The following images are for the Missouri-A model. The method used was
linear prediction, with 3 ancestors, and values were quantized AFTER linear
prediction. (A) is the original Missouri-A model, while (B)-(C) results
(8 and 10-bit) are omitted, and (D)-(F) are its 12-16-bit counterparts.
The original Missouri-A model is 2,791,268 bytes, while compression ratios
are below their respective image.

A
B

CR = 57.99 : 1 size = 48,136 bytes
C
D

CR = 60.39 : 1 size = 46,219 bytes
CR = 52.34 : 1 size = 53,327 bytes
E
F
If you have comments or suggestions, email me at scottm@meru.cecs.missouri.edu