Data.Random.Distribution.Triangular:triangularCDF from random-fu-0.2.6.2, B

Percentage Accurate: 88.9% → 98.2%

Time: 3.2s

Alternatives: 10

Speedup: 1.0×

Specification

Math

\[\frac{x}{\left(y - z\right) \cdot \left(t - z\right)} \]

FPCore

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

C

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

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 - z))
end function

Java

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

Python

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

Julia

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

MATLAB

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

Wolfram

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

Initial Program: 88.9% accurate, 1.0× speedup

Math

\[\frac{x}{\left(y - z\right) \cdot \left(t - z\right)} \]

FPCore

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

C

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

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 - z))
end function

Java

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

Python

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

Julia

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

MATLAB

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

Wolfram

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

Alternative 1: 98.2% accurate, 0.0× speedup

Math

\[\begin{array}{l} t_1 := \frac{\left|x\right|}{\left(\mathsf{min}\left(y, t\right) - z\right) \cdot \left(\mathsf{max}\left(y, t\right) - z\right)}\\ \mathsf{copysign}\left(1, x\right) \cdot \begin{array}{l} \mathbf{if}\;t\_1 \leq -1 \cdot 10^{-310}:\\ \;\;\;\;t\_1\\ \mathbf{else}:\\ \;\;\;\;\frac{\frac{\left|x\right|}{z - \mathsf{min}\left(y, t\right)}}{z - \mathsf{max}\left(y, t\right)}\\ \end{array} \end{array} \]

FPCore

(FPCore (x y z t)
  :precision binary64
  (let* ((t_1 (/ (fabs x) (* (- (fmin y t) z) (- (fmax y t) z)))))
  (*
   (copysign 1.0 x)
   (if (<= t_1 -1e-310)
     t_1
     (/ (/ (fabs x) (- z (fmin y t))) (- z (fmax y t)))))))

C

double code(double x, double y, double z, double t) {
	double t_1 = fabs(x) / ((fmin(y, t) - z) * (fmax(y, t) - z));
	double tmp;
	if (t_1 <= -1e-310) {
		tmp = t_1;
	} else {
		tmp = (fabs(x) / (z - fmin(y, t))) / (z - fmax(y, t));
	}
	return copysign(1.0, x) * tmp;
}

Java

public static double code(double x, double y, double z, double t) {
	double t_1 = Math.abs(x) / ((fmin(y, t) - z) * (fmax(y, t) - z));
	double tmp;
	if (t_1 <= -1e-310) {
		tmp = t_1;
	} else {
		tmp = (Math.abs(x) / (z - fmin(y, t))) / (z - fmax(y, t));
	}
	return Math.copySign(1.0, x) * tmp;
}

Python

def code(x, y, z, t):
	t_1 = math.fabs(x) / ((fmin(y, t) - z) * (fmax(y, t) - z))
	tmp = 0
	if t_1 <= -1e-310:
		tmp = t_1
	else:
		tmp = (math.fabs(x) / (z - fmin(y, t))) / (z - fmax(y, t))
	return math.copysign(1.0, x) * tmp

Julia

function code(x, y, z, t)
	t_1 = Float64(abs(x) / Float64(Float64(fmin(y, t) - z) * Float64(fmax(y, t) - z)))
	tmp = 0.0
	if (t_1 <= -1e-310)
		tmp = t_1;
	else
		tmp = Float64(Float64(abs(x) / Float64(z - fmin(y, t))) / Float64(z - fmax(y, t)));
	end
	return Float64(copysign(1.0, x) * tmp)
end

MATLAB

function tmp_2 = code(x, y, z, t)
	t_1 = abs(x) / ((min(y, t) - z) * (max(y, t) - z));
	tmp = 0.0;
	if (t_1 <= -1e-310)
		tmp = t_1;
	else
		tmp = (abs(x) / (z - min(y, t))) / (z - max(y, t));
	end
	tmp_2 = (sign(x) * abs(1.0)) * tmp;
end

Wolfram

code[x_, y_, z_, t_] := Block[{t$95$1 = N[(N[Abs[x], $MachinePrecision] / N[(N[(N[Min[y, t], $MachinePrecision] - z), $MachinePrecision] * N[(N[Max[y, t], $MachinePrecision] - z), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]}, N[(N[With[{TMP1 = Abs[1.0], TMP2 = Sign[x]}, TMP1 * If[TMP2 == 0, 1, TMP2]], $MachinePrecision] * If[LessEqual[t$95$1, -1e-310], t$95$1, N[(N[(N[Abs[x], $MachinePrecision] / N[(z - N[Min[y, t], $MachinePrecision]), $MachinePrecision]), $MachinePrecision] / N[(z - N[Max[y, t], $MachinePrecision]), $MachinePrecision]), $MachinePrecision]]), $MachinePrecision]]

Derivation

1. Split input into 2 regimes

2. if (/.f64 x (*.f64 (-.f64 y z) (-.f64 t z))) < -9.9999999999999694e-311

   1. Initial program 88.9%

      \[\frac{x}{\left(y - z\right) \cdot \left(t - z\right)} \]

   if -9.9999999999999694e-311 < (/.f64 x (*.f64 (-.f64 y z) (-.f64 t z)))

   1. Initial program 88.9%

      \[\frac{x}{\left(y - z\right) \cdot \left(t - z\right)} \]
   2. Step-by-step derivation

      1. lift-/.f64 N/A

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

      2. lift-*.f64 N/A

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

      3. associate-/r* N/A

         \[\leadsto \color{blue}{\frac{\frac{x}{y - z}}{t - z}} \]

