LabVIEW "myths"

[Yamaeda]

I assume the sub-vi myth is "_noticably_ slower". 😉
It's the same as some people believing object oriented code to be much slower, it might be in some instances but the benefits truly outweighs the downsides.

A classic:
LabVIEW is not a true/full programming language. (which you might believe if you only wire the virtual instruments and measurements)

/Y

[Cory_K]

---

smercurio_fc wrote:

---

PaulG. wrote:

---

for(imstuck) wrote:

... Shift registers take time to pass data from one side of the loop to the other.

---

And the wider the loop the longer the data takes to go from one side to the other!

---

I thought it had to do with trying to stuff a whole lot of data into an itty bitty thin wire. 

---

Im pretty sure thats why they stopped using the curly kind of telephone cord.

It takes FOREVER to go through all of those loops, straight wires are much faster 

[Ray.R]

Here's a myth that I often hear...

Execution goes from the leftmost objects (controls / indicators) to objects on the right within the block diagram.

Or to word it differently:

Placing objects to the left will execute before objects to the right.

FALSE: 

Any object on the Block Diagram can execute as soon as data is available.

Message Edited by Ray.R on 02-27-2010 04:28 PM

[GregFreeman]

---

Ray.R wrote:

Here's a myth that I often hear...

Execution goes from the leftmost objects (controls / indicators) to objects on the right within the block diagram.

Or to word it differently:

Placing objects to the left will execute before objects to the right.

FALSE: 

Any object on the Block Diagram can execute as soon as data is available.

Message Edited by Ray.R on 02-27-2010 04:28 PM

---

Unless, of course, you use a dreaded flat sequence structure

[Ray.R]

that's cheating 😉

[Henrik_Volkers]

1. C is faster than G (is a myth?)

2. Execution speed is alphabetical:

 Assembler

 Basic 

 C

 Delphi

 Elan

 Fortran

 G 

 HPVee

 ...

 ...

Sidenote: I looked up E in wiki and didn't found G (nor Brainfuck) in the list

[MattBradley]

But since HP Vee is now Agilent VEE, does that mean it's faster than assembler?  (ducking)

[Ben]

Here is a myth that nobody cares about is is wiling to verify.

"Adding extra connection to your icon connector does not affect its speed."

FALSE (sorta)

I measured that using a pair of PXI controllers andalong with high speed counters and one of those fancy oven-controlled oscillators and syncing the two chassis together. We were not able to measure the over-head of a single call but we did manage to measure a time difference when doing multiple calls and if my memory serves me it was something along the lines of 8 picoseconds per un-wired input. So if you are try to close a fast RT loop, don't use extra terminals.

So the correct answer yes they take time, the practical answer is "it takes so little time it is hardly worth worrying about".

Ben

[Ray.R]

So that's not a myth, Ben 😉

Same as having wires with many bends add to the VI size.  It does.. I posted something on this many years ago.  I don't think it slows down performance.  At least I never measured it.  I might say (unconfirmed) that wire bends slow down a VI is a myth.  😉

Are we going to have LabVIEW MythBusters?

[Hornless.Rhino]

---

Ray.R wrote:

So that's not a myth, Ben 😉

Same as having wires with many bends add to the VI size.  It does.. I posted something on this many years ago.  I don't think it slows down performance.  At least I never measured it.  I might say (unconfirmed) that wire bends slow down a VI is a myth.  😉

Are we going to have LabVIEW MythBusters?

---

Someone should make a VI that outputs the monalisa in a picture control, powered by a DB that itself looks like the monalisa (using different types of wires to get different colours etc)