Size of finest-resolution (base) chunk in object coordinates.
Axis-aligned bounding box lower bound of object in object coordinates to use for clipping and level-of-detail calculations.
Axis-aligned bounding box upper bound of object in object coordinates to use for clipping and level-of-detail calculations.
Specifies the number of levels of detail (as lodScales.length), and the resolution in object
coordinates for each level of detail. If lodScales[lod] === 0, then level-of-detail lod
does not exist.
Level of detail 0 is the finest resolution.
The non-zero values must be non-decreasing.
For each chunk, the chosen lod is the largest value such that
lodScales[lod] <= detailCutoff * pixelSize, where pixelSize is the maximum spatial distance
spanned by a single viewport pixel within the projected image of the chunk.
Row-major [n, 5] array where each row is of the form [x, y, z, start, end_and_empty], where
x, y, and z are the chunk grid coordinates of the entry at a particular level of detail.
Row n-1 corresponds to level of detail lodScales.length - 1, the root of the octree. Given
a row corresponding to an octree node at level of detail lod, bits start specifies the row
number of the first child octree node at level of detail lod-1, and bits [0,30] of
end_and_empty specify one past the row number of the last child octree node. Bit 31 of
end_and_empty is set to 1 if the mesh for the octree node is empty and should not be
requested/rendered.
C order [numLods, 3] array specifying the xyz vertex position offset in object coordinates
for each level of detail.
Generated using TypeDoc
Starting corner position of (0, 0, 0) chunk in object coordinates.