UKTAG Meeting 3

[.aCe.]

The third meeting was held on Friday the 28th of February 2020 from 1pm to 5pm

Location: Abaco Systems, Old Tiffield Rd, Towcester NN12 6PF and streamed on Microsoft Teams.

Videos: http://bit.ly/UK_TAG3_Playlist 

Agenda:

Time	Topic	Abstract	Presented by
1:00 – 1:10	Welcome and Introduction
1:15 – 1:55	TestStand – Batch and Parallel Testing Presentation | Video	One of the strengths of NI TestStand is its ability to perform batch and parallel testing with relative ease. This presentation delves into the Do’s and Don’ts, synchronisation and data sharing techniques of Parallel and Batch testing in TestStand.	Mathis Baumert (CLA, CTA) - Abaco Systems
2.00 – 2:40	TestStand Plugins Presentation | Video | Examples	In TestStand 2012, report generation ceased to be tightly coupled with the process model. All thanks to the introduction of process model plug-in architecture. This presentation walks through creation of plug-ins in TestStand.	Michal Bienkowski, (CLA, CTA) Averna Technologies
2:40 – 2:55	Coffee Break
3:00 – 3:20	LabVIEW UI Tips and Tricks Presentation | Video | Video	A demonstration and code walk-through of a user interface prototype that utilises several UI tips and tricks that give applications a professional look and feel.	Tom McQuillan (CLA, CTA, CLED, CPI) – Scientifica
3:20 – 3:40	TestStand OI Tips and Tricks Presentation | Video	This presentation looks in the use of messaging techniques between TestStand Sequence and the Operator Interface. It focuses on user messages, reporting progress, errors and status for single and multiple test sockets.	Gavin Clarke (CLA, CTA) - Abaco Systems
3:45 – 4:25	LabVIEW OOP and Actor framework 101 Presentation | Video | Video	Object oriented programming lends itself to modularity, code reuse and extensibility. This talk will go through the why and how to get started with OOP, and the Actor Framework that utilises its benefits for large applications.	Tom McQuillan (CLA, CTA, CLED, CPI) – Scientifica
4.25 - 5.00	Wrap Up and Networking

[Oli_Wachno]

Hi Sree,

are you going to offer remote access again?

Cheers

Oli

[.aCe.]

---

@Oli_Wachno wrote:

Hi Sree,

are you going to offer remote access again?

Cheers

Oli

---

Hello Oli,
Yes. Link to Microsoft Teams meeting will be posted soon or it will be sent to you if you register for the event on NI events page. Details to follow.

[danny_t]

Quick question, is registration required to join the remote Team session ?

cheers 

Danny

[.aCe.]

Hello Danny,

I will borrow the LabVIEW connector pane terminology and say it is not required but recommended.

[danny_t]

---

@.aCe. wrote:

Hello Danny,

I will borrow the LabVIEW connector pane terminology and say it is not required but recommended.

---

🤣 thank you I will do 

[bienieck]

Yesterday, after my presentation about the plug-ins, Oliver asked if there is any advantage of using plug-ins compared to simply overwriting callbacks.
I have responded that if you have some weird requirements for the report format, then you can create your plug-in. But this might not be a satisfying response. When I thought about it later, I was not able to convince myself with my example 🙂
It is because you may simply implement what I had in my UUT Done in your TestReport callback.

So is there any reason to use custom plug-ins for reporting? I think so, and here is why:

• The callback would, most probably, be in your client sequence - in the product-specific part. Plug-in is independent of the client file. So you have separated product-specific parts from the system.
• Plug-in is modular and more universal. You can maintain it separately from the initial project.
• In your plug-in you can have only those parts which are needed, therefore you have batter performance.

Besides that, you can see that plug-ins are one of the best practices in the TestStand environment. Tools like SystemLink or SkyWATS are shared in the form of plug-ins because that's the most natural way. It's just very easy to hook to the process model with your custom functionality thanks to plug-ins.

But at the end of the day, you would need to decide yourself about the solution, based on your particular requirements. If you decide to use plug-ins then I can help 😉

Am I more convincing now?

[RayFarmer]

Hi,

The help suggest if you are using 2012 or later then you will be using the plugins.

if you what to use pre TestStand 2012, (legacy Teststand 2010) maybe because your systems are not all at the same standard then use the Legacy Model Switcher. Then you can use the Callbacks.

I am not sure if this is any better of an answer.

[McQuillan]

Hi Team,

The videos from my presentations can be found here:

• UKTAG#3: Object Oriented Programming, The Actor Model, and The Actor Framework
• UKTAG#3: User Interface Design - Splitters, Panes and SubPanels

Cheers,

Tom

[WireWeaver]

---

@bienieck wrote:
Plug-in is modular and more universal. You can maintain it separately from the initial project.

---

Same as this comment, but worth expanding on, is that the burden of maintenance can be lower with plug-ins. If you need to upgrade TestStand versions for example, you only need to upgrade the types in your plug-in and test it. If your customizations are spread across a bunch of different client files, then any changes need to get propagated... more testing, more complexity, etc...

TestStand Semiconductor Module is another example where NI uses plug-ins to customize TestStand - in this case it's a lot more extensive than just results processing (interfacing with handlers, mapping to dut pins, etc..). Because of the usage of plug-ins, you don't even need a special process model to use TSM - it works with the default ones or any custom process model that uses the plug-in interface.

-Trent