02-23-2017 06:33 PM
I provide my users the ability to enter formulas based on recorded data and pass those formulas into the Eval Multi-Variable array VI (After performing name substitutions and validation). These formulas tend to consist of 1 to 3 source data channels mixed with basic math (EX: 5*a+(2*b - c) + 42). That kinda thing... The exact formulas are not readily predictable.
The trouble is, this VI is terribly slow for large data sets (Around 10-20 times slower than a formula node for the functions I've been using). I'm sure this has everything to do with the compiled nature of formula nodes vs the parsed formula evaluation method the Eval Multi-Variable Array VI uses.
Are there any techniques to improve the performance of this VI?
02-24-2017 10:06 AM
Here's a quick contrived demo of the performance differences (And generally how I am feeding channel data into the Eval VI).