Inconsistent times are one thing, overall slowing down is another, although they may have common source. I guess it is the XY plot in your case. Do you have it
AutoScaled? If so, try turning autoscaling of both axes off.
Secondly, I don't quite understand the reason for the inner loop with
number of iterations = constant 1. What happens if you remove the loop? I understand you want to change the number of iterations in future, but let's test the simplest things first. Removal will require indexing the
Wavelength array explicitly with an apropriate block. By the way, it is customary that you
EITHER define the number of iterations of a FOR loop by connecting the N input
OR by indexing of an input array -
NEVER both at the same time. Maybe, changing the multiple XY plot into a single XY plot won't be necessary, maybe it will, try for yourself.
Thirdly, I suggest you try enclosing the XY plot block alone in a case structure (say, in its TRUE window) and set the true condition to occur at some fraction of the total iterations, not every cycle. You do this using the index
"i" of the main iteration loop, calculate modulo something (e.g. 10) and test if it is true or false. This way you update the XY plot every 10th iteration only and by choosing different modulo divider you can test, if the plot's display is or isn't the beast slowing things down.
Finally, I suspect it is not necessary to use a
Local Variable of the seconds 2 indicator. How about simply wiring the output of your
To-double block to the input of the
subtraction block? That should let you remove
seconds 2 local variable with no harm done.
Daniel
