + x \]

      3. lower-fma.f64 N/A

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

      4. lift--.f64 60.8

         \[\leadsto \mathsf{fma}\left(t - x, y, x\right) \]

   4. Applied rewrites 60.8%

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

   5. Taylor expanded in x around 0

      \[\leadsto \mathsf{fma}\left(t, y, x\right) \]
   6. Step-by-step derivation

   7. Applied rewrites 42.0%

      \[\leadsto \mathsf{fma}\left(t, y, x\right) \]

   if 7.70000000000000011e-299 < z < 7.2e75

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in z around 0

      \[\leadsto \color{blue}{x + y \cdot \left(t - x\right)} \]
   3. Step-by-step derivation

      1. +-commutative N/A

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

      2. *-commutative N/A

         \[\leadsto \left(t - x\right) \cdot y + x \]

      3. lower-fma.f64 N/A

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

      4. lift--.f64 60.8

         \[\leadsto \mathsf{fma}\left(t - x, y, x\right) \]

   4. Applied rewrites 60.8%

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

   5. Taylor expanded in x around inf

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

      1. *-commutative N/A

         \[\leadsto \left(1 + -1 \cdot y\right) \cdot x \]

      2. mul-1-neg N/A

         \[\leadsto \left(1 + \left(\mathsf{neg}\left(y\right)\right)\right) \cdot x \]

      3. sub-flip N/A

         \[\leadsto \left(1 - y\right) \cdot x \]

      4. lower-*.f64 N/A

         \[\leadsto \left(1 - y\right) \cdot x \]

      5. lower--.f64 38.1

         \[\leadsto \left(1 - y\right) \cdot x \]

   7. Applied rewrites 38.1%

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

   if 4.05e242 < z

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in t around 0

      \[\leadsto \color{blue}{x + -1 \cdot \left(x \cdot \left(y - z\right)\right)} \]
   3. Step-by-step derivation

      1. +-commutative N/A

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

      2. associate-*r* N/A

         \[\leadsto \left(-1 \cdot x\right) \cdot \left(y - z\right) + x \]

      3. mul-1-neg N/A

         \[\leadsto \left(\mathsf{neg}\left(x\right)\right) \cdot \left(y - z\right) + x \]

      4. lower-fma.f64 N/A

         \[\leadsto \mathsf{fma}\left(\mathsf{neg}\left(x\right), \color{blue}{y - z}, x\right) \]

      5. lower-neg.f64 N/A

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

      6. lift--.f64 55.7

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

   4. Applied rewrites 55.7%

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

   5. Taylor expanded in z around inf

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

      1. *-commutative N/A

         \[\leadsto z \cdot x \]

      2. lower-*.f64 21.8

         \[\leadsto z \cdot x \]

   7. Applied rewrites 21.8%

      \[\leadsto z \cdot \color{blue}{x} \]
3. Recombined 5 regimes into one program.
4. Add Preprocessing

Alternative 9: 52.2% accurate, 0.6× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} t_1 := \left(-z\right) \cdot t\\ \mathbf{if}\;z \leq -1.6 \cdot 10^{+140}:\\ \;\;\;\;t\_1\\ \mathbf{elif}\;z \leq -14.5:\\ \;\;\;\;\mathsf{fma}\left(z, x, x\right)\\ \mathbf{elif}\;z \leq 4600000:\\ \;\;\;\;\mathsf{fma}\left(t, y, x\right)\\ \mathbf{elif}\;z \leq 4.05 \cdot 10^{+242}:\\ \;\;\;\;t\_1\\ \mathbf{else}:\\ \;\;\;\;z \cdot x\\ \end{array} \end{array} \]

FPCore

(FPCore (x y z t)
 :precision binary64
 (let* ((t_1 (* (- z) t)))
   (if (<= z -1.6e+140)
     t_1
     (if (<= z -14.5)
       (fma z x x)
       (if (<= z 4600000.0) (fma t y x) (if (<= z 4.05e+242) t_1 (* z x)))))))

C

double code(double x, double y, double z, double t) {
	double t_1 = -z * t;
	double tmp;
	if (z <= -1.6e+140) {
		tmp = t_1;
	} else if (z <= -14.5) {
		tmp = fma(z, x, x);
	} else if (z <= 4600000.0) {
		tmp = fma(t, y, x);
	} else if (z <= 4.05e+242) {
		tmp = t_1;
	} else {
		tmp = z * x;
	}
	return tmp;
}

Julia

function code(x, y, z, t)
	t_1 = Float64(Float64(-z) * t)
	tmp = 0.0
	if (z <= -1.6e+140)
		tmp = t_1;
	elseif (z <= -14.5)
		tmp = fma(z, x, x);
	elseif (z <= 4600000.0)
		tmp = fma(t, y, x);
	elseif (z <= 4.05e+242)
		tmp = t_1;
	else
		tmp = Float64(z * x);
	end
	return tmp
end

