2-ancestry mixing, negative discriminant

Percentage Accurate: 98.5% → 100.0%
Time: 5.4s
Alternatives: 3
Speedup: 1.1×

Specification

?
\[\begin{array}{l} \\ 2 \cdot \cos \left(\frac{2 \cdot \mathsf{PI}\left(\right)}{3} + \frac{\cos^{-1} \left(\frac{-g}{h}\right)}{3}\right) \end{array} \]
(FPCore (g h)
 :precision binary64
 (* 2.0 (cos (+ (/ (* 2.0 (PI)) 3.0) (/ (acos (/ (- g) h)) 3.0)))))
\begin{array}{l}

\\
2 \cdot \cos \left(\frac{2 \cdot \mathsf{PI}\left(\right)}{3} + \frac{\cos^{-1} \left(\frac{-g}{h}\right)}{3}\right)
\end{array}

Sampling outcomes in binary64 precision:

Local Percentage Accuracy vs ?

The average percentage accuracy by input value. Horizontal axis shows value of an input variable; the variable is choosen in the title. Vertical axis is accuracy; higher is better. Red represent the original program, while blue represents Herbie's suggestion. These can be toggled with buttons below the plot. The line is an average while dots represent individual samples.

Accuracy vs Speed?

Herbie found 3 alternatives:

AlternativeAccuracySpeedup
The accuracy (vertical axis) and speed (horizontal axis) of each alternatives. Up and to the right is better. The red square shows the initial program, and each blue circle shows an alternative.The line shows the best available speed-accuracy tradeoffs.

Initial Program: 98.5% accurate, 1.0× speedup?

\[\begin{array}{l} \\ 2 \cdot \cos \left(\frac{2 \cdot \mathsf{PI}\left(\right)}{3} + \frac{\cos^{-1} \left(\frac{-g}{h}\right)}{3}\right) \end{array} \]
(FPCore (g h)
 :precision binary64
 (* 2.0 (cos (+ (/ (* 2.0 (PI)) 3.0) (/ (acos (/ (- g) h)) 3.0)))))
\begin{array}{l}

\\
2 \cdot \cos \left(\frac{2 \cdot \mathsf{PI}\left(\right)}{3} + \frac{\cos^{-1} \left(\frac{-g}{h}\right)}{3}\right)
\end{array}

Alternative 1: 100.0% accurate, 1.0× speedup?

\[\begin{array}{l} \\ -2 \cdot \sin \left(\mathsf{fma}\left(0.3333333333333333, \cos^{-1} \left(\frac{-g}{h}\right) - \mathsf{PI}\left(\right), \frac{\mathsf{PI}\left(\right)}{2}\right)\right) \end{array} \]
(FPCore (g h)
 :precision binary64
 (*
  -2.0
  (sin (fma 0.3333333333333333 (- (acos (/ (- g) h)) (PI)) (/ (PI) 2.0)))))
\begin{array}{l}

\\
-2 \cdot \sin \left(\mathsf{fma}\left(0.3333333333333333, \cos^{-1} \left(\frac{-g}{h}\right) - \mathsf{PI}\left(\right), \frac{\mathsf{PI}\left(\right)}{2}\right)\right)
\end{array}
Derivation
  1. Initial program 98.4%

    \[2 \cdot \cos \left(\frac{2 \cdot \mathsf{PI}\left(\right)}{3} + \frac{\cos^{-1} \left(\frac{-g}{h}\right)}{3}\right) \]
  2. Add Preprocessing
  3. Taylor expanded in g around 0

    \[\leadsto \color{blue}{2 \cdot \cos \left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right)} \]
  4. Step-by-step derivation
    1. metadata-evalN/A

      \[\leadsto \color{blue}{\left(-1 \cdot -2\right)} \cdot \cos \left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right) \]
    2. metadata-evalN/A

      \[\leadsto \left(-1 \cdot \color{blue}{\left(-1 + -1\right)}\right) \cdot \cos \left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right) \]
    3. *-commutativeN/A

      \[\leadsto \color{blue}{\left(\left(-1 + -1\right) \cdot -1\right)} \cdot \cos \left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right) \]
    4. associate-*l*N/A

      \[\leadsto \color{blue}{\left(-1 + -1\right) \cdot \left(-1 \cdot \cos \left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right)\right)} \]
    5. mul-1-negN/A

