Before I updated LINDA to 2.6.6, I've always used the hat switch on my Thrustmaster Hotas Warthog joystick to control the elevator and aileron trim. Since I'm using the new version, this is no longer possible, because it keeps trimming endlessly. In the settings, it looks like this:

In the past, after I for example pressed up on the hat switch and let go, you could see in LINDA that LINDA selected H1: 0 again. Now this doesn't happen anymore and it seems that LINDA keeps sending trim commands. In previous versions of LINDA the text "DO NOT USE" behind H1: 0 wasn't there. Also, when I don't assign anything in the second column (on repeat) it just works, but then I have to keep pressing the hat switch while trimming, which isn't realistic. When I program the same in FSUIPC (latest version 4.948e) everything works fine.

 

Any thoughts on this?

What version of LINDA were you using previously? The H1:0 DO NOT USE was introduced around v2.5 because of issues getting the HAT to work correctly. In the configuration shown the repeat function is set but should cease when you remove your input. I always use the FSX/P3D HAT function for viewing left/right/up/down.

 

I will have to do some deeper investigation to bottom this problem out.

Below is a screenshot of the settings window of my old LINDA installation, so I guess it was a pretty old version! I didn't update it earlier because it did exactly what it needed to. Last week I needed to update because the profile for the Aerosoft Airbus didn't work with the older version.

 

I use the hat switch for trimming as it's meant for that particular function and I use TrackIR, so no need to use it for viewing around.

 

In the configuration shown the repeat function is set but should cease when you remove your input.

 

That's exactly what it doesn't do on my pc. The repeat function keeps continuing, even after I've used some other buttons. It only stops when restarting FSX or selecting another aircraft. I've already tried a clean LINDA installation. As I said, it works perfectly when configured directly in FSUIPC, but since I have 99% of buttons and switches configured in LINDA it would be nice if also these functions could be configured there.

There is an issue with On Repeat as it stores up a load of repetitions of a button push and then continues to apply the assigned function after the button is released. I need to find a way of killing these stored repeats when the button in released. I have seen a similar problem with trim wheels that have a up and down button operation.

You will be pleased to hear that I have managed to replicate the repeat fault with the HAT switch. This is the first stage of working out where the error is. This does not occur with normal non-HAT buttons. I will keep you updated.

That sounds great! I hope it can be fixed.

That sounds great! I hope it can be fixed.

 

 

After several days looking I have identified the cause and, due to the special way HATs are handled and passed to LINDA, it is not possible to use the On Repeat function with them.

 

The reason is that as soon as a HAT position is detected and passed to LINDA (say button 5 - down) the button is immediately released and a zero button is returned with a nil function. I have managed to stop the runaway button action but there is no easy way to permit a repeat function while the button is pushed.

 

At present, I advise that no On Repeat functions are used and that if you want to program a trim function then FSUIPC4 should be used. 

Thanks for investigating. The problem is not that big; as I've said before I already started using FSUIPC for this. However, how come it did work in the old version of LINDA?

Thanks for investigating. The problem is not that big; as I've said before I already started using FSUIPC for this. However, how come it did work in the old version of LINDA?

 

Which version?

Version 1.xx, I don't know it exactly. I think 1.11, but I don't know for sure.

Update: I have located the immediate release of HAT inputs to the core code (Linda.exe) and the special handling given to HATs. The code is complex and I am still investigating. It is a good exercise in reading my way back into the code.