Hi rick,

>Who's going to look at 1,000,000 records? On screen or in print?

>Database 101 - only pull what you need and no more...

Well, one million records would be a bit too much for most cases, but if you're doing statistical analysis, it is not unusual to have lots of data points.

what about more practical problems, like having downloading tables that have large memofields or blobs in them ? For example it contains images that could be a several 100K or a few MB each the problem of non disk flushing cursors becomes more obvious very quickly.

Walter,