What is your alternative for Vision Run-Time License

[MengHuiHanTang]

Don't know when it happened, but just noticed today that the price of NI Vision Run-Time License is now $582, was $440 last year. 

Even your code only uses one IMAQ functions, you still need to pay $582 for your deployment. 

I called NI and was told that one Run-Time License can be used by a single user in up to 3 PCs, but you cannot use the 3 PCs simultaneously.  

If you need to deploy to multiple PCs, this is a significant cost. I want to ask you how you handled this situation? What's your alternatives to NI Vision?

[SamiF]

Actually, even the new price remains one of the cheapest of the machine vision libraries market... .

Sami

[1100fps]

A few years ago, I ran across a company that offered third party solution to the Vision Acquisition / Vision Development module.  It cost a few hundred dollars, and there was no runtime cost.

I never got it to work.  It was set up to convert images into an array and had a range of subVI that would do image processing on the array image.

For a lot of the work I do, this would have been fine.  Sadly, I could not get it to work, so it is gone and forgotten.  If I stumble across it, I'll post the information, either as a cautionary tale, or as an option.  The supplier may show up at the Automate Show in Detroit this year, or at the Vision show in Boston next year.

As SamiF pointed out, the NI runtime license for vision is one of the least expensive ones on the market.  The vision development environment is also inexpensive compared to some other development environments, but that is tempered by the fact that NI does not appear to be putting a lot of resources in making the library grow.

I have one customer that uses ImageJ (FIJI) scripts, which I can call from LabVIEW.   It's not production fast, but it works, but does not have a camera interface.  

It would be possible to use the camera manufacturer's API to control the camera, and grab images, and call the processing through imageJ.  I'm not that ambitious. Yet.

[MengHuiHanTang]

Thanks for your reply. Appreciate the feedback.

I agree with you, the imaging library does not grow for a number of years (as far as I know, might be completely wrong). What justifies the 30% increase of price? Inflation?

I have no complaint with the Vision module, i.e., image acquisition and processing. So I don't think I will switch to other programming environment for coding. However, the deployment cost is definitely a factor to consider, since it costs ~$600 just to deploy your application to a computer. 20 PCs equals 13K. 

This might sound like nothing to other corporates in terms of budgeting. I am simply asking whether anyone else put any consideration into this issue. 

[p4keal]

I was asking the same thing for LabVIEW Community Edition. Just to let children play with an USB cam.

Maybe you can rewrite the corresponding modules in Python using harvesters and integrate them into your LV application. But honestly, I can't guarantee success and I can't estimate the effort. Can you briefly describe your application?

https://github.com/genicam/harvesters

If you have cameras of a specific type, you may be able to use the manufacturer's SDK. Here is an example from the context of the Tango Control System for Allied Vision Vimba Camera written in C++.

https://www.tango-controls.org/developers/dsc/ds/55/

May be, you can do something simular and make LV wrapper around.

No matter how, each variant looks like a lot of work...

P.S.

I am lucky enough to work at an university and we can use Academic License Suite, so there is no cost for our lab PCs.But at the moment NI seems to be in a state of flux in terms of licensing policy. And I would also like to have an alternative in the event of a fall.

[1100fps]

Just to toss some gas on this fire..

I had a customer a few years back that needed a simple inspection.  The obvious answer was a smart camera.  The NI Smart camera was fully up to the task and easily programmable, with an estimated programming effort of about 30 minutes.  The hardware / mounting would take about a day and a half.  The customer could not cough up the $8K for camera, engineering, and fixture.  The guy was in a bit of a pickle, and had been good to me in the past, so (with the help of a friend that knew python), we cobbled together a Raspberry Pi, with a cheap MIPI camera, and using Open MV library worked up the code to do the job.  Total cost, including lights and hardware from Lowes, was around $150.  The customer came to see what we had, was impressed and wanted it installed .  This was all out of my pocket, and we had no travel budget.  Next thing I know, a PO for $6K showed up and the system was installed.  He could have taken it himself, it was self contained and good to go.  No matter, that's how we funded our first SLA printer.  As I understand it the system is still performing well, and is now a mandated inspection by the FAA.

The point being that there may be occasions where LabVIEW may be the easiest solution, Tools you know, and all that, but may not be the most cost-effective solution.  Especially if there is a need to replicate installations.