      4. frac-2neg N/A

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

      5. lift--.f64 N/A

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

      6. sub-negate-rev N/A

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

      7. lower-/.f64 N/A

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

      8. distribute-neg-frac2 N/A

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

      9. lift--.f64 N/A

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

      10. sub-negate-rev N/A

          \[\leadsto \frac{\frac{x}{\color{blue}{z - y}}}{z - t} \]

      11. lower-/.f64 N/A

          \[\leadsto \frac{\color{blue}{\frac{x}{z - y}}}{z - t} \]

      12. lower--.f64 N/A

          \[\leadsto \frac{\frac{x}{\color{blue}{z - y}}}{z - t} \]

      13. lower--.f64 97.1%

          \[\leadsto \frac{\frac{x}{z - y}}{\color{blue}{z - t}} \]

   3. Applied rewrites 97.1%

      \[\leadsto \color{blue}{\frac{\frac{x}{z - y}}{z - t}} \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 2: 97.2% accurate, 0.8× speedup

Math

\[\frac{\frac{x}{z - t}}{z - y} \]

FPCore

(FPCore (x y z t)
  :precision binary64
  (/ (/ x (- z t)) (- z y)))

C

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

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 / (z - t)) / (z - y)
end function

Java

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

Python

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

Julia

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

MATLAB

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

Wolfram

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

Derivation

1. Initial program 88.9%

   \[\frac{x}{\left(y - z\right) \cdot \left(t - z\right)} \]
2. Step-by-step derivation

   1. lift-/.f64 N/A

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

   2. lift-*.f64 N/A

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

   3. *-commutative N/A

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

   4. associate-/r* N/A

      \[\leadsto \color{blue}{\frac{\frac{x}{t - z}}{y - z}} \]

   5. lift--.f64 N/A

      \[\leadsto \frac{\frac{x}{\color{blue}{t - z}}}{y - z} \]

   6. sub-negate-rev N/A

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

   7. distribute-frac-neg2 N/A

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

   8. lift--.f64 N/A

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

   9. sub-negate-rev N/A

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

   10. frac-2neg-rev N/A

       \[\leadsto \color{blue}{\frac{\frac{x}{z - t}}{z - y}} \]

   11. lower-/.f64 N/A

       \[\leadsto \color{blue}{\frac{\frac{x}{z - t}}{z - y}} \]

   12. lower-/.f64 N/A

       \[\leadsto \frac{\color{blue}{\frac{x}{z - t}}}{z - y} \]

   13. lower--.f64 N/A

       \[\leadsto \frac{\frac{x}{\color{blue}{z - t}}}{z - y} \]

   14. lower--.f64 97.2%

       \[\leadsto \frac{\frac{x}{z - t}}{\color{blue}{z - y}} \]

3. Applied rewrites 97.2%

   \[\leadsto \color{blue}{\frac{\frac{x}{z - t}}{z - y}} \]
4. Add Preprocessing

Alternative 3: 93.5% accurate, 0.1× speedup

Math

\[\begin{array}{l} t_1 := \frac{\frac{x}{z}}{z - \mathsf{min}\left(y, t\right)}\\ \mathbf{if}\;z \leq -7.4 \cdot 10^{+127}:\\ \;\;\;\;t\_1\\ \mathbf{elif}\;z \leq 4.2 \cdot 10^{+126}:\\ \;\;\;\;\frac{x}{\left(\mathsf{min}\left(y, t\right) - z\right) \cdot \left(\mathsf{max}\left(y, t\right) - z\right)}\\ \mathbf{else}:\\ \;\;\;\;t\_1\\ \end{array} \]

FPCore

(FPCore (x y z t)
  :precision binary64
  (let* ((t_1 (/ (/ x z) (- z (fmin y t)))))
  (if (<= z -7.4e+127)
    t_1
    (if (<= z 4.2e+126)
      (/ x (* (- (fmin y t) z) (- (fmax y t) z)))
      t_1))))

C

double code(double x, double y, double z, double t) {
	double t_1 = (x / z) / (z - fmin(y, t));
	double tmp;
	if (z <= -7.4e+127) {
		tmp = t_1;
	} else if (z <= 4.2e+126) {
		tmp = x / ((fmin(y, t) - z) * (fmax(y, t) - z));
	} else {
		tmp = t_1;
	}
	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) :: t_1
    real(8) :: tmp
    t_1 = (x / z) / (z - fmin(y, t))
    if (z <= (-7.4d+127)) then
        tmp = t_1
    else if (z <= 4.2d+126) then
        tmp = x / ((fmin(y, t) - z) * (fmax(y, t) - z))
    else
        tmp = t_1
    end if
    code = tmp
end function

Java

public static double code(double x, double y, double z, double t) {
	double t_1 = (x / z) / (z - fmin(y, t));
	double tmp;
	if (z <= -7.4e+127) {
		tmp = t_1;
	} else if (z <= 4.2e+126) {
		tmp = x / ((fmin(y, t) - z) * (fmax(y, t) - z));
	} else {
		tmp = t_1;
	}
	return tmp;
}

Python

def code(x, y, z, t):
	t_1 = (x / z) / (z - fmin(y, t))
	tmp = 0
	if z <= -7.4e+127:
		tmp = t_1
	elif z <= 4.2e+126:
		tmp = x / ((fmin(y, t) - z) * (fmax(y, t) - z))
	else:
		tmp = t_1
	return tmp

Julia

function code(x, y, z, t)
	t_1 = Float64(Float64(x / z) / Float64(z - fmin(y, t)))
	tmp = 0.0
	if (z <= -7.4e+127)
		tmp = t_1;
	elseif (z <= 4.2e+126)
		tmp = Float64(x / Float64(Float64(fmin(y, t) - z) * Float64(fmax(y, t) - z)));
	else
		tmp = t_1;
	end
	return tmp
end

