FSUIPC5 - LINDA Interface Changes

[pcssundahl]

Yes they do use FSS Server to interface with fsuipc. It was just an offer. I'm not sure how it could help you (your skills are far more advanced than mine). A lot of people including myself appreciate what you do and I'm just trying to help. I was thinking if you wanted to recreate the write issue for yourself. I think the issue scotflieger is having is very close to the issue I'm experiencing. Also when the errors occur it effects the eng gen write and apu and apu gen write. 

[Pete Dowson]

Just now, pcssundahl said:

It was just an offer.

Thanks.

1 minute ago, pcssundahl said:

I think the issue scotflieger is having is very close to the issue I'm experiencing. Also when the errors occur it effects the eng gen write and apu and apu gen write. 

I have a complete list of the ones it could happen with, I just don't know why at present.

So, I could use one of your apps to generate the error so I can debug it, right? Which app, and what would i need to do to provoke it with 100% certainly, please. If you could instruct me on this it very well might help speed things up. So thanks in advance.

Pete

 

 

[Pete Dowson]

10 minutes ago, pcssundahl said:

It was just an offer.

So, I need to pay to download then into my phone anyway, from the App store?

Pete

 

[pcssundahl]

No I can give you a promo code so you don't have to pay. Give me a few minutes I'm a bit busy but I will get it to you asap. Do you have an iPad? I will also need to privately send you the promo code so someone else doesn't use it before you enter it. I think FSS Overhead+ would be the best to recreate but I believe FSS MCP/EFIS will show the issue. 

[Pete Dowson]

11 minutes ago, pcssundahl said:

I can give you a promo code so you don't have to pay.

Thanks! Does the App Store ask you if you have one? I've never noticed.

11 minutes ago, pcssundahl said:

Do you have an iPad?

Yes, three of various vintages plus a Mini I use on holidays (steam trains). I use two iPads in the cockpit, one mounted each side, for charts and so on. That is, if I ever get a chance to fly!

11 minutes ago, pcssundahl said:

I will also need to privately send you the promo code so someone else doesn't use it before you enter it.

My email is petedowson@btconnect.com.

11 minutes ago, pcssundahl said:

I think FSS Overhead+ would be the best to recreate but I believe FSS MCP/EFIS will show the issue. 

Whichever is easiest. I'm a bit overloaded at present and the quickest and easiest method possible would be good. Do I need instructions on how to force it? I assume the Server is on the website somewhere?

Pete

 

[Scotfleiger]

I have been tidying my code to ensure writes only occur when a button is pushed. The Offset Writes are occurring only as I operate a button/knob and shows in the log as an offset command and interpreted as a control event. This works fine until, as you picked up, the log starts showing SimWrites and the Exception 20 Data Errors. It is happening independent of the timer events. I can not explain what may be going wrong. I am using the correct offset writes (address and length) and I am unable to trigger the fault with P3Dv3.4 and FSUIPC4.

fsuipc5.log8

[Pete Dowson]

3 minutes ago, Scotfleiger said:

I can not explain what may be going wrong.

It is not anything wrong you are doing. There are other confirmations, and I can see what is happening from the logs. I just don't understand why, and I need to reproduce it here so I can track it down.

I think you posted without reading the preceding messages. I'll reproducce part where I was appealing to you for help in reproducing it, here:

 

Quote

 

 4 hours ago, Scotfleiger said:

I see a large number of Exception 20s before LINDA was up and running.

 

Yes, this is because the TestOptions parameter added does a write test on all of the variables during FSUIPC initialisatin. Don't worry about those now, the info was for me.

   179406 *** EVENT: Cntrl= 65814 (0x00010116), Param= 0 (0x00000000) AP_APR_HOLD_OFF
   179406 *** EVENT: Cntrl= 65813 (0x00010115), Param= 0 (0x00000000) AP_LOC_HOLD_OFF
   179406 *** EVENT: Cntrl= 65811 (0x00010113), Param= 0 (0x00000000) AP_NAV1_HOLD_ON
   179406 SimWrite[145]: 0804="AUTOPILOT BACKCOURSE HOLD", DDef=00000007, Ref=3880, Size=4 [flt=0.000000, int=0, 0x00000000]
   179406 Exception 20 "DATA_ERROR", Ref 3880, Index param 1 on write SetData for "AUTOPILOT BACKCOURSE HOLD"
   179406 SimWrite[132]: 07D0="AUTOPILOT ALTITUDE LOCK", DDef=00000007, Ref=3881, Size=4 [flt=0.000000, int=0, 0x00000000]
   179406 Exception 20 "DATA_ERROR", Ref 3881, Index param 1 on write SetData for "AUTOPILOT ALTITUDE LOCK"

Something really strange is occurring. It is quite happily sending events for the assorted AP offsets you are writing, and then, suddenly, switches to the SimVar writing -- to non-writeable SimVars (as determined, in fact, by the initialisation test.