Wolfram

code[x_, y_, z_, t_] := Block[{t$95$1 = N[((-z) * t), $MachinePrecision]}, If[LessEqual[z, -1.6e+140], t$95$1, If[LessEqual[z, -14.5], N[(z * x + x), $MachinePrecision], If[LessEqual[z, 4600000.0], N[(t * y + x), $MachinePrecision], If[LessEqual[z, 4.05e+242], t$95$1, N[(z * x), $MachinePrecision]]]]]]

Derivation

1. Split input into 4 regimes

2. if z < -1.60000000000000005e140 or 4.6e6 < z < 4.05e242

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in x around 0

      \[\leadsto \color{blue}{t \cdot \left(y - z\right)} \]
   3. Step-by-step derivation

      1. *-commutative N/A

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

      2. lower-*.f64 N/A

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

      3. lift--.f64 49.6

         \[\leadsto \left(y - z\right) \cdot t \]

   4. Applied rewrites 49.6%

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

   5. Taylor expanded in y around 0

      \[\leadsto \left(-1 \cdot z\right) \cdot t \]
   6. Step-by-step derivation

      1. mul-1-neg N/A

         \[\leadsto \left(\mathsf{neg}\left(z\right)\right) \cdot t \]

      2. lower-neg.f64 26.9

         \[\leadsto \left(-z\right) \cdot t \]

   7. Applied rewrites 26.9%

      \[\leadsto \left(-z\right) \cdot t \]

   if -1.60000000000000005e140 < z < -14.5

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in t around 0

      \[\leadsto \color{blue}{x + -1 \cdot \left(x \cdot \left(y - z\right)\right)} \]
   3. Step-by-step derivation

      1. +-commutative N/A

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

      2. associate-*r* N/A

         \[\leadsto \left(-1 \cdot x\right) \cdot \left(y - z\right) + x \]

      3. mul-1-neg N/A

         \[\leadsto \left(\mathsf{neg}\left(x\right)\right) \cdot \left(y - z\right) + x \]

      4. lower-fma.f64 N/A

         \[\leadsto \mathsf{fma}\left(\mathsf{neg}\left(x\right), \color{blue}{y - z}, x\right) \]

      5. lower-neg.f64 N/A

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

      6. lift--.f64 55.7

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

   4. Applied rewrites 55.7%

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

   5. Taylor expanded in y around 0

      \[\leadsto x + \color{blue}{x \cdot z} \]
   6. Step-by-step derivation

      1. +-commutative N/A

         \[\leadsto x \cdot z + x \]

      2. *-commutative N/A

         \[\leadsto z \cdot x + x \]

      3. lower-fma.f64 36.8

         \[\leadsto \mathsf{fma}\left(z, x, x\right) \]

   7. Applied rewrites 36.8%

      \[\leadsto \mathsf{fma}\left(z, \color{blue}{x}, x\right) \]

   if -14.5 < z < 4.6e6

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in z around 0

      \[\leadsto \color{blue}{x + y \cdot \left(t - x\right)} \]
   3. Step-by-step derivation

      1. +-commutative N/A

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

      2. *-commutative N/A

         \[\leadsto \left(t - x\right) \cdot y + x \]

      3. lower-fma.f64 N/A

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

      4. lift--.f64 60.8

         \[\leadsto \mathsf{fma}\left(t - x, y, x\right) \]

   4. Applied rewrites 60.8%

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

   5. Taylor expanded in x around 0

      \[\leadsto \mathsf{fma}\left(t, y, x\right) \]
   6. Step-by-step derivation

   7. Applied rewrites 42.0%

      \[\leadsto \mathsf{fma}\left(t, y, x\right) \]

   if 4.05e242 < z

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in t around 0

      \[\leadsto \color{blue}{x + -1 \cdot \left(x \cdot \left(y - z\right)\right)} \]
   3. Step-by-step derivation

      1. +-commutative N/A

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

      2. associate-*r* N/A

         \[\leadsto \left(-1 \cdot x\right) \cdot \left(y - z\right) + x \]

      3. mul-1-neg N/A

         \[\leadsto \left(\mathsf{neg}\left(x\right)\right) \cdot \left(y - z\right) + x \]

      4. lower-fma.f64 N/A

         \[\leadsto \mathsf{fma}\left(\mathsf{neg}\left(x\right), \color{blue}{y - z}, x\right) \]

      5. lower-neg.f64 N/A

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

      6. lift--.f64 55.7

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

   4. Applied rewrites 55.7%

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

   5. Taylor expanded in z around inf

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

      1. *-commutative N/A

         \[\leadsto z \cdot x \]

      2. lower-*.f64 21.8

         \[\leadsto z \cdot x \]

   7. Applied rewrites 21.8%

      \[\leadsto z \cdot \color{blue}{x} \]
3. Recombined 4 regimes into one program.
4. Add Preprocessing

