Unless I'm just totally missing something (which is very possible), I think I've found a bug. It appears when trying to replace all the elements of a 1-D array of error clusters with an empty error cluster (using a for loop and the "Replace Array Subset" function). Please see the attached picture.
That I *had* already picked up on. I think we were worried about memory allocation, since this VI is running in LV RT (i.e. we wanted to allocate the memory once in the "Init" case and then just do everything else in place.