MATLAB

function tmp_2 = code(x, y, z, t)
	t_1 = (x / z) / (z - min(y, t));
	tmp = 0.0;
	if (z <= -7.4e+127)
		tmp = t_1;
	elseif (z <= 4.2e+126)
		tmp = x / ((min(y, t) - z) * (max(y, t) - z));
	else
		tmp = t_1;
	end
	tmp_2 = tmp;
end

Wolfram

code[x_, y_, z_, t_] := Block[{t$95$1 = N[(N[(x / z), $MachinePrecision] / N[(z - N[Min[y, t], $MachinePrecision]), $MachinePrecision]), $MachinePrecision]}, If[LessEqual[z, -7.4e+127], t$95$1, If[LessEqual[z, 4.2e+126], N[(x / N[(N[(N[Min[y, t], $MachinePrecision] - z), $MachinePrecision] * N[(N[Max[y, t], $MachinePrecision] - z), $MachinePrecision]), $MachinePrecision]), $MachinePrecision], t$95$1]]]

Derivation

1. Split input into 2 regimes

2. if z < -7.3999999999999996e127 or 4.1999999999999998e126 < z

   1. Initial program 88.9%

      \[\frac{x}{\left(y - z\right) \cdot \left(t - z\right)} \]
   2. Step-by-step derivation

      1. lift-/.f64 N/A

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

      2. lift-*.f64 N/A

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

      3. *-commutative N/A

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

      4. associate-/r* N/A

         \[\leadsto \color{blue}{\frac{\frac{x}{t - z}}{y - z}} \]

      5. lift--.f64 N/A

         \[\leadsto \frac{\frac{x}{\color{blue}{t - z}}}{y - z} \]

      6. sub-negate-rev N/A

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

      7. distribute-frac-neg2 N/A

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

      8. lift--.f64 N/A

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

      9. sub-negate-rev N/A

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

      10. frac-2neg-rev N/A

          \[\leadsto \color{blue}{\frac{\frac{x}{z - t}}{z - y}} \]

      11. lower-/.f64 N/A

          \[\leadsto \color{blue}{\frac{\frac{x}{z - t}}{z - y}} \]

      12. lower-/.f64 N/A

          \[\leadsto \frac{\color{blue}{\frac{x}{z - t}}}{z - y} \]

      13. lower--.f64 N/A

          \[\leadsto \frac{\frac{x}{\color{blue}{z - t}}}{z - y} \]

      14. lower--.f64 97.2%

          \[\leadsto \frac{\frac{x}{z - t}}{\color{blue}{z - y}} \]

   3. Applied rewrites 97.2%

      \[\leadsto \color{blue}{\frac{\frac{x}{z - t}}{z - y}} \]
   4. Step-by-step derivation

      1. *-rgt-identity N/A

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

      2. metadata-eval N/A

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

      3. lower-/.f64 N/A

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

      4. mult-flip N/A

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

      5. *-commutative N/A

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

      6. lower-*.f64 N/A

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

      7. metadata-eval N/A

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

      8. lower-/.f64 N/A

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

      9. metadata-eval N/A

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

      10. metadata-eval N/A

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

      11. *-rgt-identity 97.1%

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

   5. Applied rewrites 97.1%

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

   6. Taylor expanded in z around inf

      \[\leadsto \frac{\color{blue}{\frac{x}{z}}}{z - y} \]
   7. Step-by-step derivation

      1. lower-/.f64 58.2%

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

   8. Applied rewrites 58.2%

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

   if -7.3999999999999996e127 < z < 4.1999999999999998e126

   1. Initial program 88.9%

      \[\frac{x}{\left(y - z\right) \cdot \left(t - z\right)} \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 4: 92.0% accurate, 0.7× speedup

Math

\[\begin{array}{l} t_1 := \frac{\frac{x}{z}}{z}\\ \mathbf{if}\;z \leq -3.9 \cdot 10^{+130}:\\ \;\;\;\;t\_1\\ \mathbf{elif}\;z \leq 10^{+130}:\\ \;\;\;\;\frac{x}{\left(y - z\right) \cdot \left(t - z\right)}\\ \mathbf{else}:\\ \;\;\;\;t\_1\\ \end{array} \]

FPCore

(FPCore (x y z t)
  :precision binary64
  (let* ((t_1 (/ (/ x z) z)))
  (if (<= z -3.9e+130)
    t_1
    (if (<= z 1e+130) (/ x (* (- y z) (- t z))) t_1))))

C

double code(double x, double y, double z, double t) {
	double t_1 = (x / z) / z;
	double tmp;
	if (z <= -3.9e+130) {
		tmp = t_1;
	} else if (z <= 1e+130) {
		tmp = x / ((y - z) * (t - z));
	} else {
		tmp = t_1;
	}
	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) :: t_1
    real(8) :: tmp
    t_1 = (x / z) / z
    if (z <= (-3.9d+130)) then
        tmp = t_1
    else if (z <= 1d+130) then
        tmp = x / ((y - z) * (t - z))
    else
        tmp = t_1
    end if
    code = tmp
end function

Java

public static double code(double x, double y, double z, double t) {
	double t_1 = (x / z) / z;
	double tmp;
	if (z <= -3.9e+130) {
		tmp = t_1;
	} else if (z <= 1e+130) {
		tmp = x / ((y - z) * (t - z));
	} else {
		tmp = t_1;
	}
	return tmp;
}

Python

def code(x, y, z, t):
	t_1 = (x / z) / z
	tmp = 0
	if z <= -3.9e+130:
		tmp = t_1
	elif z <= 1e+130:
		tmp = x / ((y - z) * (t - z))
	else:
		tmp = t_1
	return tmp

