@crossrulz wrote:
You may need to use Transpose 2D Array before and after the Build Array to get it to line up the way you want.
Here's some background to the point crossrulz is making:
A 2x8652 array means 2 rows, 8652 columns. Total of 17304 elements. These are stored in continguous memory somewhere in 17304 consecutive memory locations. LabVIEW keeps track of how to navigate that memory to treat the values as a 2D array, but they are always *literally* stored in a 1D arrangement.
Next point, let's consider a simpler 2x3 array. Rows 0,1, and Columns 0,1,2. Here is the order that LabVIEW stores them in memory:
R0C0,R0C1,R0C2,R1C0,R1C1,R1C2.
Rows take precedence, so all of Row0 comes first, then all of Row1, etc.
If you think about the nature of this layout, you can start to see why it's not *natural* to append 2D arrays in the manner you'd like. There's no choice but to copy and move around in memory virtually *all* the data that was previously stored.
However, if you originally stored the *transpose* of your 2D array, it'd be 8652 rows x 2 columns. Then it would be much more natural to append another 8652 rows x 2 columns without touching any of the data that was previously stored.
Hence the tip from crossrulz that transposing might be needed.
(Behind the scenes, the LabVIEW compiler is pretty smart with memory management. When it sees that you are appending to an array in a loop, it will probably know to allocate a big memory space in the first place that won't get filled up for many iterations. The transpose approach will let you append naturally for a while before LabVIEW will eventually need to allocate more memory.)
-Kevin P
ALERT! LabVIEW's subscription-only policy came to an end (finally!). Unfortunately, pricing favors the captured and committed over new adopters -- so tread carefully.
