0
@@ -709,4 +709,26 @@
0
 In this example, @\[*D] sup 1@'s value of 26 is the smallest.
0
 Therefore, when compressing this set of samples in a FIXED subframe,
0
 it is best to use a predictor order of 1.
0
+.PP
0
+The predictor order indicates how many warm-up samples to take from
0
+the PCM stream.
0
+Determining the residual values can then be done automatically
0
+based on the current @"Sample" sub i@ and previously encoded samples, or
0
+warm-up samples.
0
+.TS
0
+tab(:);
0
+| c | c |
0
+| c | c |
0
+| c | l |.
0
+_
0
+Predictor:Calculation
0
+Order:\^
0
+=
0
+0:@"Residual" sub i = "Sample" sub i@
0
+1:@"Residual" sub i = "Sample" sub i~~-~~"Sample" sub {i~-~1}@
0
+2:@"Residual" sub i = "Sample" sub i~~-~~((2 * "Sample" sub {i~-~1})~~-~~"Sample" sub {i~-~2})@
0
+3:@"Residual" sub i = "Sample" sub i~~-~~((3 * "Sample" sub {i~-~1})~~-~~(3 * "Sample" sub {i~-~2})~~+~~"Sample" sub {i~-~3})@
0
+4:@"Residual" sub i = "Sample" sub i~~-~~((4 * "Sample" sub {i~-~1})~~-~~(6 * "Sample" sub {i~-~2})~~+~~(4 * "Sample" sub {i~-~3})~~-~~"Sample" sub {i~-~4})@
0
+_
0
+.TE