Problem with AM4 and Logitech Panels (gadget lagging and crashing)

[Crunchmeister]

The current Working Title CJ4 uses mostly custom vars and events. They won't work with Logitech panels. That will be "fixed" in about 2 weeks when the Aircraft and Avionics Update I update is launched. Most of the systems will run with normal events and vars (where applicable). Although no mention was made, the CJ4 instruments will ONLY work with the CJ4 mod by Working Title. They won't work with the default CJ4, nor will they work with the new CJ4 that is replacing the WT mod in the aforementioned AAU I update.

Logitech panels lagging and cutting out randomly is typically due to an overloaded USB bus (power). I used to have some. I had to run them on their own powered USB hub as those things suck back ridiculous amount of power. Particularly the ones with the LED displays. If they shared a USB bus with anything else, they would consistently cut out on me. I used to run the USB connections for my touch screen on the same hub, and it was a no-no. Once I moved the Logitech stuff on their own discreet hub plugged into a dedicated USB port on the computer, they worked fine.

[GiorgioB]

Probably something in those CJ4 instruments then. Those aren't ours, so you would have to ask the developer.
My Logitech joystick has been together with Air Manager for over an hour, but nothing happened. Have you ever tried a different USB port?

Yes I have tested another USB. My joystick is not Logitech, it is a CH Eclipse Yoke. My Logitech gadgets are the panels.
I have removed EVERY single instrument besides the 6 original ones and I still get the joystick "spikes".

About the Logitech Panels, even though I don't get any error now on Airmanager Log, it still breaks connection with Logitech Multi Panel at some point.

The current Working Title CJ4 uses mostly custom vars and events. They won't work with Logitech panels. That will be "fixed" in about 2 weeks when the Aircraft and Avionics Update I update is launched. Most of the systems will run with normal events and vars (where applicable). Although no mention was made, the CJ4 instruments will ONLY work with the CJ4 mod by Working Title. They won't work with the default CJ4, nor will they work with the new CJ4 that is replacing the WT mod in the aforementioned AAU I update.

Logitech panels lagging and cutting out randomly is typically due to an overloaded USB bus (power). I used to have some. I had to run them on their own powered USB hub as those things suck back ridiculous amount of power. Particularly the ones with the LED displays. If they shared a USB bus with anything else, they would consistently cut out on me. I used to run the USB connections for my touch screen on the same hub, and it was a no-no. Once I moved the Logitech stuff on their own discreet hub plugged into a dedicated USB port on the computer, they worked fine.

Thanks for your input! I deleted all the CJ4 panels from air manager, cleared persistence and cache just to be sure, but I still have the same problems.
My saitek panels are running from a dedicated USB HUB and this problem only happens with AM4 on so I don't really think it is an energy matter. I have 1000+ hours using these panels and only when using AM4 they break down.

Any ideas I could test, maybe?

[GiorgioB]

I was really willing I would be able to add Air Manager to my simulator's setup but I can confirm Air Manager is causing yoke inputs and random multi panel disconnections.
Is there any way to use AM4 only as mouse input and totally disconnect it from any hardware?

[jph]

Hi GiorgioB
As you rightly say, it is NOT a usb power issue as you have proven.
With the ch products yoke - what software / drivers are you running ? - also I presume you must be using SPAD for your Saitek panels ?
Joe

[GiorgioB]

Hi GiorgioB
As you rightly say, it is NOT a usb power issue as you have proven.
With the ch products yoke - what software / drivers are you running ? - also I presume you must be using SPAD for your Saitek panels ?
Joe

Hello jph! I really appreciate your effort on helping me and I think you are totally right about your questions.
I am using Spad.Next and I think I managed to solve the issue yesterday. I tested it for about 2 hours with no errors whatsoever.

I think the issue is exactly related to the pedal and yoke drivers. Without the drivers (using windows native) there was a lot of 'jittering' and flickering on the X axis of the yoke and rudder axis of pedals. After installing the original CH and Saitek drivers (for pedals) I get much steadier signal.
I had never needed drivers for these 2 hardwares before but I think the flickering was the cause of an excessive information traffic on sim connect causing it to crash the multi panel and somehow generate button presses on yoke.

I CAN CONFIRM that after installing yoke's drivers I don't have any random button presses anymore (FIRST ISSUE RESOLVED), but I still had multi panel failing on me. So I have also installed the pedal's drivers and after that 2 hours flying with no bugs. I'll be able to confirm after work today when I fly again.
Does my theory make sense? I saw on spad.next that flickering joystick signal can indeed crash something or make it lag. I never had issues using only spad.next but maybe air manager + spad.next + 3 panels + 1 yoke + 1 pedal (the last 2 flickering) might be too much information for sim connect to deal with properly?

THANKS AGAIN!

[GiorgioB]

OMG, this bug is about to drive me crazy!

I managed to not have joystick flickering or random inputs anymore, so the less terrible bug is dealt with.
But I am still getting multi panel disconnection (first it lags, then it crashes) after a while flying when using Air Manager. :*-(

Switch panel and radio panel keep working normally, is just the multi panel that is crashing out of nothing. If I close AM4 it never happens (I have 1000+ hours using these panels).

Any ideas, please?

EDIT: I realized the joystick parameters were back to "default" after restart because I didn't save it.
I am testing it again, fingers crossed!

[Ralph]

Logitech thingies might be the issue we were thinking of, but it is hard to say. Every x amount of seconds, Air Manager asks Windows which USB devices are connected. That's how we detect certain hardware devices, like the Raspberry Pi Pico and the Knobster. It might be that these screens get confused by these requests. But as mentioned in the ticket, please send an email to Corjan, and you'll get a build without the USB stuff.

[GiorgioB]

Logitech thingies might be the issue we were thinking of, but it is hard to say. Every x amount of seconds, Air Manager asks Windows which USB devices are connected. That's how we detect certain hardware devices, like the Raspberry Pi Pico and the Knobster. It might be that these screens get confused by these requests. But as mentioned in the ticket, please send an email to Corjan, and you'll get a build without the USB stuff.

Thank you very much for your input Ralph! I have tested every possibility. I made an 8 hour long flight today just testing it. Indeed with less joystick noise the multi panel crash won't happen so often but it is still unusable as I know it will fail at some point, and it could be on a crucial moment of the flight.
I sent Corjan a PM. If it is best to send an e-mail would you mind sending me his address via PM?

THANK YOU! I built my cockpit having Air Manager in mind and I really hope I'll be able to fulfill this long time dream on its full potential.

[Ralph]

Yes, no PM please, send an email.

[jph]

Logitech thingies might be the issue we were thinking of, but it is hard to say. Every x amount of seconds, Air Manager asks Windows which USB devices are connected. That's how we detect certain hardware devices, like the Raspberry Pi Pico and the Knobster. It might be that these screens get confused by these requests. But as mentioned in the ticket, please send an email to Corjan, and you'll get a build without the USB stuff.

HI @Ralph @Corjan and @GiorgioB
That was exactly on the lines of what I was hoping this may lead. There was another issues a year or so back with un-commanded inputs which is possibly / probably also related to whatever the solution may be. It was also around the time of the 'game controller' addition to AM which will be on a similar path as they are all HID. It may be worth looping out these as well for the test ?
Thanks for going the extra mile.

I look forward to the results.
Joe