Julia

function code(x, y, z, t)
	t_1 = Float64(Float64(x / z) / z)
	tmp = 0.0
	if (z <= -3.9e+130)
		tmp = t_1;
	elseif (z <= 1e+130)
		tmp = Float64(x / Float64(Float64(y - z) * Float64(t - z)));
	else
		tmp = t_1;
	end
	return tmp
end

MATLAB

function tmp_2 = code(x, y, z, t)
	t_1 = (x / z) / z;
	tmp = 0.0;
	if (z <= -3.9e+130)
		tmp = t_1;
	elseif (z <= 1e+130)
		tmp = x / ((y - z) * (t - z));
	else
		tmp = t_1;
	end
	tmp_2 = tmp;
end

Wolfram

code[x_, y_, z_, t_] := Block[{t$95$1 = N[(N[(x / z), $MachinePrecision] / z), $MachinePrecision]}, If[LessEqual[z, -3.9e+130], t$95$1, If[LessEqual[z, 1e+130], N[(x / N[(N[(y - z), $MachinePrecision] * N[(t - z), $MachinePrecision]), $MachinePrecision]), $MachinePrecision], t$95$1]]]

Derivation

1. Split input into 2 regimes

2. if z < -3.9000000000000002e130 or 1.0000000000000001e130 < z

   1. Initial program 88.9%

      \[\frac{x}{\left(y - z\right) \cdot \left(t - z\right)} \]
   2. Step-by-step derivation

      1. lift-/.f64 N/A

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

      2. lift-*.f64 N/A

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

      3. *-commutative N/A

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

      4. associate-/r* N/A

         \[\leadsto \color{blue}{\frac{\frac{x}{t - z}}{y - z}} \]

      5. lift--.f64 N/A

         \[\leadsto \frac{\frac{x}{\color{blue}{t - z}}}{y - z} \]

      6. sub-negate-rev N/A

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

      7. distribute-frac-neg2 N/A

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

      8. lift--.f64 N/A

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

      9. sub-negate-rev N/A

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

      10. frac-2neg-rev N/A

          \[\leadsto \color{blue}{\frac{\frac{x}{z - t}}{z - y}} \]

      11. lower-/.f64 N/A

          \[\leadsto \color{blue}{\frac{\frac{x}{z - t}}{z - y}} \]

      12. lower-/.f64 N/A

          \[\leadsto \frac{\color{blue}{\frac{x}{z - t}}}{z - y} \]

      13. lower--.f64 N/A

          \[\leadsto \frac{\frac{x}{\color{blue}{z - t}}}{z - y} \]

      14. lower--.f64 97.2%

          \[\leadsto \frac{\frac{x}{z - t}}{\color{blue}{z - y}} \]

   3. Applied rewrites 97.2%

      \[\leadsto \color{blue}{\frac{\frac{x}{z - t}}{z - y}} \]
   4. Step-by-step derivation

      1. *-rgt-identity N/A

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

      2. metadata-eval N/A

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

      3. lower-/.f64 N/A

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

      4. mult-flip N/A

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

      5. *-commutative N/A

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

      6. lower-*.f64 N/A

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

      7. metadata-eval N/A

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

      8. lower-/.f64 N/A

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

      9. metadata-eval N/A

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

      10. metadata-eval N/A

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

      11. *-rgt-identity 97.1%

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

   5. Applied rewrites 97.1%

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

   6. Taylor expanded in z around inf

      \[\leadsto \frac{\color{blue}{\frac{x}{z}}}{z - y} \]
   7. Step-by-step derivation

      1. lower-/.f64 58.2%

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

   8. Applied rewrites 58.2%

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

   9. Taylor expanded in y around 0

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

   11. Applied rewrites 44.1%

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

   if -3.9000000000000002e130 < z < 1.0000000000000001e130

   1. Initial program 88.9%

      \[\frac{x}{\left(y - z\right) \cdot \left(t - z\right)} \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 5: 77.6% accurate, 0.1× speedup

Math

\[\begin{array}{l} \mathbf{if}\;\mathsf{min}\left(y, t\right) \leq -2.9:\\ \;\;\;\;\frac{x}{\mathsf{min}\left(y, t\right) \cdot \left(\mathsf{max}\left(y, t\right) - z\right)}\\ \mathbf{elif}\;\mathsf{min}\left(y, t\right) \leq 4 \cdot 10^{-185}:\\ \;\;\;\;\frac{x}{z \cdot \left(z - \mathsf{max}\left(y, t\right)\right)}\\ \mathbf{else}:\\ \;\;\;\;\frac{x}{\left(\mathsf{min}\left(y, t\right) - z\right) \cdot \mathsf{max}\left(y, t\right)}\\ \end{array} \]

FPCore

(FPCore (x y z t)
  :precision binary64
  (if (<= (fmin y t) -2.9)
  (/ x (* (fmin y t) (- (fmax y t) z)))
  (if (<= (fmin y t) 4e-185)
    (/ x (* z (- z (fmax y t))))
    (/ x (* (- (fmin y t) z) (fmax y t))))))

C

double code(double x, double y, double z, double t) {
	double tmp;
	if (fmin(y, t) <= -2.9) {
		tmp = x / (fmin(y, t) * (fmax(y, t) - z));
	} else if (fmin(y, t) <= 4e-185) {
		tmp = x / (z * (z - fmax(y, t)));
	} else {
		tmp = x / ((fmin(y, t) - z) * fmax(y, t));
	}
	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 (fmin(y, t) <= (-2.9d0)) then
        tmp = x / (fmin(y, t) * (fmax(y, t) - z))
    else if (fmin(y, t) <= 4d-185) then
        tmp = x / (z * (z - fmax(y, t)))
    else
        tmp = x / ((fmin(y, t) - z) * fmax(y, t))
    end if
    code = tmp
end function

