Do shift registers have less propagation delay than feedback/forward nodes?

[Cory_K]

---

Idratherberacing wrote: 
... i had no issues until i started using feedback nodes, which led me to ask this question...

---

I dont think that a feedback node vs shift register would make that much of a time difference.

[Idratherberacing]

I'll know for sure when i recompile with the new subVI's.  I'll let everyone know what I come up with.  Thanks for the feedback so far.

[Ben]

---

Cory K wrote:

...

I dont think that a feedback node vs shift register would make that much of a time difference.

---

Somehow (No I can't explain it) reading that post prompted an image of Alice talking to the catapilar to pop into my head. Maybe it was a line that read somthing like "Things are often not as they appear around here" (poor paraphrase I am sure).

Ben

[Intaris]

---

The first implementation of the feedback nodes in LabVIEW 8.0 were a lot slower than a shift register because the person implementing them was creating completely new (and rather bad) code for them. In 8.2 or 8.5 the feedback node got redisigned to use internally exactly the same (highly optimized) algorithme that is also used for shift registers. As such I would not expect any measurable difference between the two but have never benchmarked them since I find the shift registers to be more intuitive to use.

Rolf Kalbermatter

Message Edited by rolfk on 08-12-2009 05:03 PM

---

Me too.

I implemented my first feedback node a week or so ago and I was so unsure about it I almost posted on the NI Forum to ask if I had understood it properly.  I decided against it because I didn't want to appear to be such a n00b.

Shane.

[tst]

Shift registers are more intuitive, but with the improvements FNs have recieved in 8.x (forward facing, initialized from the outside of a structure, doesn't require a loop), they make some things much easier and cleaner. There are some more details here.

[JICHFI]

If you need to keep data inside one specific case, to avoid wiring your shift register inside each case, it's better to use FNs.