Suggestions for simulation output changes?

[Jonathan]

A user (and instructor) recently suggested some changes to the output format including the following.  I am wondering if people in the community think they would be useful or preferable to the current output (which is modeled after @Risk circa 2001).

1. Standard error of the mean reported for each output
2. 95% confidence interval around the mean reported for each output
3. In the simulation dialog box, users can select their desired granularity of percentile outputs (5%, 2%, 1%, etc.)
4. Percentiles are reported in a separate table from the means and standard deviations.  This table has outputs in columns and percentiles in rows, transposing the current percentile display.

Apparently his main interest in finer-grain percentiles is in tail probabilities, so another option could be to display the tails percentiles at finer granularity than the default 5% used in the middle of the distribution.

Again,  your comments and suggestions are welcomed.