      \[\leadsto \left(-1 + -1\right) \cdot \color{blue}{\left(\mathsf{neg}\left(\cos \left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right)\right)\right)} \]
    6. lower-*.f64N/A

      \[\leadsto \color{blue}{\left(-1 + -1\right) \cdot \left(\mathsf{neg}\left(\cos \left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right)\right)\right)} \]
    7. metadata-evalN/A

      \[\leadsto \color{blue}{-2} \cdot \left(\mathsf{neg}\left(\cos \left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right)\right)\right) \]
    8. cos-neg-revN/A

      \[\leadsto -2 \cdot \left(\mathsf{neg}\left(\color{blue}{\cos \left(\mathsf{neg}\left(\left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right)\right)\right)}\right)\right) \]
    9. cos-+PI-revN/A

      \[\leadsto -2 \cdot \color{blue}{\cos \left(\left(\mathsf{neg}\left(\left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right)\right)\right) + \mathsf{PI}\left(\right)\right)} \]
    10. remove-double-negN/A

      \[\leadsto -2 \cdot \cos \left(\left(\mathsf{neg}\left(\left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right)\right)\right) + \color{blue}{\left(\mathsf{neg}\left(\left(\mathsf{neg}\left(\mathsf{PI}\left(\right)\right)\right)\right)\right)}\right) \]
    11. distribute-neg-outN/A

      \[\leadsto -2 \cdot \cos \color{blue}{\left(\mathsf{neg}\left(\left(\left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right) + \left(\mathsf{neg}\left(\mathsf{PI}\left(\right)\right)\right)\right)\right)\right)} \]
    12. cos-negN/A

      \[\leadsto -2 \cdot \color{blue}{\cos \left(\left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right) + \left(\mathsf{neg}\left(\mathsf{PI}\left(\right)\right)\right)\right)} \]
    13. lower-cos.f64N/A

      \[\leadsto -2 \cdot \color{blue}{\cos \left(\left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right) + \left(\mathsf{neg}\left(\mathsf{PI}\left(\right)\right)\right)\right)} \]
  5. Applied rewrites98.5%

    \[\leadsto \color{blue}{-2 \cdot \cos \left(\mathsf{fma}\left(\cos^{-1} \left(\frac{-g}{h}\right), 0.3333333333333333, -0.3333333333333333 \cdot \mathsf{PI}\left(\right)\right)\right)} \]
  6. Step-by-step derivation
    1. Applied rewrites100.0%

      \[\leadsto -2 \cdot \sin \left(\mathsf{fma}\left(0.3333333333333333, \cos^{-1} \left(\frac{-g}{h}\right) - \mathsf{PI}\left(\right), \frac{\mathsf{PI}\left(\right)}{2}\right)\right) \]
    2. Add Preprocessing

    Alternative 2: 100.0% accurate, 1.1× speedup?

    \[\begin{array}{l} \\ 2 \cdot \sin \left(\mathsf{fma}\left(\cos^{-1} \left(\frac{-g}{h}\right), -0.3333333333333333, \mathsf{PI}\left(\right) \cdot -0.16666666666666666\right)\right) \end{array} \]
    (FPCore (g h)
     :precision binary64
     (*
      2.0
      (sin
       (fma
        (acos (/ (- g) h))
        -0.3333333333333333
        (* (PI) -0.16666666666666666)))))
    \begin{array}{l}
    
    \\
    2 \cdot \sin \left(\mathsf{fma}\left(\cos^{-1} \left(\frac{-g}{h}\right), -0.3333333333333333, \mathsf{PI}\left(\right) \cdot -0.16666666666666666\right)\right)
    \end{array}
    
    Derivation
    1. Initial program 98.4%

      \[2 \cdot \cos \left(\frac{2 \cdot \mathsf{PI}\left(\right)}{3} + \frac{\cos^{-1} \left(\frac{-g}{h}\right)}{3}\right) \]
    2. Add Preprocessing
    3. Step-by-step derivation
      1. lift-cos.f64N/A

        \[\leadsto 2 \cdot \color{blue}{\cos \left(\frac{2 \cdot \mathsf{PI}\left(\right)}{3} + \frac{\cos^{-1} \left(\frac{-g}{h}\right)}{3}\right)} \]
      2. sin-+PI/2-revN/A

