XMVector2Normalize function (directxmath.h)

Article
02/22/2024

Returns the normalized version of a 2D vector.

Syntax

XMVECTOR XM_CALLCONV XMVector2Normalize(
  [in] FXMVECTOR V
) noexcept;

Parameters

[in] V

2D vector.

Return value

Returns the normalized version of V.

Remarks

For a vector of length 0, this function returns a zero vector. For a vector with infinite length, it returns a vector of QNaN.

Note that for most graphics applications, ensuring the vectors have well-defined lengths that don't cause problems for normalization is common practice. However, if you need a robust normalization that works for all floating-point inputs, you can use the following code instead:


inline XMVECTOR XMVector2NormalizeRobust( FXMVECTOR V )
{
    // Compute the maximum absolute value component.
    XMVECTOR vAbs = XMVectorAbs(V);
    XMVECTOR max0 = XMVectorSplatX(vAbs);
    XMVECTOR max1 = XMVectorSplatY(vAbs);
    max0 = XMVectorMax(max0, max1);

    // Divide by the maximum absolute component.
    XMVECTOR normalized = XMVectorDivide(V, max0);

    // Set to zero when the original length is zero.
    XMVECTOR mask = XMVectorNotEqual(g_XMZero, max0);
    normalized = XMVectorAndInt(normalized, mask);

    XMVECTOR t0 = XMVector2LengthSq(normalized);
    XMVECTOR length = XMVectorSqrt(t0);

    // Divide by the length to normalize.
    normalized = XMVectorDivide(normalized, length);

    // Set to zero when the original length is zero or infinity.  In the
    // latter case, this is considered to be an unexpected condition.
    normalized = XMVectorAndInt(normalized, mask);
    return normalized;
}

Platform Requirements

Microsoft Visual Studio 2010 or Microsoft Visual Studio 2012 with the Windows SDK for Windows 8. Supported for Win32 desktop apps, Windows Store apps, and Windows Phone 8 apps.

Requirements

Requirement	Value
Target Platform	Windows
Header	directxmath.h (include DirectXMath.h)