XMFLOAT3PK::XMFLOAT3PK(constfloat*) function (directxpackedvector.h)

Article
02/22/2024

Initializes a new instance of XMFLOAT3PK from a three element float array argument.

This constructor initializes a new instance of XMFLOAT3PK from a three element float array argument.

Note This constructor is only available under C++.

Syntax

void XMFLOAT3PK(
  const float *pArray
) noexcept;

Parameters

pArray

Three element floating point array containing the values used to initialize the three components of a new instance of XMFLOAT3PK.

Return value

None

Remarks

Values contained in pArray[0] and pArray[1] are stored, respectively, in the x-component and the y-component of the new instance of XMFLOAT3PK.

The values obtained from pArray[0] and pArray[1] are transformed from the standard 32 bit floating point format (sign bit, 8 bit exponent, 23 bit mantissa), and stored as an 11 bit floating point format (5 bit exponent, 6 bit mantissa).

Value contained in pArray[2] is stored, in the Z-component the new instance of XMFLOAT3PK. The value obtained from pArray[2] is transformed from the standard 32-bit floating point format (sign bit, 8 bit exponent, 23 bit mantissa), and stored as a 10 bit floating point format (5 bit exponent, 5 bit mantissa).

As no target formats do not support a sign bit, all member of pArray must be greater than zero.

Because of the change in floating point format during the instantiation of an instance of XMFLOAT3PK by this constructor, some loss of precision can be expected.

Requirements

Requirement	Value
Target Platform	Windows
Header	directxpackedvector.h