The viewer can use base layer groups to optimize image rendering times. Layers in a base layer group are rendered together to generate a single raster image that can be sent to the viewer. The image is divided into tiles so only the required tiles need to be rendered and sent, instead of the entire image. Tiles are cached on the server; if a tile already exists in the cache it does not need to be rendered before being sent.
Each base layer group has a series of pre-defined scales that are used for rendering. When a request is made to view a portion of the map at a given scale, the viewer renders the tiles at the pre-defined scale that is closest to the requested map view scale.
Layers within a base layer group are rendered together. Visibility settings for individual layers are ignored and the visibility setting for the group is used instead.
Layers above the base layers will generally be vector layers with transparent backgrounds. This makes the images small and relatively quick to load in the viewer.
You may have more than one base layer group. Lower layers will be hidden by higher layers unless the higher layers have transparent areas or have their visibility turned off.
A layer can only belong to one group at a time. It cannot be part of both a base layer group and a regular group.