Alternative 10: 49.4% accurate, 0.9× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;y \leq -8.6 \cdot 10^{+66}:\\ \;\;\;\;t \cdot y\\ \mathbf{elif}\;y \leq 2.4 \cdot 10^{+29}:\\ \;\;\;\;\mathsf{fma}\left(z, x, x\right)\\ \mathbf{else}:\\ \;\;\;\;\left(-x\right) \cdot y\\ \end{array} \end{array} \]

FPCore

(FPCore (x y z t)
 :precision binary64
 (if (<= y -8.6e+66) (* t y) (if (<= y 2.4e+29) (fma z x x) (* (- x) y))))

C

double code(double x, double y, double z, double t) {
	double tmp;
	if (y <= -8.6e+66) {
		tmp = t * y;
	} else if (y <= 2.4e+29) {
		tmp = fma(z, x, x);
	} else {
		tmp = -x * y;
	}
	return tmp;
}

Julia

function code(x, y, z, t)
	tmp = 0.0
	if (y <= -8.6e+66)
		tmp = Float64(t * y);
	elseif (y <= 2.4e+29)
		tmp = fma(z, x, x);
	else
		tmp = Float64(Float64(-x) * y);
	end
	return tmp
end

Wolfram

code[x_, y_, z_, t_] := If[LessEqual[y, -8.6e+66], N[(t * y), $MachinePrecision], If[LessEqual[y, 2.4e+29], N[(z * x + x), $MachinePrecision], N[((-x) * y), $MachinePrecision]]]

Derivation

1. Split input into 3 regimes

2. if y < -8.60000000000000054e66

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in y around inf

      \[\leadsto \color{blue}{y \cdot \left(t - x\right)} \]
   3. Step-by-step derivation

      1. *-commutative N/A

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

      2. lower-*.f64 N/A

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

      3. lift--.f64 45.7

         \[\leadsto \left(t - x\right) \cdot y \]

   4. Applied rewrites 45.7%

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

   5. Taylor expanded in x around 0

      \[\leadsto t \cdot y \]
   6. Step-by-step derivation

   7. Applied rewrites 27.1%

      \[\leadsto t \cdot y \]

   if -8.60000000000000054e66 < y < 2.4000000000000001e29

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in t around 0

      \[\leadsto \color{blue}{x + -1 \cdot \left(x \cdot \left(y - z\right)\right)} \]
   3. Step-by-step derivation

      1. +-commutative N/A

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

      2. associate-*r* N/A

         \[\leadsto \left(-1 \cdot x\right) \cdot \left(y - z\right) + x \]

      3. mul-1-neg N/A

         \[\leadsto \left(\mathsf{neg}\left(x\right)\right) \cdot \left(y - z\right) + x \]

      4. lower-fma.f64 N/A

         \[\leadsto \mathsf{fma}\left(\mathsf{neg}\left(x\right), \color{blue}{y - z}, x\right) \]

      5. lower-neg.f64 N/A

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

      6. lift--.f64 55.7

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

   4. Applied rewrites 55.7%

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

   5. Taylor expanded in y around 0

      \[\leadsto x + \color{blue}{x \cdot z} \]
   6. Step-by-step derivation

      1. +-commutative N/A

         \[\leadsto x \cdot z + x \]

      2. *-commutative N/A

         \[\leadsto z \cdot x + x \]

      3. lower-fma.f64 36.8

         \[\leadsto \mathsf{fma}\left(z, x, x\right) \]

   7. Applied rewrites 36.8%

      \[\leadsto \mathsf{fma}\left(z, \color{blue}{x}, x\right) \]

   if 2.4000000000000001e29 < y

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in y around inf

      \[\leadsto \color{blue}{y \cdot \left(t - x\right)} \]
   3. Step-by-step derivation

      1. *-commutative N/A

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

      2. lower-*.f64 N/A

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

      3. lift--.f64 45.7

         \[\leadsto \left(t - x\right) \cdot y \]

   4. Applied rewrites 45.7%

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

   5. Taylor expanded in x around inf

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

      1. associate-*r* N/A

         \[\leadsto \left(-1 \cdot x\right) \cdot y \]

      2. lower-*.f64 N/A

         \[\leadsto \left(-1 \cdot x\right) \cdot y \]

      3. mul-1-neg N/A

         \[\leadsto \left(\mathsf{neg}\left(x\right)\right) \cdot y \]

      4. lift-neg.f64 23.2

         \[\leadsto \left(-x\right) \cdot y \]

   7. Applied rewrites 23.2%

      \[\leadsto \left(-x\right) \cdot \color{blue}{y} \]
3. Recombined 3 regimes into one program.
4. Add Preprocessing

Alternative 11: 39.4% accurate, 1.0× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;y \leq -7 \cdot 10^{-10}:\\ \;\;\;\;t \cdot y\\ \mathbf{elif}\;y \leq 3.2 \cdot 10^{+58}:\\ \;\;\;\;\left(-z\right) \cdot t\\ \mathbf{else}:\\ \;\;\;\;\left(-x\right) \cdot y\\ \end{array} \end{array} \]

