LabWindows/CVI Idea Exchange

Community Browser
cancel
Showing results for 
Search instead for 
Did you mean: 
Post an idea

 

What is the point of this Idea Exchange, when there is such a low uptake rate from NI?

 

There have been many very good suggestions posted here, but only 6 are labelled as "In Beta", which I assume means will be in the CVI release this autumn. There are 106 "Under Consideration", but there is no indication what is required to move them to "In Beta". For instance, Wolfgang posted "Reduce and document the NI services running after Windows logon" on ‎04-01-2011. It went to "Under Consideration"  on 08-15-2011. Two Years from the original posting, we still have nothing!


How do I justify renewing our Volume Licence Agreement when the LabWindows development effort within NI is so small?

 
How many people actually are on the full time LabWindows team within NI?

 

It would be nice if eventData1 and eventData2 were used when EVENT_VAL_COERCED is triggered.

 

I have an application in which I want to know if too high or low of a number entered.

 

GetCtrlVal always returns the coerced to number that is in range.

 

if eventData1 is 1 the max value was hit, if -1 the min value was hit.

 

eventData2 is always 0.

 

You can check the value returned by GetCtrlVal against the min and max of the control using GetCtrlAttribute.

 

I am running into an issue where my ranges change based on other control ranges and sometimes the min and max of a control are set to the same

number.  I want to know if the number entered was too high or low.

I believe there is a Fuzzy Logic package for LabView. I suggest that a Fuzzy Logic package also be provided for LabWindows/CVI.

 

I am working on a motor control project and I do have my own Fuzzy Logic code that I developed in C a few years ago for a Z-World board and recently ported it to a MicroChip dsPIC processor, also in C. I can port my own code to LabWindows but I believe it would be better to use code developed by NI folks as the NI code would probably more stable, reliable, flexible etc.

 

I won't attempt to expound on the advantages of Fuzzy Logic (FL) over PID except to say I have found FL to be very useful and easy to get working as it doesn't require extensive and exhaustive (and exhausting) system testing in order to come up with a working control system. Just a set of rules and a FL compiler to produce a "control surface" that is used with system feedback values to compute output values to produce a working control system.

 

So how about providing Fuzzy Logic code for LabWindows/CVI?

Hello all,

when a numeric is an integer, there's the option of having the radix preceeding the value in the control. For float/double, I'd like this tiny radix box to be extended for arbitrary strings in order to show the physical unit (m/s, kV, etc) of that numeric.

 

Right now the workarounds are:

- use a text box and display the value and the unit inside. Problem: works only for indicators, no arrows for changeable controls.

- display the unit in the label of the control.

It would be convienent if I could use ATTR_TITLE_FONT attribute with GetPanelAttribute() on top-level panels to obtain the Windows font properties for the title bar (this would be the font that the user can customize in the "Display Properties" in Windows Control Panel).  Currently this attribute only works for retrieving the font for child panels.  Thanks.

The addition of structured exception handling (SEH) would be a welcomed addition to CVI. This would allow for much cleaner handling of complex exception situations.

 

 

Hello,

 

Does anyone know if there are any plans for CVI and LabvVew Real-Time to support the Realtek NIC's? Supporting the Realtek NIC's would greatly expand the choices of boards that would be compatible with Real-Time modules.

 

Regards,

 

Tom Doyle

 

Hi community

 

To have more flexibility on the UIR conception, it would be convenient that controls or panel have a structure and behavior similar to Java Object (Java swing tutorial).

 

It would be necessary that CVI offers a list of basic control as it is the way today. But it would be interesting that CVI offers some special controls which could be a composition of multiple basic controls (or simply a composition of panel with its control). (Refers to Java containers)

 

IE: I create a panel composed with a picture control, a text control and a button control. Then I would like to have a control which able to react like a list control but each item of this list would be my composed panel.

 

IE: I create a preformatted “top panel” with some control to search, apply … then into this panel I have an area able to contain another panel. With this, I’m in position to change the panel into the area and preserving the design of the “top panel”.

 

IE: I could insert into a combo box an object like a radio button or a button instead of an numerical ou string item as it the way today.

 

I don’t know if this suggestion is understandable?

I like the Combo Box controls which are available in CVI, and I think there is some room for some improvements.

 

1.  Make their appearance more similar to the available Ring controls.  Right now the Combo Box has a look unlike the other available controls which can make a UIR that mixes Combo Boxes with Rings and Lists rather jarring.

2.  A numeric Combo Box control.

 

Thanks.

Hi all

how about making CVI operatable in MAC OS.

 

 

Starting with CVI2012 NI provides Win7 style popup menus to select files or a directory. Unfortunately, with the new functions is is not possible to select files from more than one directory, and it is also not possibly to select multiple directories. So one has to choose either to use the old-fashioned, outdated look and feel from the MultiFileSelectPopup function, or to use the new design but limited functionality... This choice is not a good one

 

I vote for a combination of the two - the look of the new and the features of the old functions

Hello,

 

the FileSelectPopup(Ex) provides the option to choose 'Select' or 'Load'... buttons with the benefit that the users choice of files can be limited to existing files only.

