12-06-2012 02:48 PM
Ok, I thought about something like that. Just one more question to be sure: what do I do with those "partial partial derivatives" I get from every column/row to make one partial derivative? I repeat for every y=0,1,2...256, I receive 256 dZ/dx's and how do I combine them in one?
12-06-2012 03:38 PM
@Oficmajster wrote:
... I receive 256 dZ/dx's and how do I combine them in one?
They are all independent, so you cannot really combine them without losing a lot of information. Please explain what you want at the end.....
12-06-2012 04:04 PM
I think I need an array, cause I need to make a double integral of it later. I just combined those 1D arrays into one 2D array. One for dx, and one for dy. The result seems acceptable.
What do you mean by loosing a lot of information, and how can I prevent that?
12-06-2012 04:10 PM - edited 12-06-2012 04:20 PM
You did not say what you mean by "combine into one". One number is insufficient, but one complex 2D array can hold the data just fine. Just wire your two arrays to the RE and IM input of RE/IM to complex, for example.
12-06-2012 04:19 PM
Ok... so if I took my 1D arrays and simply used Insert Into Array.vi to combine them into one 2D array, using FOR loop? And the very same algorithm with the second partial derivative.
12-06-2012 04:22 PM - edited 12-06-2012 04:23 PM
Way too much code!
You can autoindex on the 2D array at the left loop boundary (transpose if needed), then autoindex the 1D result on the right side to generate the 2D array. Don't wire N, it will be automatic based on the array size.
12-06-2012 04:28 PM
Roger that.
But this 2D array is my partial derivative, right? I mean: is it mathematically correct, that this array represent a function that is partial derivative from my basic function?
Sorry for taking so much of your time. You helped me a lot. Thanks 🙂
12-06-2012 04:37 PM - edited 12-06-2012 04:38 PM
12-07-2012 08:35 AM
Thanks a lot!