FPCore

(FPCore (x y z t)
 :precision binary64
 (if (<= y -7e-10) (* t y) (if (<= y 3.2e+58) (* (- z) t) (* (- x) y))))

C

double code(double x, double y, double z, double t) {
	double tmp;
	if (y <= -7e-10) {
		tmp = t * y;
	} else if (y <= 3.2e+58) {
		tmp = -z * t;
	} else {
		tmp = -x * y;
	}
	return tmp;
}

Fortran

module fmin_fmax_functions
    implicit none
    private
    public fmax
    public fmin

    interface fmax
        module procedure fmax88
        module procedure fmax44
        module procedure fmax84
        module procedure fmax48
    end interface
    interface fmin
        module procedure fmin88
        module procedure fmin44
        module procedure fmin84
        module procedure fmin48
    end interface
contains
    real(8) function fmax88(x, y) result (res)
        real(8), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(x, max(x, y), y /= y), x /= x)
    end function
    real(4) function fmax44(x, y) result (res)
        real(4), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(y, merge(x, max(x, y), y /= y), x /= x)
    end function
    real(8) function fmax84(x, y) result(res)
        real(8), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(dble(y), merge(x, max(x, dble(y)), y /= y), x /= x)
    end function
    real(8) function fmax48(x, y) result(res)
        real(4), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(dble(x), max(dble(x), y), y /= y), x /= x)
    end function
    real(8) function fmin88(x, y) result (res)
        real(8), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(x, min(x, y), y /= y), x /= x)
    end function
    real(4) function fmin44(x, y) result (res)
        real(4), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(y, merge(x, min(x, y), y /= y), x /= x)
    end function
    real(8) function fmin84(x, y) result(res)
        real(8), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(dble(y), merge(x, min(x, dble(y)), y /= y), x /= x)
    end function
    real(8) function fmin48(x, y) result(res)
        real(4), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(dble(x), min(dble(x), y), y /= y), x /= x)
    end function
end module

real(8) function code(x, y, z, t)
use fmin_fmax_functions
    real(8), intent (in) :: x
    real(8), intent (in) :: y
    real(8), intent (in) :: z
    real(8), intent (in) :: t
    real(8) :: tmp
    if (y <= (-7d-10)) then
        tmp = t * y
    else if (y <= 3.2d+58) then
        tmp = -z * t
    else
        tmp = -x * y
    end if
    code = tmp
end function

Java

public static double code(double x, double y, double z, double t) {
	double tmp;
	if (y <= -7e-10) {
		tmp = t * y;
	} else if (y <= 3.2e+58) {
		tmp = -z * t;
	} else {
		tmp = -x * y;
	}
	return tmp;
}

Python

def code(x, y, z, t):
	tmp = 0
	if y <= -7e-10:
		tmp = t * y
	elif y <= 3.2e+58:
		tmp = -z * t
	else:
		tmp = -x * y
	return tmp

Julia

function code(x, y, z, t)
	tmp = 0.0
	if (y <= -7e-10)
		tmp = Float64(t * y);
	elseif (y <= 3.2e+58)
		tmp = Float64(Float64(-z) * t);
	else
		tmp = Float64(Float64(-x) * y);
	end
	return tmp
end

MATLAB

function tmp_2 = code(x, y, z, t)
	tmp = 0.0;
	if (y <= -7e-10)
		tmp = t * y;
	elseif (y <= 3.2e+58)
		tmp = -z * t;
	else
		tmp = -x * y;
	end
	tmp_2 = tmp;
end

Wolfram

code[x_, y_, z_, t_] := If[LessEqual[y, -7e-10], N[(t * y), $MachinePrecision], If[LessEqual[y, 3.2e+58], N[((-z) * t), $MachinePrecision], N[((-x) * y), $MachinePrecision]]]

Derivation

1. Split input into 3 regimes

2. if y < -6.99999999999999961e-10

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in y around inf

      \[\leadsto \color{blue}{y \cdot \left(t - x\right)} \]
   3. Step-by-step derivation

      1. *-commutative N/A

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

      2. lower-*.f64 N/A

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

      3. lift--.f64 45.7

         \[\leadsto \left(t - x\right) \cdot y \]

   4. Applied rewrites 45.7%

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

   5. Taylor expanded in x around 0

      \[\leadsto t \cdot y \]
   6. Step-by-step derivation

   7. Applied rewrites 27.1%

      \[\leadsto t \cdot y \]

   if -6.99999999999999961e-10 < y < 3.20000000000000015e58

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in x around 0

      \[\leadsto \color{blue}{t \cdot \left(y - z\right)} \]
   3. Step-by-step derivation

      1. *-commutative N/A

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

      2. lower-*.f64 N/A

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

      3. lift--.f64 49.6

         \[\leadsto \left(y - z\right) \cdot t \]