        \[\leadsto 2 \cdot \color{blue}{\sin \left(\left(\frac{2 \cdot \mathsf{PI}\left(\right)}{3} + \frac{\cos^{-1} \left(\frac{-g}{h}\right)}{3}\right) + \frac{\mathsf{PI}\left(\right)}{2}\right)} \]
      3. lower-sin.f64N/A

        \[\leadsto 2 \cdot \color{blue}{\sin \left(\left(\frac{2 \cdot \mathsf{PI}\left(\right)}{3} + \frac{\cos^{-1} \left(\frac{-g}{h}\right)}{3}\right) + \frac{\mathsf{PI}\left(\right)}{2}\right)} \]
      4. +-commutativeN/A

        \[\leadsto 2 \cdot \sin \color{blue}{\left(\frac{\mathsf{PI}\left(\right)}{2} + \left(\frac{2 \cdot \mathsf{PI}\left(\right)}{3} + \frac{\cos^{-1} \left(\frac{-g}{h}\right)}{3}\right)\right)} \]
      5. lower-+.f64N/A

        \[\leadsto 2 \cdot \sin \color{blue}{\left(\frac{\mathsf{PI}\left(\right)}{2} + \left(\frac{2 \cdot \mathsf{PI}\left(\right)}{3} + \frac{\cos^{-1} \left(\frac{-g}{h}\right)}{3}\right)\right)} \]
      6. lift-PI.f64N/A

        \[\leadsto 2 \cdot \sin \left(\frac{\color{blue}{\mathsf{PI}\left(\right)}}{2} + \left(\frac{2 \cdot \mathsf{PI}\left(\right)}{3} + \frac{\cos^{-1} \left(\frac{-g}{h}\right)}{3}\right)\right) \]
      7. lower-/.f6497.6

        \[\leadsto 2 \cdot \sin \left(\color{blue}{\frac{\mathsf{PI}\left(\right)}{2}} + \left(\frac{2 \cdot \mathsf{PI}\left(\right)}{3} + \frac{\cos^{-1} \left(\frac{-g}{h}\right)}{3}\right)\right) \]
      8. lift-+.f64N/A

        \[\leadsto 2 \cdot \sin \left(\frac{\mathsf{PI}\left(\right)}{2} + \color{blue}{\left(\frac{2 \cdot \mathsf{PI}\left(\right)}{3} + \frac{\cos^{-1} \left(\frac{-g}{h}\right)}{3}\right)}\right) \]
      9. lift-/.f64N/A

        \[\leadsto 2 \cdot \sin \left(\frac{\mathsf{PI}\left(\right)}{2} + \left(\color{blue}{\frac{2 \cdot \mathsf{PI}\left(\right)}{3}} + \frac{\cos^{-1} \left(\frac{-g}{h}\right)}{3}\right)\right) \]
      10. lift-/.f64N/A

        \[\leadsto 2 \cdot \sin \left(\frac{\mathsf{PI}\left(\right)}{2} + \left(\frac{2 \cdot \mathsf{PI}\left(\right)}{3} + \color{blue}{\frac{\cos^{-1} \left(\frac{-g}{h}\right)}{3}}\right)\right) \]
      11. div-add-revN/A

        \[\leadsto 2 \cdot \sin \left(\frac{\mathsf{PI}\left(\right)}{2} + \color{blue}{\frac{2 \cdot \mathsf{PI}\left(\right) + \cos^{-1} \left(\frac{-g}{h}\right)}{3}}\right) \]
      12. lower-/.f64N/A

        \[\leadsto 2 \cdot \sin \left(\frac{\mathsf{PI}\left(\right)}{2} + \color{blue}{\frac{2 \cdot \mathsf{PI}\left(\right) + \cos^{-1} \left(\frac{-g}{h}\right)}{3}}\right) \]
      13. lift-*.f64N/A

        \[\leadsto 2 \cdot \sin \left(\frac{\mathsf{PI}\left(\right)}{2} + \frac{\color{blue}{2 \cdot \mathsf{PI}\left(\right)} + \cos^{-1} \left(\frac{-g}{h}\right)}{3}\right) \]
      14. *-commutativeN/A

        \[\leadsto 2 \cdot \sin \left(\frac{\mathsf{PI}\left(\right)}{2} + \frac{\color{blue}{\mathsf{PI}\left(\right) \cdot 2} + \cos^{-1} \left(\frac{-g}{h}\right)}{3}\right) \]
      15. lower-fma.f6497.6

