Nonlinear Simplex Algorithm too slow

[zerotolerance]

Is there a way of speeding up the attached VI. Its just unrealisticly slow. It can't even estimate 1 point, let alone 1000.

[Jack._W]

Hello,

I have had a look at your VI and have some questions.

When I run your code the program seems to run smoothly until it gets to the 'Downhill Simplex nD.vi'. On my PC with the reference and sample text files you have provided the program will run to that VI which it then holds at indefinitely. Is this the same problem you are seeing?

If you could also give me a little more information on what the code is doing that may be helpful. Has it been functioning correctly or faster before a change? What has changed?

Kind regards

[zerotolerance]

Hello Jack, thanks for the reply, go to the following link. Its another discussion about this.

http://forums.ni.com/t5/LabVIEW/complicated-curve-fitting/td-p/1748948/page/2

My latest attachment "ZIP" contains another form of this Simplex algorithm. This time, it accepts objective function instead of formula string. So it makes things much simpler and easier to follow. But even with this method it still hangs indefinetly, and I don't understand why. When I used the same objective function using the Lev-Mar algorithm, tis worked fine, but not with this Simplex algorithm.

The "Main" Vi also contains some comments on what I'm trying to do. Hopefully it makes it easier to understand.

kas