class Gaussian_Integer():
    #入力定義
    def __init__(self,Real_part=0,Imaginary_part=0):
        self.re=Real_part
        self.im=Imaginary_part

    #表示定義
    def __str__(self):
        s=""
        s=Gaussian_Integer.__strmake(s,self.re,"")
        s=Gaussian_Integer.__strmake(s,self.im,"i")
        if s=="":
            return "0"
        else:
            return s

    def __strmake(self,coefficient,axis):
        if coefficient==0:
            return self
        else:
            if self=="":
                if axis=="":
                    self+=str(coefficient)
                else:
                    if coefficient==1:self+=axis
                    elif coefficient==-1:self+="-"+axis
                    else:self+=str(coefficient)+axis
            else:
                if coefficient>0:
                    if coefficient==1:self+="+"+axis
                    else:self+="+"+str(coefficient)+axis
                else:
                    if coefficient==-1:self+="-"+axis
                    else:self+=str(coefficient)+axis

        return self

    #四則演算定義
    #加法
    def __add__(self,other):
        if isinstance(other,Gaussian_Integer):
            return Gaussian_Integer(self.re+other.re,self.im+other.im)
        else:
            return Gaussian_Integer(self.re+other,self.im)

    def __radd__(self,other):
        if isinstance(other,int):
            return Gaussian_Integer(self.re+other,self.im)

    #減法
    def __sub__(self,other):
        return self+(-other)

    def __rsub__(self,other):
        if isinstance(other,int):
            return (-self)+other

    #乗法
    def __mul__(self,other):
        a,b=self.re,self.im
        if isinstance(other,Gaussian_Integer):
            c,d=other.re,other.im
            return Gaussian_Integer(a*c-b*d,a*d+b*c)
        else:
            return Gaussian_Integer(other*a,other*b)

    def __rmul__(self,other):
        if isinstance(other,int):
            a,b=self.re,self.im
            return Gaussian_Integer(other*a,other*b)

    #除法
    def __truediv__(self,other):
        pass

    def __rtruediv__(self,other):
        pass

    def __floordiv__(self,other):
        if isinstance(other,int):
            other=Gaussian_Integer(other,0)

        a,b=self.re,self.im
        c,d=other.re,other.im

        n=other.norm()

        p=(2*(a*c+b*d)+n)//(2*n)
        q=(2*(b*c-a*d)+n)//(2*n)

        return Gaussian_Integer(p,q)

    def __mod__(self,other):
        return  self-other*(self//other)
    #比較演算子
    def __eq__(self,other):
        if isinstance(other,Gaussian_Integer):
            return (self.re==other.re) and (self.im==other.im)
        else:
            return (self-other)==Gaussian_Integer(0,0)

    def __bool__(self):
        return not(self==0)

    #その他
    def conjugate(self):
        return Gaussian_Integer(self.re,-self.im)

    def __abs__(self):
        import math
        return math.sqrt(self.norm())

    def norm(self):
        return self.re*self.re+self.im*self.im

    #実数から複素数に変換
    def Real_to_Complex(self):
        pass

    #正負判定

    #要約

    #逆数
    def __inverse(self):
        pass

    #符号
    def __pos__(self):
        return self

    def __neg__(self):
        return Gaussian_Integer(-self.re,-self.im)

#最大公約数
def gcd(x,y):
    """Gauss整数 x,yの最大公約数を求める.

    x,y:Gauss整数
    """

    if x.norm()<y.norm():
        x,y=y,x

    while y!=0:
        x,y=y,x%y

    return x
#================================================
A,B=map(int,input().split())
alpha=Gaussian_Integer(A,B)
beta =alpha.conjugate()
gamma=gcd(alpha,beta)

if alpha==0:
    print(0.25)

assert gamma%1==0
print(0.3371877158)