Setting different precisions for Y-scale and cursor position

[Bob_Schor]

This may be (slowly) changing, but from my listening to recommendations from NI, if you do not specifically require the "large memory" model of 64-bit LabVIEW, the "more likely to work and be problem-free" installation appears to be 32-bit LabVIEW installed on 64-bit Windows 7.  LabVIEW 2015 will also work on 64-bit Windows 10.

LabVIEW is designed as a data acquisition and control language (though that certainly doesn't mean that you can't do "pure computation", including data analysis, with it).  When viewed in this light, it makes sense (to me) to develop for either the MAC or PC, depending on the hardware requirements.  LabVIEW definitely started out on the MAC, but in recent years (say, the last decade) appears to be running more on PCs.

Bob Schor

[altenbach]

No, I mean when you open the VI in the Mac or linux development system they will open typically just fine.

In one of my programs, I call a FORTRAN dll via a call library function node (by specified file name). I have a conditional disable structure that gives one dll name if LabVIEW is 32bit and another dll for 64bit. (Since the dlls are windows only, this VI would not run on a Mac, of course).

Here are my two cases of the conditional disable structure. As you can see, the 64bit case is bleached because I an in LabVIEW 32bit at the moment. However, if I would open the same VI in 64bit LabVIEW, the 32bit case would be bleached. Similarly, you could make your VI aware to know about the current OS and other factors.

[blessedk]

Thanks Altenbach. Will read up those documentation. Bob you are absolutely right about stability of 64-bit. So far so good for me though, just a few minor "cosmetic" issues like the yscale digit 🙂 The 64-bit memory capacity is surprisingly huge and works great even with large computations / data analysis. Bottom line is ensuring code efficiency. I read up a lot of material on managing memory ( didn't understand much though but the little I did understand did the magic 🙂 ) Virtuallly all the bottle neck I faced earlier on are all sorted out

[jamiva]

---

@blessedk wrote:
Thanks Bob. My question is simple. On the vi I sent earlier. Can you set the Yscale digit to zero and set the cursor Y precision digits to four ? That's all I am asking. I have long resorted to placing a fresh graph and gotten what I wanted. But I just want to establish that we both agree that if the scale legend has been adjusted, ( or fiddled with in your own words) that it can't be restored to the default.

---

I too have issues with the way LabVIEW displays the format/precision of cursors in the Cursor Legend. There is a property node to set the Display Format for the x-scale and the y-scale, but I have yet to find one for the Cursors.

My workaround to get the cursor legend to display the format that I want, is to set the x- and y-scale format BEFORE I create the cursor. The x-scale and y-scale display settings (format/precision) are passed to the newly created cursor. I can then change the x- and y-scale to something else and it has no effect on the cursor legend. If someone out there knows how to change cursor display format after the fact please let me know.

p.s., With my workaround you don't need to place a fresh graph. Just create a new cursor.

[jamiva]

I recently upgraded to LabVIEW 2015 and my workaround the I just posted does not work. It seems that NOW whatever your y-scale format is copied to the cursor format in the cursor legend.

Now I see what blessedk was referring to.

[Peter_B]

I concur !  After trying to set the precision of any plot's cursors (especially in multi plot graphs) there were only two ways I stumbled across being able to do so.  Firstly ensure that all plots axes (probably at least plot 0) have the precsion you are after, then

1. Either Add a new cursor which inherits the precision settings, or

2. rename the cursor which also inherits the precision settings,

From that point no matter what you change you plot precision to be, your cursor precision will remain fixed unless you follow either of those points again.

Very unobvious and worth documenting in the Help file if that is by design (or even by accident!)

[DangerDave]

Hey everyone,

It sounds like this post has been working through the details of a potential bug.  Am I correct in summarizing that the numeric precision seems to be shared between the x and y axis, but only in LabVIEW 2015 SP1 64-bit?

If we can distill this down into a simple description of the problem and reproducing case, I can file a CAR myself or you can feel free to post a bug report here:

Bug Report - National Instruments

http://digital.ni.com/betaprogram/labview_BetaCust.nsf/webcar?openform&ParentUNID=6B5924AF8BF4A41886...

David Randolph

Applications Engineering

National Instruments

www.ni.com/support

[Peter_B]

---

---

 Am I correct in summarizing that the numeric precision seems to be shared between the x and y axis, but only in LabVIEW 2015 SP1 64-bit?

Hi David,

that isn't my experience.  I am using  LabVIEW 2015 32-bit (with no service pack).  Additionally I don't see a sharing of the precision between the X and Y axes, instead there is a dependency of the cursor's numerical precision on its associated axis in the graph.  i.e.

• The cursor's X-value can only be set/changed under certain circumstances and is solely dependent on the precision of the graph's X-axis at the time of those circumstances.
• The cursor's Y-value can only be set/changed under certain circumstances and is solely dependent on the precision of the graph's Y-axis at the time of those circumstances.

Those circumstances are described above in my last post.

[DangerDave]

Hi Peter,

Thanks for the reply!  I read back through your previous statement (hopefully second time's the charm):

"From that point no matter what you change you plot precision to be, your cursor precision will remain fixed unless you follow either of those points again"

I created a simple VI trying to replicate the cursor position not matching the data.

In my provided code, changing the precision input of the Numer to Fractional String VI (and running the code again) updates the precision of the Y values within the cursor legend.  I'm currently using LabVIEW 2015 on my machine.  

Am I still misunderstanding the behavior in question?

As an aside, I added the Numer to Fractional String VI and the Fract/Exp String to Number VI based on this KB:

How Do I Set the Significant Digits on a Double-Precision Wire?

http://digital.ni.com/public.nsf/allkb/7E14F54169EEECFD862571EE004B7F7F

Best,

David Randolph

Applications Engineering

National Instruments

www.ni.com/support

[Simon_B]

You can go to "Display Format" in the graph Properties and check "Hide trailing zeros" found under the number of digits you want. It should hide the zeros on the axis (which seems to be what you want) and the cursor legend.  However, it may be an added feature in LabVIEW 2016.