At 03:19 PM 9/30/2002 -0400, Ann W. Harrison wrote:

>OK, time for some actual tests. Theoretically that guy is all wet.

The first index puts the less selective column (varchar 50 with
two distinct values) first. The second puts it last. Notice
the average data lengths....

Index B1 (0)
Depth: 3, leaf buckets: 510, nodes: 265954
Average data length: 1.00, total dup: 0, max dup: 0
Fill distribution:
0 - 19% = 0
20 - 39% = 1
40 - 59% = 0
60 - 79% = 0
80 - 99% = 509

Index B2 (1)
Depth: 4, leaf buckets: 4502, nodes: 265954
Average data length: 60.00, total dup: 0, max dup: 0
Fill distribution:
0 - 19% = 0
20 - 39% = 1
40 - 59% = 0
60 - 79% = 0
80 - 99% = 4501