Java

public static double code(double x, double y, double z, double t) {
	double tmp;
	if (fmin(y, t) <= -2.9) {
		tmp = x / (fmin(y, t) * (fmax(y, t) - z));
	} else if (fmin(y, t) <= 4e-185) {
		tmp = x / (z * (z - fmax(y, t)));
	} else {
		tmp = x / ((fmin(y, t) - z) * fmax(y, t));
	}
	return tmp;
}

Python

def code(x, y, z, t):
	tmp = 0
	if fmin(y, t) <= -2.9:
		tmp = x / (fmin(y, t) * (fmax(y, t) - z))
	elif fmin(y, t) <= 4e-185:
		tmp = x / (z * (z - fmax(y, t)))
	else:
		tmp = x / ((fmin(y, t) - z) * fmax(y, t))
	return tmp

Julia

function code(x, y, z, t)
	tmp = 0.0
	if (fmin(y, t) <= -2.9)
		tmp = Float64(x / Float64(fmin(y, t) * Float64(fmax(y, t) - z)));
	elseif (fmin(y, t) <= 4e-185)
		tmp = Float64(x / Float64(z * Float64(z - fmax(y, t))));
	else
		tmp = Float64(x / Float64(Float64(fmin(y, t) - z) * fmax(y, t)));
	end
	return tmp
end

MATLAB

function tmp_2 = code(x, y, z, t)
	tmp = 0.0;
	if (min(y, t) <= -2.9)
		tmp = x / (min(y, t) * (max(y, t) - z));
	elseif (min(y, t) <= 4e-185)
		tmp = x / (z * (z - max(y, t)));
	else
		tmp = x / ((min(y, t) - z) * max(y, t));
	end
	tmp_2 = tmp;
end

Wolfram

code[x_, y_, z_, t_] := If[LessEqual[N[Min[y, t], $MachinePrecision], -2.9], N[(x / N[(N[Min[y, t], $MachinePrecision] * N[(N[Max[y, t], $MachinePrecision] - z), $MachinePrecision]), $MachinePrecision]), $MachinePrecision], If[LessEqual[N[Min[y, t], $MachinePrecision], 4e-185], N[(x / N[(z * N[(z - N[Max[y, t], $MachinePrecision]), $MachinePrecision]), $MachinePrecision]), $MachinePrecision], N[(x / N[(N[(N[Min[y, t], $MachinePrecision] - z), $MachinePrecision] * N[Max[y, t], $MachinePrecision]), $MachinePrecision]), $MachinePrecision]]]

Derivation

1. Split input into 3 regimes

2. if y < -2.8999999999999999

   1. Initial program 88.9%

      \[\frac{x}{\left(y - z\right) \cdot \left(t - z\right)} \]

   2. Taylor expanded in z around 0

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

   4. Applied rewrites 57.1%

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

   5. Taylor expanded in y around inf

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

      1. lower-*.f64 N/A

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

      2. lower--.f64 57.1%

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

   7. Applied rewrites 57.1%

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

   if -2.8999999999999999 < y < 4e-185

   1. Initial program 88.9%

      \[\frac{x}{\left(y - z\right) \cdot \left(t - z\right)} \]
   2. Step-by-step derivation

      1. lift-/.f64 N/A

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

      2. lift-*.f64 N/A

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

      3. associate-/r* N/A

         \[\leadsto \color{blue}{\frac{\frac{x}{y - z}}{t - z}} \]

      4. frac-2neg N/A

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

      5. lift--.f64 N/A

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

      6. sub-negate-rev N/A

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

      7. lower-/.f64 N/A

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

      8. distribute-neg-frac2 N/A

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

      9. lift--.f64 N/A

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

      10. sub-negate-rev N/A

          \[\leadsto \frac{\frac{x}{\color{blue}{z - y}}}{z - t} \]

      11. lower-/.f64 N/A

          \[\leadsto \frac{\color{blue}{\frac{x}{z - y}}}{z - t} \]

      12. lower--.f64 N/A

          \[\leadsto \frac{\frac{x}{\color{blue}{z - y}}}{z - t} \]

      13. lower--.f64 97.1%

          \[\leadsto \frac{\frac{x}{z - y}}{\color{blue}{z - t}} \]

   3. Applied rewrites 97.1%

      \[\leadsto \color{blue}{\frac{\frac{x}{z - y}}{z - t}} \]

   4. Taylor expanded in y around 0

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

      1. lower-/.f64 N/A

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

      2. lower-*.f64 N/A

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

      3. lower--.f64 53.6%

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

   6. Applied rewrites 53.6%

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

   if 4e-185 < y

   1. Initial program 88.9%

      \[\frac{x}{\left(y - z\right) \cdot \left(t - z\right)} \]

   2. Taylor expanded in z around 0

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

   4. Applied rewrites 57.1%

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

Alternative 6: 76.1% accurate, 0.1× speedup

Math

\[\begin{array}{l} t_1 := \frac{x}{\mathsf{min}\left(y, t\right) \cdot \left(\mathsf{max}\left(y, t\right) - z\right)}\\ \mathbf{if}\;\mathsf{min}\left(y, t\right) \leq -2.9:\\ \;\;\;\;t\_1\\ \mathbf{elif}\;\mathsf{min}\left(y, t\right) \leq 6.8 \cdot 10^{-60}:\\ \;\;\;\;\frac{x}{z \cdot \left(z - \mathsf{max}\left(y, t\right)\right)}\\ \mathbf{else}:\\ \;\;\;\;t\_1\\ \end{array} \]

