Muokkaa

Jaa


Vector128.MaxNumber<T>(Vector128<T>, Vector128<T>) Method

Definition

Compares two vectors, on a per-element basis, to compute which is greater and returning the other value if an element is NaN.

public:
generic <typename T>
 static System::Runtime::Intrinsics::Vector128<T> MaxNumber(System::Runtime::Intrinsics::Vector128<T> left, System::Runtime::Intrinsics::Vector128<T> right);
public static System.Runtime.Intrinsics.Vector128<T> MaxNumber<T> (System.Runtime.Intrinsics.Vector128<T> left, System.Runtime.Intrinsics.Vector128<T> right);
static member MaxNumber : System.Runtime.Intrinsics.Vector128<'T> * System.Runtime.Intrinsics.Vector128<'T> -> System.Runtime.Intrinsics.Vector128<'T>
Public Function MaxNumber(Of T) (left As Vector128(Of T), right As Vector128(Of T)) As Vector128(Of T)

Type Parameters

T

Parameters

left
Vector128<T>

The vector to compare with right.

right
Vector128<T>

The vector to compare with left.

Returns

A vector where the corresponding element comes from left if it is greater than right; otherwise, right.

Exceptions

The type of the elements in the vector (T) is not supported.

Remarks

For IFloatingPoint<TSelf> this method matches the IEEE 754:2019 maximumNumber function. This requires NaN inputs to not be propagated back to the caller and for -0.0 to be treated as less than +0.0.

Applies to