I just want to elaborate more for you guys so you understand how I got led on the path of the measures being the issue instead of the date conversion. Where the date conversion is the root of the problem, tinkering with the measures was having an affect on the results. ie: there was still a bit of a relationship if you will.
For example my script type calculation view contains;
1) SCRIPT BOX
Contains SQL code including my many attributes, 2 calculated measures and 2 bad date conversions. All of these fields were outputted to my 'Output of script'
2) OUTPUT BOX
This box contains a list of all the objects derived from my 'output of script' but here I experimented with removing these from the TRUE final output list.
Now, removing all but a single attribute from the OUTPUT box made the view run without error. (Whilst leaving the bad date conversion still in step 1!). When I added fields back one at a time I noticed when I added the calculated measures back the view would again fail. So even though I have kept the bad date conversion in the step 1) script box I was able to get the error to occur or disappear by tweaking the calculated measures. It's as though the resources required to do the bad date conversion was affecting how much data could be also converted in the calculated measure. So there was a relationship. But removing the bad date conversion completely from the script box then allowed my calculated measures to work fully.
I hope this makes sense, it's difficult to explain textually.
-Patrick
