I would like to add that in 10G you still can battle correlation problems using higher levels of dynamic sampling (to certain degree).

In 10.2.0.3 the correct cardinality was reported on level 4:

SQL> select /*+ dynamic_sampling(4) */ count(*) from t_vc where n1=10 and n2=10;

Execution Plan
———————————————————-
Plan hash value: 2872596891

—————————————————————————
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
—————————————————————————
| 0 | SELECT STATEMENT | | 1 | 6 | 8 (0)| 00:00:01 |
| 1 | SORT AGGREGATE | | 1 | 6 | | |
|* 2 | TABLE ACCESS FULL| T_VC | 100 | 600 | 8 (0)| 00:00:01 |
—————————————————————————

Predicate Information (identified by operation id):
—————————————————

2 – filter(“N1″=10 AND “N2″=10)

Note
—–
– dynamic sampling used for this statement