vpRenderRawVolume, vpRenderClassifiedVolume - render a volume
- vpContext *vpc;
- vpContext *vpc;
- VolPack context from vpCreateContext.
These routines are used to render a volume using the parameters in a rendering
renders the data in the current voxel array
and stores the result in the current image array. If an octree is present in
the context (see vpCreateMinMaxOctree(3)) then it is used to accelerate
is identical in function except
that it renders the current preclassified volume (see vpClassifyVolume(3))
instead of the voxel array, and the octree is not used.
should be used when the same volume is rendered
multiple times with the same opacity transfer function. This rendering mode is
the fastest. The opacity transfer function must not change because the voxel
opacities must be fixed to compute a preclassified volume.
should be used when the volume data or the opacity
transfer function changes in between renderings. If the volume data remains
fixed then a min-max octree should be computed to accelerate rendering.
Rendering is significantly faster with an octree than without, although it is
still slower than rendering with vpRenderClassifiedVolume
Before the rendering routines are called the rendering context must be
initialized to set the volume parameters, classification parameters (
only), viewing parameters, shading and lighting
parameters, and image array. See VolPack(3) for a list of relevant functions.
One important state variable should be adjusted before calling the rendering
routines: the maximum opacity threshold. This number specifies a threshold
value for the opacity of an image pixel. If the opacity of the pixel reaches
the threshold then no more voxels are composited into the pixel. The threshold
should be a number slightly less than one (0.95 is a good value); numbers
closer to 1.0 result in longer rendering times, but less error. The value of
the threshold is changed by calling vpSetd
with the VP_MAX_RAY_OPACITY
option. The default value is 1.0.
There are two additional state variables that can be adjusted to set the size of
an internal data structure. During rendering, a work buffer called the
intermediate image is used to hold temporary results. The size of this data
structure depends on the size of the volume and the current view
transformation matrix. Over the course of an animation sequence the
intermediate image may have to be enlarged multiple times, resulting in many
calls to the memory allocator. Normally the overhead of reallocating the data
structure is negligible and can be ignored. However, this overhead can be
eliminated by specifying hints for the maximum size of the intermediate image.
The data structure will then be allocated once, and will never be reallocated
unless an even larger intermediate image is required. To set the size hints,
with the VP_INT_WIDTH_HINT and VP_INT_HEIGHT_HINT state
variable codes. If the viewing transformation does not include any scaling,
then the maximum possible size of each dimension of the intermediate image is
twice the size of the largest dimension of the volume. To get a more precise
bound, the intermediate image size required for rendering a volume with the
current viewing parameters can be found by calling vpGeti
VP_INTERMEDIATE_WIDTH and VP_INTERMEDIATE_HEIGHT state variable codes.
The normal return value is VP_OK. The following error return values are
- The shading lookup tables are sized incorrectly, or one of the voxel
fields used to index the tables is invalid, or the number of material
types is invalid.
- One of the current view transformation matrices is singular.
- The volume size has not been set or is invalid, or there is no volume
- The voxel size or the voxel fields have not been specified or have been
- The opacity transfer function tables have invalid sizes or are associated
with invalid voxel fields or have been incorrectly specified. (
- The angle between the shadow light vector and the viewing direction is too
large (must be less than 45 degrees). Either change the light direction or
disable shadows (see vpSetShadowLookupShader).