Stagg54

You could try running the linux version in a VM.

You shouldn't need to get the key names from the file. It appears you already have them.

I've used Postrgres with ODBC and I've had no issues (other than the 32/64 bit problems that it seems you have already figured out). As far as installation and setting it up, I just download the turnkey linux appliance. It already has everything set up.

I think you are going to find you want a seperate emulated class for each instrument. It really depends on what are doing with the data from each instrument when you get it back. If you are just logging it, then maybe everything is fine with just one emulated class, but if you are making any decisions based on it, then it will probably require a seperate emulator for each instrument. And as for manually entering the response, that works as long as the calling entity isn't going to time out. So you either have to add intelligent code to send back the expected response, or you have disable the timeout in the calling code.

We got a new head IT security guy. Shortly after starting, he somehow got a virus on his laptop from a thumb drive he picked up somewhere. The solution was to ban thumbdrives. So he sent out an email to the entire company that said basically: "Due to security risks, we are banning all USB thumbdrives except for these special ones that are encrypted and have fingerprint readers built into them. They cost $200 (for a 4 GB drive). Oh and they are backordered and won't be available for 6 months. If you want one, let us know and we will put you on the waitlist." It was almost universally ignored...

I wrote software that was able to run in nuclear power plants, but was apparently not trustworthy enough to have administrator access on my own laptop. They had no problem giving me administrator access on the machines that the end software got installed on. So I could run the installer on the actual machine, but couldn't test the installer on my own laptop. Seemed kind of schizophrenic to me. One hand they trusted me, and yet on the other hand they didn't. I also worked in an R&D lab where we were constantly evaluating new hardware, which of course involved installing drivers. I would have to put in a ticket and have the local IT guy install it for me. After several hours he would walk over. He had no clue what I was installing and didn't really seem to care much either. I would just start the installer and when it asked for admin credentials, he would come over eventually and put his in. Then he would walk away.

Is there a way to use vi server to communicate across the app instance boundaries to achieve what you want? Seems like that might be doable.

Looks like you are doing sequencing. Instead of reinventing the wheel, perhaps consider using Teststand...

They could also be worried about this: https://www.usbkill.com/

Dynamic Dispatch is your friend. It sounds like you are on the right track, but you just haven't had that epiphany moment yet. It will come. Try and find some example code to look at. I'm sure there's plenty out there (just don't know where off the top of my head).

Do you happen to have a link to that article?

Title says it all. I'm a CLA (been one for over 4 years now) and I'm thinking of moving to Denver and looking for a job out there.

It might be kind of clunky but in this particular case you could convert the I32 to a double and append to either the front or back of the array (so you would just be sending a fixed size array of doubles) and then remove it on the other end.

It you only want to change what is displayed on the graph do the following:1. Right click on the x axis -> Deselect autoscale X2. Highlight the leftmost value on the x axis (in this case 0) and you should be able to change it to whatever you want. You can do the same thing for the leftmost value.If you want to actually remove the samples from the signal and then do some calculations, then it is more complicated.

Yes and No. Consider the case where I lock my bike up and lose the key. It's not illegal for me to break the lock and take my bike back.

Maybe it would be worth writing an xcontrol with a string indicator and have the xcontrol monitor the parent vi (not sure if that is the right terminology) and whenever the user hovers over a control, the xcontrol would display the help description of the control (that is being hovered over). It seems like you ought to be able to do that. Although that seems like a lot of work. Also the other way that context help breaks down is with subpanels. When you hover over a control inside the subpanel, LV only displays the context help for the subpanel itself, not for the control within the subpanel vi that you are hovering over. Hope that description makes sense. What I've done to compensate is just write a simple vi that runs in parallel and knows what vi is loaded and has a reference to the subpanel control and then whenever you hover over a control in that vi (detected using dynamic events), it updates the description (and possibly caption) of the subpanel control (using VI server), effectively "fooling" LV. LV is still displaying the description (and caption) for the subpanel control, the description is just changing on the fly to match which control the user is hovering over. Hope that all makes sense. That seems to work for us.

Typically we just use the context help. If you enter a description for a control/indicator, that will show up in the context help window. You can programmatically show/hide that. So you just have a help button that pops it up. Then when they hover over a control/indicator the description shows up in the context help window.

When i first read your post I immediately had the same idea. Yair beat me to it.

I agree with Jack. I think it is a feature that should not have been implemented. From the point of human performance and error prevention it is horrible.

Why not just have a seperate TCP/IP process (call it a moderator) running as a seperate app? Have all your logic in there. When each of your seperate apps startsup they checking to see if the moderator is running (by attempting to open a connection to it). If not, then it can spin one off.

Second this idea. You could definitely make an xcontrol do what you want. Xcontrols are usually a pain, but if you've got that many of them... The first thing I would ask myself is how important is this "feature"? My inclination would be for the end user to just deal with it and ok maybe it's not that intuitive, but it's a lot easier than writing an xcontrol. Then again if they want it bad enough and are willing to pay for it...

There are nodes that let you drop Matlab code directly into LabVIEW (with a limited function set). However I believe they require more than the trial version of LV (not sure what version you need - check the NI website). Note this also requires a Matlab license which I am assuming you already have.

Sounds like something for the idea exchange.

to add to Rolf's post: for an abortable wait loop, I just use a notifier as the loop timer. I set the timeout so that it will time out at the correct point in time. Then I can abort it early by sending a notification. I'm sure other people have other ways of doing this, but I've found it works for me and seems fairly clean to me. YMMV

Perhaps I should have explained more, but I figured if someone had to ask what the heap peak window was, then they wouldn't know the answer to my question.