... (echoing Hatef) and I never use those two methods, but rather a (small) modification to Crossrulz's method. Crossrulz noted that if you already have each channel's worth of points in a 1D array, you could simply use Build Array to create a 2D array (row 1 being Channel 1's points, row 2 being Channel 2's points).
However, it makes a difference to what you wire the 2D array. Suppose you wire it to a (Waveform) Chart. Recall that Charts expect to get their points one-at-a-time, and processes 2D arrays by row. So if you have two channels of 500 points and Build Array, you will have a 2-row, 500-column 2D array, and the Chart will think you are plotting 500 channels (all the data in Row 1) with two data points (the rows). Oops. Fortunately, there's the Transpose 2D Array function on the Array Palette which changes the array to one of two columns (= 2 Channels) and 500 rows (= 500 points), which plots just fine.
Got it? OK, here's a Quiz -- what if (knowing you are dealing with 500 points-at-a-time instead of one-at-a-time) you decide to use a Graph instead of a Chart? [Warning -- this is a Trick Question]. Because the Graph expects rows to be Channels (or "plots"), you do not want to use the Transpose 2D Array when plotting multi-channel graphs.
A fun thing to do is to take Hatef's nice example and add on the Crossrulz/Schor Method, and to "expand" both by creating a set of Graphs in addition to Charts.
I hope the somewhat-long-winded explanation of "Why did they do it that way?" that I provided makes sense to you. Just remember that 2D arrays are usually processed a row at a time, and ask yourself how LabVIEW made the 2D Array, was it a point-at-a-time (= Charts) or a plot-at-a-time (= Graphs). And, if you can't figure it out, just try it and see if it makes sense -- if you instead see hundreds of different plots, haul out Transpose 2D Arrays and see if that fixes things.
Bob Schor