   4. Applied rewrites 49.6%

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

   5. Taylor expanded in y around 0

      \[\leadsto \left(-1 \cdot z\right) \cdot t \]
   6. Step-by-step derivation

      1. mul-1-neg N/A

         \[\leadsto \left(\mathsf{neg}\left(z\right)\right) \cdot t \]

      2. lower-neg.f64 26.9

         \[\leadsto \left(-z\right) \cdot t \]

   7. Applied rewrites 26.9%

      \[\leadsto \left(-z\right) \cdot t \]

   if 3.20000000000000015e58 < y

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in y around inf

      \[\leadsto \color{blue}{y \cdot \left(t - x\right)} \]
   3. Step-by-step derivation

      1. *-commutative N/A

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

      2. lower-*.f64 N/A

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

      3. lift--.f64 45.7

         \[\leadsto \left(t - x\right) \cdot y \]

   4. Applied rewrites 45.7%

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

   5. Taylor expanded in x around inf

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

      1. associate-*r* N/A

         \[\leadsto \left(-1 \cdot x\right) \cdot y \]

      2. lower-*.f64 N/A

         \[\leadsto \left(-1 \cdot x\right) \cdot y \]

      3. mul-1-neg N/A

         \[\leadsto \left(\mathsf{neg}\left(x\right)\right) \cdot y \]

      4. lift-neg.f64 23.2

         \[\leadsto \left(-x\right) \cdot y \]

   7. Applied rewrites 23.2%

      \[\leadsto \left(-x\right) \cdot \color{blue}{y} \]
3. Recombined 3 regimes into one program.
4. Add Preprocessing

Alternative 12: 39.2% accurate, 0.7× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;z \leq -1400000000000:\\ \;\;\;\;z \cdot x\\ \mathbf{elif}\;z \leq 2.9 \cdot 10^{-295}:\\ \;\;\;\;t \cdot y\\ \mathbf{elif}\;z \leq 2.5 \cdot 10^{+72}:\\ \;\;\;\;\left(-x\right) \cdot y\\ \mathbf{else}:\\ \;\;\;\;z \cdot x\\ \end{array} \end{array} \]

FPCore

(FPCore (x y z t)
 :precision binary64
 (if (<= z -1400000000000.0)
   (* z x)
   (if (<= z 2.9e-295) (* t y) (if (<= z 2.5e+72) (* (- x) y) (* z x)))))

C

double code(double x, double y, double z, double t) {
	double tmp;
	if (z <= -1400000000000.0) {
		tmp = z * x;
	} else if (z <= 2.9e-295) {
		tmp = t * y;
	} else if (z <= 2.5e+72) {
		tmp = -x * y;
	} else {
		tmp = z * x;
	}
	return tmp;
}

Fortran

module fmin_fmax_functions
    implicit none
    private
    public fmax
    public fmin

    interface fmax
        module procedure fmax88
        module procedure fmax44
        module procedure fmax84
        module procedure fmax48
    end interface
    interface fmin
        module procedure fmin88
        module procedure fmin44
        module procedure fmin84
        module procedure fmin48
    end interface
contains
    real(8) function fmax88(x, y) result (res)
        real(8), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(x, max(x, y), y /= y), x /= x)
    end function
    real(4) function fmax44(x, y) result (res)
        real(4), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(y, merge(x, max(x, y), y /= y), x /= x)
    end function
    real(8) function fmax84(x, y) result(res)
        real(8), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(dble(y), merge(x, max(x, dble(y)), y /= y), x /= x)
    end function
    real(8) function fmax48(x, y) result(res)
        real(4), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(dble(x), max(dble(x), y), y /= y), x /= x)
    end function
    real(8) function fmin88(x, y) result (res)
        real(8), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(x, min(x, y), y /= y), x /= x)
    end function
    real(4) function fmin44(x, y) result (res)
        real(4), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(y, merge(x, min(x, y), y /= y), x /= x)
    end function
    real(8) function fmin84(x, y) result(res)
        real(8), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(dble(y), merge(x, min(x, dble(y)), y /= y), x /= x)
    end function
    real(8) function fmin48(x, y) result(res)
        real(4), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(dble(x), min(dble(x), y), y /= y), x /= x)
    end function
end module

real(8) function code(x, y, z, t)
use fmin_fmax_functions
    real(8), intent (in) :: x
    real(8), intent (in) :: y
    real(8), intent (in) :: z
    real(8), intent (in) :: t
    real(8) :: tmp
    if (z <= (-1400000000000.0d0)) then
        tmp = z * x
    else if (z <= 2.9d-295) then
        tmp = t * y
    else if (z <= 2.5d+72) then
        tmp = -x * y
    else
        tmp = z * x
    end if
    code = tmp
end function