        \[\leadsto 2 \cdot \sin \left(\frac{\mathsf{PI}\left(\right)}{2} + \frac{\color{blue}{\mathsf{fma}\left(\mathsf{PI}\left(\right), 2, \cos^{-1} \left(\frac{-g}{h}\right)\right)}}{3}\right) \]
    4. Applied rewrites97.6%

      \[\leadsto 2 \cdot \color{blue}{\sin \left(\frac{\mathsf{PI}\left(\right)}{2} + \frac{\mathsf{fma}\left(\mathsf{PI}\left(\right), 2, \cos^{-1} \left(\frac{-g}{h}\right)\right)}{3}\right)} \]
    5. Taylor expanded in g around 0

      \[\leadsto 2 \cdot \sin \color{blue}{\left(\frac{1}{3} \cdot \left(\cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + 2 \cdot \mathsf{PI}\left(\right)\right) + \frac{1}{2} \cdot \mathsf{PI}\left(\right)\right)} \]
    6. Applied rewrites97.6%

      \[\leadsto 2 \cdot \sin \color{blue}{\left(\mathsf{fma}\left(\cos^{-1} \left(\frac{-g}{h}\right), 0.3333333333333333, 1.1666666666666667 \cdot \mathsf{PI}\left(\right)\right)\right)} \]
    7. Taylor expanded in g around 0

      \[\leadsto 2 \cdot \color{blue}{\sin \left(\frac{1}{3} \cdot \left(\cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + 2 \cdot \mathsf{PI}\left(\right)\right) + \frac{1}{2} \cdot \mathsf{PI}\left(\right)\right)} \]
    8. Applied rewrites100.0%

      \[\leadsto 2 \cdot \color{blue}{\sin \left(\mathsf{fma}\left(\cos^{-1} \left(\frac{-g}{h}\right), -0.3333333333333333, \mathsf{PI}\left(\right) \cdot -0.16666666666666666\right)\right)} \]
    9. Add Preprocessing

    Alternative 3: 98.5% accurate, 1.1× speedup?

    \[\begin{array}{l} \\ -2 \cdot \cos \left(\cos^{-1} \left(\frac{g}{h}\right) \cdot -0.3333333333333333\right) \end{array} \]
    (FPCore (g h)
     :precision binary64
     (* -2.0 (cos (* (acos (/ g h)) -0.3333333333333333))))
    double code(double g, double h) {
    	return -2.0 * cos((acos((g / h)) * -0.3333333333333333));
    }
    
    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(g, h)
    use fmin_fmax_functions
        real(8), intent (in) :: g
        real(8), intent (in) :: h
        code = (-2.0d0) * cos((acos((g / h)) * (-0.3333333333333333d0)))
    end function
    
    public static double code(double g, double h) {
    	return -2.0 * Math.cos((Math.acos((g / h)) * -0.3333333333333333));
    }
    
    def code(g, h):
    	return -2.0 * math.cos((math.acos((g / h)) * -0.3333333333333333))
    
    function code(g, h)
    	return Float64(-2.0 * cos(Float64(acos(Float64(g / h)) * -0.3333333333333333)))
    end
    
    function tmp = code(g, h)
    	tmp = -2.0 * cos((acos((g / h)) * -0.3333333333333333));
    end
    
    code[g_, h_] := N[(-2.0 * N[Cos[N[(N[ArcCos[N[(g / h), $MachinePrecision]], $MachinePrecision] * -0.3333333333333333), $MachinePrecision]], $MachinePrecision]), $MachinePrecision]
    
    \begin{array}{l}
    
    \\
    -2 \cdot \cos \left(\cos^{-1} \left(\frac{g}{h}\right) \cdot -0.3333333333333333\right)
    \end{array}
    