FPCore

(FPCore (x y z t)
  :precision binary64
  (let* ((t_1 (/ x (* (fmin y t) (- (fmax y t) z)))))
  (if (<= (fmin y t) -2.9)
    t_1
    (if (<= (fmin y t) 6.8e-60) (/ x (* z (- z (fmax y t)))) t_1))))

C

double code(double x, double y, double z, double t) {
	double t_1 = x / (fmin(y, t) * (fmax(y, t) - z));
	double tmp;
	if (fmin(y, t) <= -2.9) {
		tmp = t_1;
	} else if (fmin(y, t) <= 6.8e-60) {
		tmp = x / (z * (z - fmax(y, t)));
	} else {
		tmp = t_1;
	}
	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) :: t_1
    real(8) :: tmp
    t_1 = x / (fmin(y, t) * (fmax(y, t) - z))
    if (fmin(y, t) <= (-2.9d0)) then
        tmp = t_1
    else if (fmin(y, t) <= 6.8d-60) then
        tmp = x / (z * (z - fmax(y, t)))
    else
        tmp = t_1
    end if
    code = tmp
end function

Java

public static double code(double x, double y, double z, double t) {
	double t_1 = x / (fmin(y, t) * (fmax(y, t) - z));
	double tmp;
	if (fmin(y, t) <= -2.9) {
		tmp = t_1;
	} else if (fmin(y, t) <= 6.8e-60) {
		tmp = x / (z * (z - fmax(y, t)));
	} else {
		tmp = t_1;
	}
	return tmp;
}

Python

def code(x, y, z, t):
	t_1 = x / (fmin(y, t) * (fmax(y, t) - z))
	tmp = 0
	if fmin(y, t) <= -2.9:
		tmp = t_1
	elif fmin(y, t) <= 6.8e-60:
		tmp = x / (z * (z - fmax(y, t)))
	else:
		tmp = t_1
	return tmp

Julia

function code(x, y, z, t)
	t_1 = Float64(x / Float64(fmin(y, t) * Float64(fmax(y, t) - z)))
	tmp = 0.0
	if (fmin(y, t) <= -2.9)
		tmp = t_1;
	elseif (fmin(y, t) <= 6.8e-60)
		tmp = Float64(x / Float64(z * Float64(z - fmax(y, t))));
	else
		tmp = t_1;
	end
	return tmp
end

MATLAB

function tmp_2 = code(x, y, z, t)
	t_1 = x / (min(y, t) * (max(y, t) - z));
	tmp = 0.0;
	if (min(y, t) <= -2.9)
		tmp = t_1;
	elseif (min(y, t) <= 6.8e-60)
		tmp = x / (z * (z - max(y, t)));
	else
		tmp = t_1;
	end
	tmp_2 = tmp;
end

Wolfram

code[x_, y_, z_, t_] := Block[{t$95$1 = N[(x / N[(N[Min[y, t], $MachinePrecision] * N[(N[Max[y, t], $MachinePrecision] - z), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]}, If[LessEqual[N[Min[y, t], $MachinePrecision], -2.9], t$95$1, If[LessEqual[N[Min[y, t], $MachinePrecision], 6.8e-60], N[(x / N[(z * N[(z - N[Max[y, t], $MachinePrecision]), $MachinePrecision]), $MachinePrecision]), $MachinePrecision], t$95$1]]]

Derivation

1. Split input into 2 regimes

2. if y < -2.8999999999999999 or 6.8000000000000001e-60 < y

   1. Initial program 88.9%

      \[\frac{x}{\left(y - z\right) \cdot \left(t - z\right)} \]

   2. Taylor expanded in z around 0

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

   4. Applied rewrites 57.1%

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

   5. Taylor expanded in y around inf

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

      1. lower-*.f64 N/A

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

      2. lower--.f64 57.1%

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

   7. Applied rewrites 57.1%

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

   if -2.8999999999999999 < y < 6.8000000000000001e-60

   1. Initial program 88.9%

      \[\frac{x}{\left(y - z\right) \cdot \left(t - z\right)} \]
   2. Step-by-step derivation

      1. lift-/.f64 N/A

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

      2. lift-*.f64 N/A

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

      3. associate-/r* N/A

         \[\leadsto \color{blue}{\frac{\frac{x}{y - z}}{t - z}} \]

      4. frac-2neg N/A

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

      5. lift--.f64 N/A

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

      6. sub-negate-rev N/A

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

      7. lower-/.f64 N/A

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

      8. distribute-neg-frac2 N/A

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

      9. lift--.f64 N/A

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

      10. sub-negate-rev N/A

          \[\leadsto \frac{\frac{x}{\color{blue}{z - y}}}{z - t} \]

      11. lower-/.f64 N/A

          \[\leadsto \frac{\color{blue}{\frac{x}{z - y}}}{z - t} \]

      12. lower--.f64 N/A

          \[\leadsto \frac{\frac{x}{\color{blue}{z - y}}}{z - t} \]

      13. lower--.f64 97.1%

          \[\leadsto \frac{\frac{x}{z - y}}{\color{blue}{z - t}} \]

   3. Applied rewrites 97.1%

      \[\leadsto \color{blue}{\frac{\frac{x}{z - y}}{z - t}} \]

   4. Taylor expanded in y around 0

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

      1. lower-/.f64 N/A

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

      2. lower-*.f64 N/A

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

      3. lower--.f64 53.6%

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

   6. Applied rewrites 53.6%

      \[\leadsto \color{blue}{\frac{x}{z \cdot \left(z - t\right)}} \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 7: 68.2% accurate, 0.7× speedup

