DD::Image::VArray Class Reference

#include <VertexContext.h>

Public Member Functions
	VArray (const ChannelSetInit c)
	VArray (const ChannelSet &c)
	VArray (Channel c)
	VArray (const VArray &b)
float &	operator[] (Channel z)
const float &	operator[] (Channel z) const
float *	array ()
VArray &	operator= (const VArray &v)
Vector4 &	P () const
float	x () const
float	y () const
float	z () const
float	w () const
Vector3 &	PL () const
Vector3 &	PW () const
Vector3 &	MB () const
Vector4 &	UV () const
Vector3 &	N () const
Vector4 &	Cf () const
float	r () const
float	g () const
float	b () const
float	a () const
Vector3 &	VEL () const
void	lerp (const VArray &b, float t)
void	offset (const VArray &b, const VArray &delta, float t)
void	scale (float s)
void	offset (const VArray &b, const VArray &dx, float x, const VArray &dy, float y)
void	midpoint (const VArray &a, const VArray &b)
void	w_normalize ()
	Divide all the channels by w, and put 1/w into w.

Public Attributes
ChannelSet	channels
float	chan [VARRAY_CHANS]

Detailed Description

Array of floating point numbers that are interpolated across a primitive. A VertexContext contains 4 of these: the current value, the derivatives horizontally and vertically, and the value at X=0.

Member Function Documentation

lerp()

void DD::Image::VArray::lerp ( const VArray &  b, float  t  )

inline

Destructive linear interpolate between this and VArray b by t.

offset() [1/2]

void DD::Image::VArray::offset ( const VArray &  b, const VArray &  delta, float  t  )

inline

Sets it to b+delta*t

scale()

void DD::Image::VArray::scale ( float  s )

inline

Scale the VArray.

offset() [2/2]

void DD::Image::VArray::offset ( const VArray &  b, const VArray &  dx, float  x, const VArray &  dy, float  y  )

inline

Sets it to b+dx*x+dy*y

midpoint()

void DD::Image::VArray::midpoint ( const VArray &  a, const VArray &  b  )

inline

Sets it to the midpoint between a and b ((a+b)/2).

w_normalize()

void VArray::w_normalize

(

)

Divide all the channels by w, and put 1/w into w.

Divide all the channels by w, and put 1/w into w. The resulting values can then be linearly interpolated in screen space. Don't forget to divide by w to get the original values back, unless the divided value is useful (which is only true for xyz).

Referenced by DD::Image::rPoint::add_to_render().