04-11-2017 11:38 AM
We tried both 15.1 and 16.0 and both fail. 15.0 works.
04-12-2017 09:56 AM
Matt,
I am going to discuss with our R&D team about this behavior and whether or not it is expected. They might not have updated the documentation or we might be implementing it incorrectly -- I will let you know over the next few days.
In order to try and avoid this issue in the future though, I would suggest changing the code such that you won't run into this problem again. If you have a UI, try making the input a control as opposed to a constant. If you're unable to do that, the safest way of doing it is to rename it is to just put the device name rather than "DAQ::DAQ DEVICE NAME[::INSTR]" because at least we'll be able to safely change it in MAX if there is a discrepancy.
04-12-2017 10:37 AM - edited 04-12-2017 10:39 AM
Code changes are costly. They can lead to bugs. They require building, testing, deployment, etc. When something works, we don't change it. Our customers don't want us to change things that work.
When we upgrade OS or driver versions, we read the documentation, read me, and upgrade instructions.
I understand this is old nomenclature, but there is nothing in your documentation or upgrade instructions that indicates it won't continue to work. In fact, your current documentation states explicitly it will work.
Suggesting we should modify our code to defend against bugs that NI might introduce seems misplaced at the least. Moreover, I don't see how we can anticipate the bugs that NI will implement.
04-19-2017 07:45 AM
Any new information?
04-19-2017 02:59 PM
R&D suspects this is a bug, especially since it's worked in previous versions. They're going to investigate and get back to me. Thank you for your patience.
04-25-2017 08:29 AM
Another week... Any word? Should I submit a support request for faster response?
04-25-2017 10:12 AM
Seeing as though it is a software bug, it will take a bit to verify if there is a workaround and how they might approach the issue in the future. Submitting a service request won't get you a faster response in this case since we are already in collaboration with our R&D team.
04-25-2017 11:53 AM
Thanks!
Have you confirmed that it is a bug? Your last post said R&D "suspects" it is a bug, and that they are going to "investigate and get back" to you.
This post says it "is" a bug, so does that mean it is confirmed?
I am trying to confirm there is a bug. If it is confirmed, then I can explain to my customer why there is a cost to develop a workaround for Windows 10.
04-25-2017 12:16 PM
Seeing as though it worked before and it doesn't work now and the documentation is the same, it's safe to say that it is a software bug. Go ahead and tell your customers that there will be a cost to development. I would suggest, if possible, to make it a control if there is a GUI associated with the code. Otherwise, keep as the generic constant and then change the name in NI MAX whenever possible so you don't have to edit the code again. Thank you so much for your patience and understanding.
04-25-2017 01:19 PM
Just wanted to let you know that I filed a CAR - 644072 for you!