This module corresponds to section 3.3 (Points) of the OpenGL 2.1 specs.
- pointSize :: StateVar GLfloat
- vertexProgramPointSize :: StateVar Capability
- pointSizeRange :: StateVar (GLfloat, GLfloat)
- pointDistanceAttenuation :: StateVar (GLfloat, GLfloat, GLfloat)
- pointFadeThresholdSize :: StateVar GLfloat
- pointSmooth :: StateVar Capability
- pointSprite :: StateVar Capability
- aliasedPointSizeRange :: GettableStateVar (GLfloat, GLfloat)
- smoothPointSizeRange :: GettableStateVar (GLfloat, GLfloat)
- smoothPointSizeGranularity :: GettableStateVar GLfloat
pointSize contains the rasterized diameter of both aliased and
antialiased points. The initial value is 1. Using a point size other than 1
has different effects, depending on whether point antialiasing is enabled
pointSmooth) or point sprites are enabled (see
are initially disabled.
The specified point size is multiplied with a distance attenuation factor
and clamped to the specified
pointSizeRange, and further clamped to the
implementation-dependent point size range to produce the derived point size
derivedSize = clamp (size * sqrt (1 / (a + b * d + c * d^2)))
where d is the eye-coordinate distance from the eye to the vertex, and a,
b, and c are the distance attenuation coefficients (see
If multisampling is disabled, the computed point size is used as the point's width.
If multisampling is enabled, the point may be faded by modifying the point
alpha value (see
instead of allowing the point width to go below a given
pointFadeThresholdSize. In this case, the width is further modified in
the following manner:
width = if derivedSize >= threshold then derivedSize else threshold
The point alpha value is modified by computing:
alpha = if derivedSize >= threshold then 1 else (derivedSize / threshold)^2
If point antialiasing is disabled, the actual size is determined by rounding the supplied size to the nearest integer. (If the rounding results in the value 0, it is as if the point size were 1.) If the rounded size is odd, then the center point (x, y) of the pixel fragment that represents the point is computed as
(x, y) = (floor xw + 0.5, floor yw + 0.5)
where xw and yw indicate window coordinates. All pixels that lie within the square grid of the rounded size centered at (x, y) make up the fragment. If the size is even, the center point is
(x, y) = (floor (xw + 0.5), floor (yw + 0.5))
and the rasterized fragment's centers are the half-integer window coordinates within the square of the rounded size centered at (x, y). All pixel fragments produced in rasterizing a nonantialiased point are assigned the same associated data, that of the vertex corresponding to the point.
If antialiasing is enabled, then point rasterization produces a fragment for each pixel square that intersects the region lying within the circle having diameter equal to the current point size and centered at the point's (xw, yw). The coverage value for each fragment is the window coordinate area of the intersection of the circular region with the corresponding pixel square. This value is saved and used in the final rasterization step. The data associated with each fragment is the data associated with the point being rasterized.
Not all sizes are supported when point antialiasing is enabled. If an
unsupported size is requested, the nearest supported size is used. Only size
1 is guaranteed to be supported; others depend on the implementation. To
query the range of supported sizes for antialiased points and the size
difference between supported sizes within the range, query
smoothPointSizeGranularity, respectively. For
aliased points, query the supported range with
The point size specified when
pointSize is set is always returned when it
is queried. Clamping and rounding for aliased and antialiased points have no
effect on the specified value.
A non-antialiased point size may be clamped to an implementation-dependent maximum. Although this maximum cannot be queried, it must be no less than the maximum value for antialiased points, rounded to the nearest integer value.
Controlling the Derived Size
The range to which the derived point size is clamped, see
that the size is further clamped to the implementation-dependent limits, see
smoothPointSizeRange. The initial range is
InvalidValue is generated if the
lower or upper bound of the range is set to a value less than zero. If the
lower bound is greater than the upper bound, the point size after clamping is
undefined, but no error is generated.
The constant, linear, and quadratic distance attenuation coefficients, see
pointSize. The initial coefficients are (1, 0, 0).
Controls whether point antialiasing is enabled. The initial state is
The smallest and largest supported size of aliased points.
The smallest and largest supported size of antialiased points.