Article - Did National Instruments forget about Virtual Instruments?

[Jim_Kring]

I believe that we are currently laking a lot of features that would enable use to easily create Virtual Instruments and I've written an article about it here:

• Did National Instruments forget about Virtual Instruments?

What are your thoughts? What does software-defined instrumentation mean to you? What gaps exist in LabVIEW that need to be filled?

[cross-post]

[Graziano]

Hi!
   I have to state first that today, I'm not a virtual instrument builder.  I began using LabWindows for data acquisition, some years ago, and now, having changed workplace, I use LabView to build distributed acquisition systems with compactFieldPoint, and occasionally, I build simulator for plants we are working with.  So, actually, I don't create Virtual instrument. 

    That's why, from my point of view, I love "design, develop, deploy" catchphrase, and sometimes I benefit also from the fast prototyping philosophy.

   Maybe, you're a Virtual instrument builder, and I can understand that you'd like to have NI focused on that, but since NI is a company, and has to deal with revenue and so on, it's not bad to turn to different target, like industrial automation, or embedded devices.  I find that these new targets have a lot in common with virtual instrumentation, and can let LabView grow, and become more used.

   Of course, that's my point of view, any feedback is wellcome!

graziano

PS.: I hope this is not off-topic...

[Spectre_Dave]

Jim

I have to agree - it is possible to create systems with multiple instances of complex code but it is not trivial.  I worked on a project where we had up to 12 instances of a complex test programs running asynchronously - they where dynamically called by a Test Manager.  The hardest part was figuring out which VIs needed to be reentrant and which ones could not be reentrant.  In the end it works quite well but there was a lot of trial and error to get it running optimally whith 12 asynchronous instances.

[Ben]

I seldom find myslef needing or developing vitual instruments. I beleive this is due to at least two reasons.

1) I can not think of any two projects that I have developed that used the same hardware.

2) My customers are not interested in the "black boxes" used to acquire or control. The HP DAS they are using today may be replaced with a NI-DAQ system when the clone the app tomorow.

So any time spent trying to develop "Virtual Instruments" that go beyond the minimum required to acquire and control is of little or no use to my customers. The Virtual Instrument  may still be useful for those who have a collection of hardware that they will re-use for more than one experiment. But nobody is paying me to code up those apps.

One more thought;

Since the time when NI was using the phrase "The Software is the...." another transition has occured. The "instrument" is no longer a piece of gear that sits next to the PC but to a large extent has become an option that is IN the PC. An oscilloscope no longer cost big bucks and actually cost less than the LV license used to control it!

Ben

[vt92]

I think you make some good points.  The term 'VI' has probably always been more about marketing than reality.