"Retain Wire Values" Issue

[Mr._Bob]

Recommend that the “retain wire values” feature continue to retain the values EVEN AFTER the program pauses, errors-out, stops, or is aborted by the programmer. What good is it if the program stops and you can’t probe the history? The system was maintaining the values up to that point anyway. There should be no reason why those values cannot be retained in memory after a stoppage, at least until someone attempts to make a change to the program.

I determined that a workaround for this issue is to create a separate while loop the never stops. Then, when the program comes out of the main loop with an error, the separate while loop keeps the program running so you can see all the retained values. Naturally, lines that did not execute have no line info.

Bob
 

[altenbach]

The "retain wire values" feature retains the wire values from the last run. You can place a probe after stopping and you should get the last wire value. If it does not happen, you found a bug and you should report it. Please start a discussion in the LabVIEW forum and post some simple example code so we can investigate.

(Wire values are of course erased if the VI needs to recompile, e.g. after you start editing the code.)

[Mr._Bob]

Replies appreciated. There is no program "bug". My expectation is to move the cursor to a line and see the value in the same manner as when the program is running (i.e. it will indicate the real-time value of the line WITHOUT invoking the probe "window".)

[altenbach]

Ah, OK that clarifies the issues. From the original post, you said that the values are not retained, but they are! You are talking about the yellow value tip strips, not probes.

The problem is that the "tip strip" feature has other functions in edit mode and e.g. shows the name of terminals and such. We probably need to carefully investigate if there are potential conflicts in edit mode. If there are, maybe the value tips could be shown in run mode (i.e. after pressing ctrl+m) while the VI is not actually running.

In principle I like the idea of showing retained wire values as tip strips even if the VI is in edit mode. Let's sort out potential problems first, though.

[Mr._Bob]

I also noticed that variables that are counting at high rates of speed do not display any information at all on the "yellow value tip strips". Bit if you look closely, you can see the line oscillating as if it wanted to display the value, but can't. In that case, a value displayed every 100 msecs would be nice.

[AristosQueue (NI)]

Tip strips never show values to the best of my knowledge. Ever. We don't use tip strips for debugging. That's what probes are for.

Are you maybe talking about the execution highlighting info? We do show values on wires while execution highlighting is turned on, and those do go away when the VI stops running.

[altenbach]

> Tip strips never show values to the best of my knowledge.

Just try it on any simple VI. (From some limiting testing it only shows if "retain wire values" is enabled. Not 100% sure). Hovering over a wire during run will show a yellow "tip-strip-like" number indicating the wire value (no need for execution highlighting).

For example, if I hover over the blue wire as shown, the autotool will turn into a probe cursor and the yellow "2" will show below. On array wires you get the number of elements (e.g. "#5")

[Mr._Bob]

My original request was, "Recommend that the “retain wire values” feature [tip strip] continue to retain the values EVEN AFTER the program pauses, errors-out, stops, or is aborted by the programmer."

This would mean that "retain wire values" is ON, "highlight execution" is OFF, the program runs all the way to the end with an error condition, the program stops, and the tip strip values are available for any line that executed up to that point. I realize that I can right-click and select "probe" [numerous times] for any value I want to see, but it would be much more efficient without the extra steps, and the probe screen commanding real estate.

As you have shown in the diagram above "hover mouse here", the editor mode would continue to allow the same feature.

Bob

[AristosQueue (NI)]

> From some limiting testing it only shows if "retain wire values" is enabled.

Weird. Doesn't match what we've said in several other feature reviews about avoiding diagram noise. Guess someone thought it was a good idea, probably because the values are small.

Speaking as R&D, it is viable, but strikes me as something we have generally said no to in the past, but apparently at some point someone decided to make happen in this limited case. It certainly could be expanded.

Speaking as a LV user with my own preferences: I definitely wouldn't want this stuff infecting my diagram readability. Tip strips are awful, IMHO. And this is the sort of thing that if you make it a Tools>>Option, no one will ever discover it to turn it on [and, yeah, I would default to it off!].

Bob, just FYI, I try to be careful to disambiguate my personal opinions from LV R&D opinions... just because I don't like a feature myself doesn't mean it won't get traction. I am just the most likely to feedback into the Idea Exchange, but others do monitor. 🙂

[Mr._Bob]

Aprreciate the advice. It's obvious that I am new to the forum. These are quite mundane examples I am soliciting [to start with.] I was basically "testing the waters" of the forum. I could do this all day, but I'm in a "time-permitting" environment.

I do have several other technical issues relating to RT controllers and FPGA programming that should be investigated. As I remember them, I will respond to the forum. I should be doing some more FPGA programming in a few days.

Thanks,

Bob

[Darren]

Any idea that has received less than 2 kudos within 2 years after posting will be automatically declined.