Math

\[\begin{array}{l} t_1 := \frac{x}{z \cdot \left(z - t\right)}\\ \mathbf{if}\;z \leq -1.6 \cdot 10^{-38}:\\ \;\;\;\;t\_1\\ \mathbf{elif}\;z \leq 1.15 \cdot 10^{-90}:\\ \;\;\;\;\frac{x}{t \cdot y}\\ \mathbf{else}:\\ \;\;\;\;t\_1\\ \end{array} \]

FPCore

(FPCore (x y z t)
  :precision binary64
  (let* ((t_1 (/ x (* z (- z t)))))
  (if (<= z -1.6e-38) t_1 (if (<= z 1.15e-90) (/ x (* t y)) t_1))))

C

double code(double x, double y, double z, double t) {
	double t_1 = x / (z * (z - t));
	double tmp;
	if (z <= -1.6e-38) {
		tmp = t_1;
	} else if (z <= 1.15e-90) {
		tmp = x / (t * y);
	} else {
		tmp = t_1;
	}
	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) :: t_1
    real(8) :: tmp
    t_1 = x / (z * (z - t))
    if (z <= (-1.6d-38)) then
        tmp = t_1
    else if (z <= 1.15d-90) then
        tmp = x / (t * y)
    else
        tmp = t_1
    end if
    code = tmp
end function

Java

public static double code(double x, double y, double z, double t) {
	double t_1 = x / (z * (z - t));
	double tmp;
	if (z <= -1.6e-38) {
		tmp = t_1;
	} else if (z <= 1.15e-90) {
		tmp = x / (t * y);
	} else {
		tmp = t_1;
	}
	return tmp;
}

Python

def code(x, y, z, t):
	t_1 = x / (z * (z - t))
	tmp = 0
	if z <= -1.6e-38:
		tmp = t_1
	elif z <= 1.15e-90:
		tmp = x / (t * y)
	else:
		tmp = t_1
	return tmp

Julia

function code(x, y, z, t)
	t_1 = Float64(x / Float64(z * Float64(z - t)))
	tmp = 0.0
	if (z <= -1.6e-38)
		tmp = t_1;
	elseif (z <= 1.15e-90)
		tmp = Float64(x / Float64(t * y));
	else
		tmp = t_1;
	end
	return tmp
end

MATLAB

function tmp_2 = code(x, y, z, t)
	t_1 = x / (z * (z - t));
	tmp = 0.0;
	if (z <= -1.6e-38)
		tmp = t_1;
	elseif (z <= 1.15e-90)
		tmp = x / (t * y);
	else
		tmp = t_1;
	end
	tmp_2 = tmp;
end

Wolfram

code[x_, y_, z_, t_] := Block[{t$95$1 = N[(x / N[(z * N[(z - t), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]}, If[LessEqual[z, -1.6e-38], t$95$1, If[LessEqual[z, 1.15e-90], N[(x / N[(t * y), $MachinePrecision]), $MachinePrecision], t$95$1]]]

Derivation

1. Split input into 2 regimes

2. if z < -1.5999999999999999e-38 or 1.1499999999999999e-90 < z

   1. Initial program 88.9%

      \[\frac{x}{\left(y - z\right) \cdot \left(t - z\right)} \]
   2. Step-by-step derivation

      1. lift-/.f64 N/A

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

      2. lift-*.f64 N/A

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

      3. associate-/r* N/A

         \[\leadsto \color{blue}{\frac{\frac{x}{y - z}}{t - z}} \]

      4. frac-2neg N/A

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

      5. lift--.f64 N/A

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

      6. sub-negate-rev N/A

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

      7. lower-/.f64 N/A

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

      8. distribute-neg-frac2 N/A

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

      9. lift--.f64 N/A

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

      10. sub-negate-rev N/A

          \[\leadsto \frac{\frac{x}{\color{blue}{z - y}}}{z - t} \]

      11. lower-/.f64 N/A

          \[\leadsto \frac{\color{blue}{\frac{x}{z - y}}}{z - t} \]

      12. lower--.f64 N/A

          \[\leadsto \frac{\frac{x}{\color{blue}{z - y}}}{z - t} \]

      13. lower--.f64 97.1%

          \[\leadsto \frac{\frac{x}{z - y}}{\color{blue}{z - t}} \]

   3. Applied rewrites 97.1%

      \[\leadsto \color{blue}{\frac{\frac{x}{z - y}}{z - t}} \]

   4. Taylor expanded in y around 0

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

      1. lower-/.f64 N/A

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

      2. lower-*.f64 N/A

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

      3. lower--.f64 53.6%

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

   6. Applied rewrites 53.6%

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

   if -1.5999999999999999e-38 < z < 1.1499999999999999e-90

   1. Initial program 88.9%

      \[\frac{x}{\left(y - z\right) \cdot \left(t - z\right)} \]

   2. Taylor expanded in z around 0

      \[\leadsto \frac{x}{\color{blue}{t \cdot y}} \]
   3. Step-by-step derivation

      1. lower-*.f64 39.4%

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

   4. Applied rewrites 39.4%

      \[\leadsto \frac{x}{\color{blue}{t \cdot y}} \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 8: 48.2% accurate, 0.1× speedup

Math

\[\begin{array}{l} t_1 := \frac{x}{\left(-z\right) \cdot \mathsf{max}\left(y, t\right)}\\ \mathbf{if}\;z \leq -1.6 \cdot 10^{-38}:\\ \;\;\;\;t\_1\\ \mathbf{elif}\;z \leq 2.15 \cdot 10^{-30}:\\ \;\;\;\;\frac{x}{\mathsf{max}\left(y, t\right) \cdot \mathsf{min}\left(y, t\right)}\\ \mathbf{else}:\\ \;\;\;\;t\_1\\ \end{array} \]

