One of the major differences with working with the 2D arrays and the full data clusters is that the timing is in true values. We recommend that you use microseconds throughout your programs and convert appropriately when desired. In doing so one will encounter situations where a pulse as entered in the 2D array can no longer be represented by a single slot on the generator. As an example consider that you have selected the 10MHz clock and want to create a pulse of 5 ms that is 50000 clock units, then one needs to split this pulse into two slots. As a user you do not want to be bothered with this and hence FLCWFG Partition Raw data.vi does the job for you.
A situation as described above can appear quite easily when you scale timing data, which can be done with FLCWFG Scale Data.vi. This VI takes a set of scaling factors and changes the 2D data accordingly, but be sure to partition the data before you mold it back into a full data cluster which is needed for downloading operations. The figure shows you a call chain describing the above situation.