Java

public static double code(double x, double y, double z, double t) {
	double tmp;
	if (z <= -1400000000000.0) {
		tmp = z * x;
	} else if (z <= 2.9e-295) {
		tmp = t * y;
	} else if (z <= 2.5e+72) {
		tmp = -x * y;
	} else {
		tmp = z * x;
	}
	return tmp;
}

Python

def code(x, y, z, t):
	tmp = 0
	if z <= -1400000000000.0:
		tmp = z * x
	elif z <= 2.9e-295:
		tmp = t * y
	elif z <= 2.5e+72:
		tmp = -x * y
	else:
		tmp = z * x
	return tmp

Julia

function code(x, y, z, t)
	tmp = 0.0
	if (z <= -1400000000000.0)
		tmp = Float64(z * x);
	elseif (z <= 2.9e-295)
		tmp = Float64(t * y);
	elseif (z <= 2.5e+72)
		tmp = Float64(Float64(-x) * y);
	else
		tmp = Float64(z * x);
	end
	return tmp
end

MATLAB

function tmp_2 = code(x, y, z, t)
	tmp = 0.0;
	if (z <= -1400000000000.0)
		tmp = z * x;
	elseif (z <= 2.9e-295)
		tmp = t * y;
	elseif (z <= 2.5e+72)
		tmp = -x * y;
	else
		tmp = z * x;
	end
	tmp_2 = tmp;
end

Wolfram

code[x_, y_, z_, t_] := If[LessEqual[z, -1400000000000.0], N[(z * x), $MachinePrecision], If[LessEqual[z, 2.9e-295], N[(t * y), $MachinePrecision], If[LessEqual[z, 2.5e+72], N[((-x) * y), $MachinePrecision], N[(z * x), $MachinePrecision]]]]

Derivation

1. Split input into 3 regimes

2. if z < -1.4e12 or 2.49999999999999996e72 < z

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in t around 0

      \[\leadsto \color{blue}{x + -1 \cdot \left(x \cdot \left(y - z\right)\right)} \]
   3. Step-by-step derivation

      1. +-commutative N/A

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

      2. associate-*r* N/A

         \[\leadsto \left(-1 \cdot x\right) \cdot \left(y - z\right) + x \]

      3. mul-1-neg N/A

         \[\leadsto \left(\mathsf{neg}\left(x\right)\right) \cdot \left(y - z\right) + x \]

      4. lower-fma.f64 N/A

         \[\leadsto \mathsf{fma}\left(\mathsf{neg}\left(x\right), \color{blue}{y - z}, x\right) \]

      5. lower-neg.f64 N/A

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

      6. lift--.f64 55.7

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

   4. Applied rewrites 55.7%

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

   5. Taylor expanded in z around inf

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

      1. *-commutative N/A

         \[\leadsto z \cdot x \]

      2. lower-*.f64 21.8

         \[\leadsto z \cdot x \]

   7. Applied rewrites 21.8%

      \[\leadsto z \cdot \color{blue}{x} \]

   if -1.4e12 < z < 2.90000000000000015e-295

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in y around inf

      \[\leadsto \color{blue}{y \cdot \left(t - x\right)} \]
   3. Step-by-step derivation

      1. *-commutative N/A

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

      2. lower-*.f64 N/A

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

      3. lift--.f64 45.7

         \[\leadsto \left(t - x\right) \cdot y \]

   4. Applied rewrites 45.7%

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

   5. Taylor expanded in x around 0

      \[\leadsto t \cdot y \]
   6. Step-by-step derivation

   7. Applied rewrites 27.1%

      \[\leadsto t \cdot y \]

   if 2.90000000000000015e-295 < z < 2.49999999999999996e72

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in y around inf

      \[\leadsto \color{blue}{y \cdot \left(t - x\right)} \]
   3. Step-by-step derivation

      1. *-commutative N/A

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

      2. lower-*.f64 N/A

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

      3. lift--.f64 45.7

         \[\leadsto \left(t - x\right) \cdot y \]

   4. Applied rewrites 45.7%

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

   5. Taylor expanded in x around inf

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

      1. associate-*r* N/A

         \[\leadsto \left(-1 \cdot x\right) \cdot y \]

      2. lower-*.f64 N/A

         \[\leadsto \left(-1 \cdot x\right) \cdot y \]

      3. mul-1-neg N/A

         \[\leadsto \left(\mathsf{neg}\left(x\right)\right) \cdot y \]

      4. lift-neg.f64 23.2

         \[\leadsto \left(-x\right) \cdot y \]

   7. Applied rewrites 23.2%

      \[\leadsto \left(-x\right) \cdot \color{blue}{y} \]
3. Recombined 3 regimes into one program.
4. Add Preprocessing

Alternative 13: 37.4% accurate, 1.0× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;z \leq -1400000000000:\\ \;\;\;\;z \cdot x\\ \mathbf{elif}\;z \leq 7.6 \cdot 10^{+48}:\\ \;\;\;\;t \cdot y\\ \mathbf{else}:\\ \;\;\;\;z \cdot x\\ \end{array} \end{array} \]