FPCore

(FPCore (x y z t)
  :precision binary64
  (let* ((t_1 (/ x (* (- z) (fmax y t)))))
  (if (<= z -1.6e-38)
    t_1
    (if (<= z 2.15e-30) (/ x (* (fmax y t) (fmin y t))) t_1))))

C

double code(double x, double y, double z, double t) {
	double t_1 = x / (-z * fmax(y, t));
	double tmp;
	if (z <= -1.6e-38) {
		tmp = t_1;
	} else if (z <= 2.15e-30) {
		tmp = x / (fmax(y, t) * fmin(y, t));
	} else {
		tmp = t_1;
	}
	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) :: t_1
    real(8) :: tmp
    t_1 = x / (-z * fmax(y, t))
    if (z <= (-1.6d-38)) then
        tmp = t_1
    else if (z <= 2.15d-30) then
        tmp = x / (fmax(y, t) * fmin(y, t))
    else
        tmp = t_1
    end if
    code = tmp
end function

Java

public static double code(double x, double y, double z, double t) {
	double t_1 = x / (-z * fmax(y, t));
	double tmp;
	if (z <= -1.6e-38) {
		tmp = t_1;
	} else if (z <= 2.15e-30) {
		tmp = x / (fmax(y, t) * fmin(y, t));
	} else {
		tmp = t_1;
	}
	return tmp;
}

Python

def code(x, y, z, t):
	t_1 = x / (-z * fmax(y, t))
	tmp = 0
	if z <= -1.6e-38:
		tmp = t_1
	elif z <= 2.15e-30:
		tmp = x / (fmax(y, t) * fmin(y, t))
	else:
		tmp = t_1
	return tmp

Julia

function code(x, y, z, t)
	t_1 = Float64(x / Float64(Float64(-z) * fmax(y, t)))
	tmp = 0.0
	if (z <= -1.6e-38)
		tmp = t_1;
	elseif (z <= 2.15e-30)
		tmp = Float64(x / Float64(fmax(y, t) * fmin(y, t)));
	else
		tmp = t_1;
	end
	return tmp
end

MATLAB

function tmp_2 = code(x, y, z, t)
	t_1 = x / (-z * max(y, t));
	tmp = 0.0;
	if (z <= -1.6e-38)
		tmp = t_1;
	elseif (z <= 2.15e-30)
		tmp = x / (max(y, t) * min(y, t));
	else
		tmp = t_1;
	end
	tmp_2 = tmp;
end

Wolfram

code[x_, y_, z_, t_] := Block[{t$95$1 = N[(x / N[((-z) * N[Max[y, t], $MachinePrecision]), $MachinePrecision]), $MachinePrecision]}, If[LessEqual[z, -1.6e-38], t$95$1, If[LessEqual[z, 2.15e-30], N[(x / N[(N[Max[y, t], $MachinePrecision] * N[Min[y, t], $MachinePrecision]), $MachinePrecision]), $MachinePrecision], t$95$1]]]

Derivation

1. Split input into 2 regimes

2. if z < -1.5999999999999999e-38 or 2.1499999999999998e-30 < z

   1. Initial program 88.9%

      \[\frac{x}{\left(y - z\right) \cdot \left(t - z\right)} \]

   2. Taylor expanded in z around 0

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

   4. Applied rewrites 57.1%

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

   5. Taylor expanded in y around 0

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

      1. lower-*.f64 31.9%

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

   7. Applied rewrites 31.9%

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

      1. lift-*.f64 N/A

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

      2. metadata-eval N/A

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

      3. distribute-lft-neg-out N/A

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

      4. metadata-eval N/A

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

      5. distribute-lft-neg-out N/A

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

      6. mul-1-neg N/A

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

      7. lower-neg.f64 N/A

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

      8. mul-1-neg N/A

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

      9. *-commutative N/A

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

      10. distribute-rgt-neg-out N/A

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

      11. metadata-eval N/A

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

      12. *-rgt-identity 31.9%

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

   9. Applied rewrites 31.9%

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

   if -1.5999999999999999e-38 < z < 2.1499999999999998e-30

   1. Initial program 88.9%

      \[\frac{x}{\left(y - z\right) \cdot \left(t - z\right)} \]

   2. Taylor expanded in z around 0

      \[\leadsto \frac{x}{\color{blue}{t \cdot y}} \]
   3. Step-by-step derivation

      1. lower-*.f64 39.4%

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

   4. Applied rewrites 39.4%

      \[\leadsto \frac{x}{\color{blue}{t \cdot y}} \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 9: 39.4% accurate, 1.4× speedup

Math

\[\frac{x}{t \cdot y} \]

FPCore

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

C

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

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 / (t * y)
end function

Java

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

Python

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

Julia

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

MATLAB

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

Wolfram

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

Derivation

1. Initial program 88.9%

   \[\frac{x}{\left(y - z\right) \cdot \left(t - z\right)} \]

2. Taylor expanded in z around 0

   \[\leadsto \frac{x}{\color{blue}{t \cdot y}} \]
3. Step-by-step derivation

   1. lower-*.f64 39.4%

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

4. Applied rewrites 39.4%

   \[\leadsto \frac{x}{\color{blue}{t \cdot y}} \]
5. Add Preprocessing

Reproduce

herbie shell --seed 2025258 
(FPCore (x y z t)
  :name "Data.Random.Distribution.Triangular:triangularCDF from random-fu-0.2.6.2, B"
  :precision binary64
  (/ x (* (- y z) (- t z))))