BigFloat Class

[SerializableAttribute]
public class BigFloat : IComparable, IFormattable, 
	IEquatable<BigFloat>, IComparable<BigFloat>

<SerializableAttribute>
Public Class BigFloat
	Implements IComparable, IFormattable, IEquatable(Of BigFloat), 
	IComparable(Of BigFloat)

[SerializableAttribute]
public ref class BigFloat : IComparable, 
	IFormattable, IEquatable<BigFloat^>, IComparable<BigFloat^>

[<SerializableAttribute>]
type BigFloat = 
    class
        interface IComparable
        interface IFormattable
        interface IEquatable<BigFloat>
        interface IComparable<BigFloat>
    end

Inheritance

Object  →  BigFloat

Implements

IComparable, IComparable<BigFloat>, IEquatable<BigFloat>, IFormattable

Use the BigFloat class to represent numbers that may not fit into the standard single or double precision floating point types, or the decimal type.

A BigFloat can have a precision of up to 2³⁶ bits. The binary exponent is an integer, which allows a decimal exponent up to roughly 646 million. This gives the BigFloat type a total range of roughly

BigFloat numbers can be constructed in a variety of ways, including from other numerical types and from strings. Implicit (widening) conversions are available for all built-in integer types and floating-point types. Explicit conversions are also available for Decimal and BigRational.

The BigFloat class provides methods for all the common operations on real numbers. Overloaded versions of the major arithmetic operators are provided for languages that support them. For languages that don't support operator overloading, equivalent static methods are supplied.

The precision of the result of an operation depends on the operation itself. Addition and subtraction by default use the smallest absolute precision of their operands. Multiplication and division use the smallest relative precision of their operands. Details on the precision of specific operations can be found in the documentation.

Results are rounded to the nearest number. There is a high probability but no guarantee that the result is correctly rounded.

The desired precision and rounding mode can be overridden by passing additional arguments to the method that performs the operation. An AccuracyGoal value specifies the precision. The precision can be absolute or relative. By default, the precision is inherited from the operands. The RoundingMode determines how the result is rounded.

The BigFloat type supports generic arithmetic. The associated arithmetic type implements IIeeeOperations<T>, which means it supports all methods available for Double in the Math class, as well as special values such as PositiveInfinity, NegativeInfinity and NaN.

Reference