FPCore

(FPCore (x y z t)
 :precision binary64
 (if (<= z -1400000000000.0) (* z x) (if (<= z 7.6e+48) (* t y) (* z x))))

C

double code(double x, double y, double z, double t) {
	double tmp;
	if (z <= -1400000000000.0) {
		tmp = z * x;
	} else if (z <= 7.6e+48) {
		tmp = t * y;
	} else {
		tmp = z * x;
	}
	return tmp;
}

Fortran

module fmin_fmax_functions
    implicit none
    private
    public fmax
    public fmin

    interface fmax
        module procedure fmax88
        module procedure fmax44
        module procedure fmax84
        module procedure fmax48
    end interface
    interface fmin
        module procedure fmin88
        module procedure fmin44
        module procedure fmin84
        module procedure fmin48
    end interface
contains
    real(8) function fmax88(x, y) result (res)
        real(8), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(x, max(x, y), y /= y), x /= x)
    end function
    real(4) function fmax44(x, y) result (res)
        real(4), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(y, merge(x, max(x, y), y /= y), x /= x)
    end function
    real(8) function fmax84(x, y) result(res)
        real(8), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(dble(y), merge(x, max(x, dble(y)), y /= y), x /= x)
    end function
    real(8) function fmax48(x, y) result(res)
        real(4), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(dble(x), max(dble(x), y), y /= y), x /= x)
    end function
    real(8) function fmin88(x, y) result (res)
        real(8), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(x, min(x, y), y /= y), x /= x)
    end function
    real(4) function fmin44(x, y) result (res)
        real(4), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(y, merge(x, min(x, y), y /= y), x /= x)
    end function
    real(8) function fmin84(x, y) result(res)
        real(8), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(dble(y), merge(x, min(x, dble(y)), y /= y), x /= x)
    end function
    real(8) function fmin48(x, y) result(res)
        real(4), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(dble(x), min(dble(x), y), y /= y), x /= x)
    end function
end module

real(8) function code(x, y, z, t)
use fmin_fmax_functions
    real(8), intent (in) :: x
    real(8), intent (in) :: y
    real(8), intent (in) :: z
    real(8), intent (in) :: t
    real(8) :: tmp
    if (z <= (-1400000000000.0d0)) then
        tmp = z * x
    else if (z <= 7.6d+48) then
        tmp = t * y
    else
        tmp = z * x
    end if
    code = tmp
end function

Java

public static double code(double x, double y, double z, double t) {
	double tmp;
	if (z <= -1400000000000.0) {
		tmp = z * x;
	} else if (z <= 7.6e+48) {
		tmp = t * y;
	} else {
		tmp = z * x;
	}
	return tmp;
}

Python

def code(x, y, z, t):
	tmp = 0
	if z <= -1400000000000.0:
		tmp = z * x
	elif z <= 7.6e+48:
		tmp = t * y
	else:
		tmp = z * x
	return tmp

Julia

function code(x, y, z, t)
	tmp = 0.0
	if (z <= -1400000000000.0)
		tmp = Float64(z * x);
	elseif (z <= 7.6e+48)
		tmp = Float64(t * y);
	else
		tmp = Float64(z * x);
	end
	return tmp
end

MATLAB

function tmp_2 = code(x, y, z, t)
	tmp = 0.0;
	if (z <= -1400000000000.0)
		tmp = z * x;
	elseif (z <= 7.6e+48)
		tmp = t * y;
	else
		tmp = z * x;
	end
	tmp_2 = tmp;
end

Wolfram

code[x_, y_, z_, t_] := If[LessEqual[z, -1400000000000.0], N[(z * x), $MachinePrecision], If[LessEqual[z, 7.6e+48], N[(t * y), $MachinePrecision], N[(z * x), $MachinePrecision]]]

Derivation

1. Split input into 2 regimes

2. if z < -1.4e12 or 7.60000000000000001e48 < z

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in t around 0

      \[\leadsto \color{blue}{x + -1 \cdot \left(x \cdot \left(y - z\right)\right)} \]
   3. Step-by-step derivation

      1. +-commutative N/A

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

      2. associate-*r* N/A

         \[\leadsto \left(-1 \cdot x\right) \cdot \left(y - z\right) + x \]

      3. mul-1-neg N/A

         \[\leadsto \left(\mathsf{neg}\left(x\right)\right) \cdot \left(y - z\right) + x \]

      4. lower-fma.f64 N/A

         \[\leadsto \mathsf{fma}\left(\mathsf{neg}\left(x\right), \color{blue}{y - z}, x\right) \]

      5. lower-neg.f64 N/A

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

      6. lift--.f64 55.7

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

   4. Applied rewrites 55.7%

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

   5. Taylor expanded in z around inf

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

      1. *-commutative N/A

         \[\leadsto z \cdot x \]

      2. lower-*.f64 21.8

         \[\leadsto z \cdot x \]

