Raster data files are representations of continuous values that are regularly sampled in rows and columns. The values may also be organized into multiple layers, though this is far more likely to be the case with images than with rasters. The
most common measurement is elevation in what is known as a DEM for Digital Elevation Model, but there are many other values that can be represented. The symbolic coloring can be set up to change gradually, or reflect discrete jumps based on distinct
classes of values. Hillshading is an option for display which adds shadowing as if the raster represented elevation values and the surface was being illuminated from an angle.
The beginner data access class is the Raster, and the Raster<T> class is the expert tool. Using the advanced tool requires knowledge of what data type is in the data file, but allows retrieving a window of data from the file without loading the
entire raster into memory. Raster<T> also provides a strong typed jagged array of data values for faster data access without converting the original values into doubles. With the Raster class, the Values property is indexed by row and column
and allows easy access regardless of what the underlying raster format is, but this comes at a performance penalty for converting the values into doubles and working through a property accessor.