    Derivation
    1. Initial program 98.4%

      \[2 \cdot \cos \left(\frac{2 \cdot \mathsf{PI}\left(\right)}{3} + \frac{\cos^{-1} \left(\frac{-g}{h}\right)}{3}\right) \]
    2. Add Preprocessing
    3. Taylor expanded in g around 0

      \[\leadsto \color{blue}{2 \cdot \cos \left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right)} \]
    4. Step-by-step derivation
      1. metadata-evalN/A

        \[\leadsto \color{blue}{\left(-1 \cdot -2\right)} \cdot \cos \left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right) \]
      2. metadata-evalN/A

        \[\leadsto \left(-1 \cdot \color{blue}{\left(-1 + -1\right)}\right) \cdot \cos \left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right) \]
      3. *-commutativeN/A

        \[\leadsto \color{blue}{\left(\left(-1 + -1\right) \cdot -1\right)} \cdot \cos \left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right) \]
      4. associate-*l*N/A

        \[\leadsto \color{blue}{\left(-1 + -1\right) \cdot \left(-1 \cdot \cos \left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right)\right)} \]
      5. mul-1-negN/A

        \[\leadsto \left(-1 + -1\right) \cdot \color{blue}{\left(\mathsf{neg}\left(\cos \left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right)\right)\right)} \]
      6. lower-*.f64N/A

        \[\leadsto \color{blue}{\left(-1 + -1\right) \cdot \left(\mathsf{neg}\left(\cos \left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right)\right)\right)} \]
      7. metadata-evalN/A

        \[\leadsto \color{blue}{-2} \cdot \left(\mathsf{neg}\left(\cos \left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right)\right)\right) \]
      8. cos-neg-revN/A

        \[\leadsto -2 \cdot \left(\mathsf{neg}\left(\color{blue}{\cos \left(\mathsf{neg}\left(\left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right)\right)\right)}\right)\right) \]
      9. cos-+PI-revN/A

        \[\leadsto -2 \cdot \color{blue}{\cos \left(\left(\mathsf{neg}\left(\left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right)\right)\right) + \mathsf{PI}\left(\right)\right)} \]
      10. remove-double-negN/A

        \[\leadsto -2 \cdot \cos \left(\left(\mathsf{neg}\left(\left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right)\right)\right) + \color{blue}{\left(\mathsf{neg}\left(\left(\mathsf{neg}\left(\mathsf{PI}\left(\right)\right)\right)\right)\right)}\right) \]
      11. distribute-neg-outN/A

        \[\leadsto -2 \cdot \cos \color{blue}{\left(\mathsf{neg}\left(\left(\left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right) + \left(\mathsf{neg}\left(\mathsf{PI}\left(\right)\right)\right)\right)\right)\right)} \]
      12. cos-negN/A

        \[\leadsto -2 \cdot \color{blue}{\cos \left(\left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right) + \left(\mathsf{neg}\left(\mathsf{PI}\left(\right)\right)\right)\right)} \]
      13. lower-cos.f64N/A

        \[\leadsto -2 \cdot \color{blue}{\cos \left(\left(\frac{1}{3} \cdot \cos^{-1} \left(-1 \cdot \frac{g}{h}\right) + \frac{2}{3} \cdot \mathsf{PI}\left(\right)\right) + \left(\mathsf{neg}\left(\mathsf{PI}\left(\right)\right)\right)\right)} \]
    5. Applied rewrites98.5%

      \[\leadsto \color{blue}{-2 \cdot \cos \left(\mathsf{fma}\left(\cos^{-1} \left(\frac{-g}{h}\right), 0.3333333333333333, -0.3333333333333333 \cdot \mathsf{PI}\left(\right)\right)\right)} \]
    6. Step-by-step derivation
      1. Applied rewrites98.5%

        \[\leadsto -2 \cdot \cos \left(\mathsf{fma}\left(\mathsf{PI}\left(\right) - \cos^{-1} \left(\frac{g}{h}\right), 0.3333333333333333, -0.3333333333333333 \cdot \mathsf{PI}\left(\right)\right)\right) \]
      2. Taylor expanded in g around 0

        \[\leadsto -2 \cdot \cos \left(\frac{-1}{3} \cdot \mathsf{PI}\left(\right) + \frac{1}{3} \cdot \left(\mathsf{PI}\left(\right) - \cos^{-1} \left(\frac{g}{h}\right)\right)\right) \]
      3. Step-by-step derivation
        1. Applied rewrites98.5%

          \[\leadsto -2 \cdot \cos \left(\cos^{-1} \left(\frac{g}{h}\right) \cdot -0.3333333333333333\right) \]
        2. Add Preprocessing

        Reproduce

        ?
        herbie shell --seed 2025011 
        (FPCore (g h)
          :name "2-ancestry mixing, negative discriminant"
          :precision binary64
          (* 2.0 (cos (+ (/ (* 2.0 (PI)) 3.0) (/ (acos (/ (- g) h)) 3.0)))))