LOD Component

The Level-of-Detail (LOD) component switches child objects named LODn (with n from 0 to 4) on and off, depending on how close this object is to the main camera.

The LOD component is used to reduce the performance impact of complex objects when they are far away. To do so, a LOD object consists of multiple states, from highly detailed to very coarse. By convention the highly detailed object is called LOD0 and the lesser detailed objects are called LOD1, LOD2, LOD3 and finally LOD4.

This component calculates how large it roughly appears on screen at it's current distance. This is called the coverage value. Using user defined coverage thresholds, it then selects which LOD child object to activate. All others get deactivated. The LODs have to be direct child objects called LOD0 to LOD4. Other child objects are not affected.

How many LODs are used depends on the number of elements in the LodThresholds array. The array describes up to which coverage value each LOD is used. Thus if it contains one value, two LODs will be used, LOD0 for coverage values above the specified threshold, and LOD1 at lower coverage values.

To see the current coverage, enable the debug drawing.

The coverage calculation uses spherical bounds. It should be configured to encompass the geometry of all LODs.

To prevent LODs switching back and forth at one exact boundary, the LOD ranges may overlap by a fixed percentage. This way once one LOD gets activated, the coverage value has to change back quite a bit, before the previous LOD gets activated. Since this behavior can make it harder to set up the LOD thresholds, it can be deactivated, but in practice it should stay enabled.