 

Unfortunately, the new MultiFileSelectPopupEx behaves differently, with the new function the user can always select new file even if it doesn't make sense in a specific context, hence the programmer has to take extra care for this scenario. Also, the response is different for "newfile.c" "existing1.c" "existing2.c" and "existing1.c" "existing2.c" "newfile.c". From a users perpective, however, this doesn't make much sense.

 

I understand that most of these restrictions result from Windows and thus can not be easily changed.

 

But may be it is possible to use a similar idea of different buttons (OK vs. Select) for the multifile popup, too, allowing the programmer to limit the user to select exisiting files only. OK would keep the current behavior, while Select would restrict the choice of files to existing files only.

 

Thanks!

After doing Find operation,Discard find results on next compile.

We do find operation lots of time and we see its find results below.

In same result area, we also see error and other messages.

But once we do find operation its results are retained till we close project.

Its actually confusing for programmer,when he compile he feels that there is error.It should be discarded on next compilation.It has no use once user see find result.So it should be discarded.

Wenn ein Rechner nicht am Nezt ist, vielleicht das Kabel nicht richtig drin oder am Platz kein Netz mehr frei ist, dann wird CVI mit einer eingeeschränkten Testlizenz gestartet und würde dann eine EXE generieren, welche in der Laufzeit eingeschränkt ist. Oft ist es der Fall, dass jemand CVI ohne Netzt started um im Quelltext etwas zu kontrollieren. Zu einem späteren Zeitpunkt erstellt jemand eine EXE neu und bemerkt nicht, dass das gestartete CVI keine Lizenz hat, obwohl der Rechner Netz hat. Er müsste erst CVI neustarten um die Lizenz zu holen.
Den Fehler bemerkt er erst am Folgetag, wo das Programm mit einem Fehler abgebrochen wird.
Es wäre aus meiner Sicht sehr sinnvoll, dass eine EXE, welche durch CVI ohne Lizenz erstellt wird, auch einen Popup mit eingebaut bekommt, welcher hiervor warnt.
Auch eine Abfrage, welche der Progrmmierer in sein Programm einbauen kann wäre sehr Hilfreich.

Hello,

 

I just discovered a strange and unexpected behavior of EVENT_VAL_CHANGED which requires that I will have to double check all my previous software written in CVI...

For some reason I have assumed that if the EVENT_VAL_CHANGED is triggered on a numeric control the new value will be different from the old one. This may not be true - the value may be still the same!

 

Since for more experienced users this seems to be a well-known fact I doubt that NI will call it a bug, hence I am suggesting an improved behavior Smiley Wink: Please, for numeric controls, fire the EVENT_VAL_CHANGED event only if the numeric value has changed...

 

(Right now, if the numeric control displays 1.0000 and I enter '1 ENTER' the value remains the same (at least), but the EVENT_VAL_CHANGED event is triggered...)

 

Thanks

 

PS: Also the documentation seems to be misleading: It (in CVI2010) states:

 

The EVENT_VAL_CHANGED event is generated continuously while the user is performing an ongoing action which results in the value of a control changing.

Is there a configuration/source management system available for LabWindows/CVI? If not, one is desperately needed.

Hi,

 

it's an old question, see here, but so far I have not heard about a practical solution. The quest is to find all instruments on a local LAN that are switched ON (and connected to the LAN, of course). Unfortunately, viFindRsrc does not work. Pinging all possible addresses seems not a viable solution due to the required time, either. So it's possible to detect all serial instruments, all instruments connected via GPIB, but until now there is no standalone solution (i.e.without involving NI-MAX) to detect instruments on a local LAN... Smiley Sad

 

Thanks,

Wolfgang

 

 

Might it be possible to implement some utility so that the debug_printf function could print to a window (or even the debug-output window in LabWindows/CVI) from a program that is NOT running under the debugger? I'm workng on a program that does some near-real-time data acquisition, and running it with the debugger from within LabWindows/CVI slows it down too much, to the point where it can't respond to the DAQmx notifications and the program crashes. Since I have a multi-core machine, I can, however, run it as a separate process at the same time as LabWindows/CVI. 

One problem that I've had, while debugging a rather large program with a complex user interface and lots of panels, is that quite often, if (when) the program crashes (say, due to an operation on a NULL pointer) the pop-up that displays the error is hidden beneath the currently open program panel which, of course, cannot be moved or otherwise touched because the program is halted. Likewise, I often find myself moving the LabWindows panel around on the screen while in debugging mode,  trying to see the currently executing line which is hidden behind some GUI panel.

 

It would be most helpful if all error pop-ups (indeed the entire LabWindows/CVI interface) had the highest priority and display on top of anything else on the screen.

* It would be nice to have a tool that automatically generated and maintained .fp files for subroutine libraries. Keeping these up to date is a pain.

 

* It would be most helpful to have a subroutine call-tree analyzer tool built into LabWindows. I'm working with a rather elaborate user interface at the moment, and I am totally losing track of what routine calls what. It's especially difficult because I use the "CallCtrlCallback" function quite a bit, especially during initialization, and I'm very afraid of accidentally implementing a recursive loop without realizing it. such a tool would, of course, have to be able to follow through the callback functions of controls, and perhaps into NI libraries that use callbacks.