Not an answer to the question, but the array manipulation in the examples you attached really needs to be commented;
You have an array of clusters containing two double precision numbers and want to convert it into a 2D single precision array to write to a file...
You wire the cluster array to a FOR loop and initialize two shift registers with empty 1D arrays...and then build the two arrays using insert array element...when the for loop is finished you build a 2D array of the two 1D arrays using a build array function...This is the absolutely worst way to do the job. This will resize the two 1D arrays on every iteration an operation that is heavy on memory operations and very slow. In addition you wire the arrays to single precision indicators...causing the whole a
rrays to be converted to SGL in a new very memory costly operation.
Never do this, instead initialize the shift registers with an array already containing the correct number of elements, and then use the replace array subset instead...OR in this case (if you really needed the 1D arrays), forget about the shift register and use the auto indexing of the for loop to produce the output arrays...
Here though you don't really need the 1D arrays, what you want to end up with is a 2D array...the fastest and most memory efficient way to do this is thus to never create those 1D arrays at all...instead initialize a shift register in the for loop with a 2D SGL array with the correct number of elements and then use replace array subset to insert the two new numbers (converted to SGL) on each iteration....
I'll attach an example later.
I would recommend the optimization section of the LabVIEW manual..reading and following it will improve the performance of your code drastically. Or read
:
http://zone.ni.com/devzone/conceptd.nsf/webmain/732CEC772AA4FBE586256A37005541D3?opendocument&node=DZ52068_US
