MDX Script: Calculated Members (vs Named Sets)

In the previous blog post I discussed the benefits of defining MDX calculations in the cube (instead of in the query using the WITH-clause). The primary benefit was having named sets computed ahead of time (when the MDX Script is executed) and available via the global cache which greatly improves query response time. The first version of that post (corrected later) neglected to specify that it was only the named set calculations that were "computed and cached" when the MDX Script was executed.and *NOT* the calculated members.which is a very important distinction (thanks again to Chris Webb for pointing out that ambiguity).