Since my communication is not critical and is asynchronous, I decided to go
with UDP communications and have it working just fine. I have one question
though: Will UDP or any of the other discussed application to application
communications work on a PC that does not have a network card installed?
This would not normally be the case but could be under some circumstances.
My guess is that File I/O and possibly VI Server would be the only ones that
would work without a network card installed.
"Greg McKaskle" wrote in message
news:Qd10c.13279$lS1.3844@fe2.texas.rr.com...
> > Does anyone have a better suggestion for communicating between built
LabVIEW
> > App.s?
> >
>
> This depends on the type of communication. Datasocket is quite easy and
> appropriate for asynchronous data updates. You could use TCP, doing
> your own flattening and unflattening. This will take less installation,
> but will be more complicated in other ways.
>
> Another choice is the VI Server. This is often the best choice when you
> want to have synchronization or handshaking of some sort. This also
> takes some installation, as the ini file needs to turn on the server and
> designate what types of access will be accepted, but that is about it.
>
> There was one comment about using queues. If you wrap them in VIs that
> call through the VI Server, these will work really well. This technique
> means cloning the VI using Save a Copy. Open the copy and delete the
> diagram, replacing it with a VI Server Call node. Then arrange for the
> VI Ref to be read from a subVI call or add them to the connector pane.
> Finally, you might want to make a small change to the icon so that you
> can tell that they are remote.
>
> After this, you can use these VIs locally and they will perform their
> operation on the remote computer. In fact, you don't even need to build
> the subVI wrappers, but it makes them much easier to use. If you wrap
> the queue, then your local diagram looks pretty much like a local usage
> of queues except for the initial connection to get the remote references.
>
> If this didn't make sense, ask questions and I'll go into more detail.
>
> Greg McKaskle
>