   7. Applied rewrites 21.8%

      \[\leadsto z \cdot \color{blue}{x} \]

   if -1.4e12 < z < 7.60000000000000001e48

   1. Initial program 100.0%

      \[x + \left(y - z\right) \cdot \left(t - x\right) \]

   2. Taylor expanded in y around inf

      \[\leadsto \color{blue}{y \cdot \left(t - x\right)} \]
   3. Step-by-step derivation

      1. *-commutative N/A

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

      2. lower-*.f64 N/A

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

      3. lift--.f64 45.7

         \[\leadsto \left(t - x\right) \cdot y \]

   4. Applied rewrites 45.7%

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

   5. Taylor expanded in x around 0

      \[\leadsto t \cdot y \]
   6. Step-by-step derivation

   7. Applied rewrites 27.1%

      \[\leadsto t \cdot y \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 14: 21.8% accurate, 3.0× speedup

Math

\[\begin{array}{l} \\ z \cdot x \end{array} \]

FPCore

(FPCore (x y z t) :precision binary64 (* z x))

C

double code(double x, double y, double z, double t) {
	return z * x;
}

Fortran

module fmin_fmax_functions
    implicit none
    private
    public fmax
    public fmin

    interface fmax
        module procedure fmax88
        module procedure fmax44
        module procedure fmax84
        module procedure fmax48
    end interface
    interface fmin
        module procedure fmin88
        module procedure fmin44
        module procedure fmin84
        module procedure fmin48
    end interface
contains
    real(8) function fmax88(x, y) result (res)
        real(8), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(x, max(x, y), y /= y), x /= x)
    end function
    real(4) function fmax44(x, y) result (res)
        real(4), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(y, merge(x, max(x, y), y /= y), x /= x)
    end function
    real(8) function fmax84(x, y) result(res)
        real(8), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(dble(y), merge(x, max(x, dble(y)), y /= y), x /= x)
    end function
    real(8) function fmax48(x, y) result(res)
        real(4), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(dble(x), max(dble(x), y), y /= y), x /= x)
    end function
    real(8) function fmin88(x, y) result (res)
        real(8), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(x, min(x, y), y /= y), x /= x)
    end function
    real(4) function fmin44(x, y) result (res)
        real(4), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(y, merge(x, min(x, y), y /= y), x /= x)
    end function
    real(8) function fmin84(x, y) result(res)
        real(8), intent (in) :: x
        real(4), intent (in) :: y
        res = merge(dble(y), merge(x, min(x, dble(y)), y /= y), x /= x)
    end function
    real(8) function fmin48(x, y) result(res)
        real(4), intent (in) :: x
        real(8), intent (in) :: y
        res = merge(y, merge(dble(x), min(dble(x), y), y /= y), x /= x)
    end function
end module

real(8) function code(x, y, z, t)
use fmin_fmax_functions
    real(8), intent (in) :: x
    real(8), intent (in) :: y
    real(8), intent (in) :: z
    real(8), intent (in) :: t
    code = z * x
end function

Java

public static double code(double x, double y, double z, double t) {
	return z * x;
}

Python

def code(x, y, z, t):
	return z * x

Julia

function code(x, y, z, t)
	return Float64(z * x)
end

MATLAB

function tmp = code(x, y, z, t)
	tmp = z * x;
end

Wolfram

code[x_, y_, z_, t_] := N[(z * x), $MachinePrecision]

Derivation

1. Initial program 100.0%

   \[x + \left(y - z\right) \cdot \left(t - x\right) \]

2. Taylor expanded in t around 0

   \[\leadsto \color{blue}{x + -1 \cdot \left(x \cdot \left(y - z\right)\right)} \]
3. Step-by-step derivation

   1. +-commutative N/A

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

   2. associate-*r* N/A

      \[\leadsto \left(-1 \cdot x\right) \cdot \left(y - z\right) + x \]

   3. mul-1-neg N/A

      \[\leadsto \left(\mathsf{neg}\left(x\right)\right) \cdot \left(y - z\right) + x \]

   4. lower-fma.f64 N/A

      \[\leadsto \mathsf{fma}\left(\mathsf{neg}\left(x\right), \color{blue}{y - z}, x\right) \]

   5. lower-neg.f64 N/A

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

   6. lift--.f64 55.7

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

4. Applied rewrites 55.7%

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

5. Taylor expanded in z around inf

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

   1. *-commutative N/A

      \[\leadsto z \cdot x \]

   2. lower-*.f64 21.8

      \[\leadsto z \cdot x \]

7. Applied rewrites 21.8%

   \[\leadsto z \cdot \color{blue}{x} \]
8. Add Preprocessing

Reproduce

herbie shell --seed 2025136 
(FPCore (x y z t)
  :name "Data.Metrics.Snapshot:quantile from metrics-0.3.0.2"
  :precision binary64
  (+ x (* (- y z) (- t x))))