/// <summary>
/// Computes the sine of the given value to full significance over the full range of arguments.
/// </summary>
/// <param name="x">The argument.</param>
/// <returns>The value of sin(x).</returns>
/// <remarks>
/// <para>This method addresses several subtle shortcommings of the <see cref="System.Math.Sin" /> method.
/// One shortcoming, quite striking but rarely encountered, is that <see cref="Math.Sin"/> returns entirely wrong results very large
/// arguments; for x larger than about 10<sup>20</sup>, it simply returns the argument as the function value!
/// (I have no idea
/// why the base class library designers did not at least choose to return <see cref="Double.NaN"/> so as to signal to the user
/// that the result should not be trusted. No floating point standard specifies this crazy behavior.)
/// Another shortcomming, more commonly encountered but often unnoticed, is that for large but not necessarily very large arguments,
/// function values loose precision, particularly near zeros of the function.</para>
/// <para>
/// One way to view these shortcommings is that they are justified by the uncertainty inherent in floating point representations. In this
/// view, any <see cref="System.Double"/> should be seen as an uncertain value with a relative error of ~10<sup>-16</sup>. If the
/// number is very large, then the absolute size of this error can be as large or larger than 2π; in this circumstance we should not
/// expect to be able to say anything about the value. (Except, of course, that it is between -1 and +1, which is violated by the designers' crazy
/// choice to return the argument as the value). Even if the absolute error is just a non-negligable faction of 2π, there is a non-neglible fraction
/// of the values between -1 and +1 in the corresponding range of function values; any of these values is as good as any other as an answer to the
/// question of the sine of our uncertain argument, so we should be satisfied with any returned value in this non-negligible range.
/// </para>
/// <para>A contrary view is that it is better to treat all arguments as infinitely precise and return the nearest representable <see cref="System.Double"/>
/// to the actual function value under this assumption. Users are unlikely to complain that we returned a more accurate value and this behavior is particularly
/// useful when the argument is an intermediate result that the programmer may not even realize has become large.</para>
/// <para>For typical arguments, say between 10<sup>-4</sup> and 10<sup>4</sup>, the extra cost of this function over
/// <see cref="Math.Sin"/> is just a couple of comparisons and a single floating point operation; less than 0.1% of
/// arguments in this range are then routed to our much slower, higher-accuracy algorithm. We therefore suggest that,
/// for general use, you prefer this method over <see cref="Math.Sin"/>; only in very unusual situations where (i) you are guaranteed
/// never to encounter very large arguments, (ii) full precision values are not required, and (iii) the run-time of your
/// application is critical and dominated by trigonometric calculations, should you prefer the base class library method.</para>
/// </remarks>
public static double Sin(double x)
{
if (x < 0.0)
{
return(-Sin(-x));
}
else if (x < 1.0)
{
// If x is small enough not to cross a zero, use the built-in function
return(Math.Sin(x));
}
else if (x < RangeReduction.xLimit)
{
// If x is in the intermediate region, try the built-in function but switch to our range-reduction
// algorithm if the result is too small.
double y = Math.Sin(x);
double ym = x / RangeReduction.xLimit;
if (Math.Abs(y) < ym)
{
return(RangeReduction.Sin(x));
}
else
{
return(y);
}
}
else
{
// If x is beyond the range of the built-in function eintirely, use our range-reduction algorithm
return(RangeReduction.Sin(x));
}
}
internal static double SinPi(double x)
{
long y0;
double y1;
RangeReduction.ReduceByOnes(2.0 * x, out y0, out y1);
return(RangeReduction.Sin(y0, y1));
}
/// <summary>
/// Computes the sine of the given multiple of π.
/// </summary>
/// <param name="x">The argument.</param>
/// <returns>The value of sin(<paramref name="x"/>π).</returns>
/// <remarks>
/// <para>This function allows the user to increase performance and avoid inaccuracies due to the finite
/// precision of the stored constant <see cref="Math.PI"/> in some cases. Suppose you need to compute
/// sin(xπ) for a large value of x.
/// </para>
/// </remarks>
/// <seealso cref="CosPi(double)"/>
public static double SinPi(double x)
{
RangeReduction.ReduceByOnes(2.0 * x, out long y0, out double y1);
return(RangeReduction.Sin(y0, y1));
}