Example "B" compressed according to RLE4

Encoded data:

 14.  0F 11  00 00
 13.  02 11  09 00  04 11  00 00
 12.  04 11  03 00  03 11  02 00  03 11  00 00
 11.  04 11  03 00  04 11  02 00  02 11  00 00
 10.  04 11  03 00  04 11  02 00  02 11  00 00
  9.  04 11  03 00  04 11  02 00  02 11  00 00
  8.  04 11  03 00  03 11  02 00  03 11  00 00
  7.  04 11  03 00  01 11  03 00  04 11  00 00
  6.  04 11  03 00  01 11  03 00  04 11  00 00
  5.  04 11  03 00  03 11  02 00  03 11  00 00
  4.  04 11  03 00  04 11  02 00  02 11  00 00
  3.  04 11  03 00  04 11  02 00  02 11  00 00
  2.  04 11  03 00  03 11  03 00  02 11  00 00
  1.  02 11  0A 00  03 11  00 00
  0.  0F 11  00 00  00 01

This example also matches the Microsoft conventions for bitmap coding (e.g. line order, word alignment).

In comparison to RLE8 this example does not profit from 4 bit encoding and only demonstrates the principle. The reason is the special architecture collecting two 4 bit pixels together. A clearly better compression would be achieved, if data consist of alternating pairs of pixels. Such a type of contents is generated by certain dithering algorithms used for colour reduction.

To encode the example above 162 byte are required which is exactly the same data volume as for RLE8, only a smaller colour table is used. The size of the data part of an uncompressed version is 120 Byte.

 <   ^   >