Rube Goldberg Code

[crossrulz]

---

@alexderjuengere wrote:

Why is there a timeout case that can never happen?

- because, I don't want any timeout effects, can it be removed?

---

Yes.  Just delete the event case.

---

@alexderjuengere wrote:

Why aren't the buttons latch action?

- because you can't use local variables with that mechanical action

---

Do you actually need the Local Variables?  You've been around long enough to know you should avoid them.

[altenbach]

I somehow missed that this discussion was in the RG thread. I was looking for the link to the quote thread.... 🙂

[JÞB]

---

wiebe@CARYA wrote:

---

@Henrik_Volkers wrote:

---

@altenbach wrote:

In a nutshell, we basically have a 2D boolean array to be turned into a single boolean according to some rules.

I have that nagging feeling that this could be simplified somewhat. 😄 (seen here)

---

shiftregisters and one OR 🙂   just keep the boolean information (edit: first SR init to false 😉 )

---

Put the indexing on Last Value, and make it conditional:

---

Nest auto indexed loops and Just enable the loop conditional terminals and STOP THE LOOPS when the first TRUE occurs. Output last value. 

Better yet: there should be a way to use recursion for this and scale it for n dim array where n is known only at runtime (Xnode, vim maybe?  I'd have to play)  just OR each autoindexed element with the previous result on a FBN. NOTE the nth+1 dimension loop will execute 0 times returning the default output F to the nth loop, so the whole dang thing stops on the first T in any dimension

for dimensions (n to 1), n--

[altenbach]

---

@JÞB wrote:
Nest auto indexed loops and Just enable the loop conditional terminals and STOP THE LOOPS when the first TRUE occurs. Output last value. 

---

Sometimes this is fine, but not always. In the original code, a lot of other things occur in the loop, including file writing and an inner FOR loop doing DAQ. I cannot tell if stopping early would be correct in that scenario. (I might study it if the code were cleaner, but there are too many hairballs... 😄 )

[JÞB]

---

@altenbach wrote:

---

@JÞB wrote:
Nest auto indexed loops and Just enable the loop conditional terminals and STOP THE LOOPS when the first TRUE occurs. Output last value. 

---

Sometimes this is fine, but not always. In the original code, a lot of other things occur in the loop, including file writing and an inner FOR loop doing DAQ. I cannot tell if stopping early would be correct in that scenario. (I might study it if the code were cleaner, but there are too many hairballs... 😄 )

---

Oh, I assumed modular code.  Silly me.

So much for just sorting and testing the default (last) element output of delete from array 😞 as an OR Array Elements replacement.   HMMM. I wonder if there is any case where that could be faster than OR Elements. Certainly either would be simpler than the example shown.   That "=0, Negate" bothers me when "≠0" is right there. 

[altenbach]

Logging four scalar (!) values to a file every 10 seconds apparently needs conversion to arrays, to dynamic data,  a diagram full of express VIs and a handful of local variables. (seen here)

[altenbach]

I am not sure where the OP dug out that old example, but turning a 1D array into a 2D array with one column does NOT need a formula node!

Seen here

[altenbach]

Sometimes I feel evil. 😄

[JÞB]

---

@altenbach wrote:

Sometimes I feel evil. 😄

---

X(radians)+pi?  Aren't there 2pi radians in a circle?  I'm sure you can find a complex solution that fits on a postage stamp (of course that might be a postage stamp that calls itself reentrantly)

[altenbach]

---

@JÞB wrote:

---

@altenbach wrote:

Sometimes I feel evil. 😄

---

X(radians)+pi?  Aren't there 2pi radians in a circle? 

---

That pi gets subtracted again later. It just makes sure that the remainder is mapped into -pi..pi instead of 0..2pi.