I now need to find out why this is occurring. Corruption of the tables in memory?  Nothing actually writes to those decode tables in my code, at least nothing programmed.

I think I'll have to write a Lua plug-in to do something like what you are doing. Writing of, what, 8 or so AP switches up to 20 times per second? I don't suppose you have a simpe (not needing LINDA) plug-in which can do that, and which gives the same symptoms on your System, have you? If not, it might speed things up if you could make one, test that it fails, then let me use it do I can debug the issue here.

Thanks,
Pete

[Scotfleiger]

Believe it or not I have never written a LUA plug-in. I work mainly with the Delphi Pascal code, Günter does the LUA.

[pcssundahl]

To recreate what I was having problems with. 

After installing FSS Server and letting all 3 servers past your firewall and installing the apps from App Store (instruction for promo code will be in email  

I deleted all files in module folder. Run fsuipc5 installer then register. All should be fine at that point. 

Start sim and confirm that autopilot and engine gen switches are working. Then I set my axis assignments and calibrate and filter options. Within 5 minutes after that I start getting the write issues. You will notice when it happens, you will engage some part of the autopilot on the app and it will change back. Or you will flip the engine gen switches on the app and they will flip back to where they were. At these moments the reads seems to still work correctly because if you flip those switches in the sim they will change on the app. 

I will send a promo for all my apps just as appreciation for what you do. Who knows you may like using one for enjoyment.

Email coming shortly. Thanks again Pete. 

[Pete Dowson]

2 minutes ago, pcssundahl said:

Within 5 minutes after that I start getting the write issues.

So is the App doing writes repeatedly, without having to operate switches on its display?

3 minutes ago, pcssundahl said:

I will send a promo for all my apps just as appreciation for what you do. Who knows you may like using one for enjoyment.

Okay. Thank you very much!

Pete

 

[pcssundahl]

No. Only when you operate things from the app. Reads are in a run loop 

[Pete Dowson]

1 hour ago, pcssundahl said:

Only when you operate things from the app.

So do i have to keep switching things on and off for 5 minutes, or are you saying these write problems occur on their own, with no attempts being made to write?

If so, then this is different to what is reported for LINDA.

Pete

 

[pcssundahl]

No I didn't. But I was using them. A basic quick startup and take off from active rwy. 

Sorry for the delay. My day job gets in the way sometimes. Lol

[pcssundahl]

I will do more test in a bit. See if I can help more. 

[Pete Dowson]

7 minutes ago, pcssundahl said:

I will do more test in a bit. See if I can help more. 

Thanks. Bed time here -- I'll catch up in the morning. I sent an email saying I couldn't find any "Redeem" button, BTW. So if I need to go that way i'll have to pay.

Pete

 

[pcssundahl]

[Pete Dowson]

Ah. So you don't select the App you want first? Just enter the code and it automatically gets the right App?

That's all new to me. Thanks. I'll try it.

Pete

 

[pcssundahl]

Correct. Automatic download.

 

So I did a bit more testing. since the last log file I sent all I did was start pc today and start p3d v4 in flight at 10000. What I notice this time is only a few autopilot writes don't work but also the spoiler armed error is coming from my Saitek throttle quadrant y axis not the app which is assigned in the axis section of fsuipc. The autopilot write error are writes from FSS MCP/EFIS. Heres the log file

FSUIPC5.log

[pcssundahl]

So the next flight I loaded the default V4 flight changed aircraft to the kingair and quick take off. The first climb out went good no problems (key part I think is there was no axis calibration done on the kingair) the last log was from a tds 737 after doing a profile specific axis calibration for my 737-MAX preferences.

 

Then I exited p3d v4 and 10 minutes later did the exact same thing. Started sim with default flight and switched to the KingAir.

Then did a profile specific axis calibration for the kingAir. Then did a quick take off. Problems started at about 2000 ft when I started using the auto pilot from the FSS MCP/EFIS app.

Prior to that I had only turned on a couple lights and raised the landing gear with FSS Overhead+. then I looked at the log. Im guessing the spoiler armed error did not show because the default king air does not have them.

BTW fss mcp/efis and fss switches is for iPhone or ipod ios 8 and later

all the others are for ipad ios 6 and later

FSUIPC5 log Kingair no axis calibration.log

FSUIPC5 log KingAir after axis calibration.log

[Pete Dowson]

7 hours ago, pcssundahl said:

(key part I think is there was no axis calibration done on the kingair)

I think that the only reason axis calibration brings the problem on is that there will then be many many more requests being made to SimConnect. I suspect it's some sort of SimConnect overloading, but how it then returns errors relating to write requests for SimVars for which FSUIPC doesn't issue them I don't know.

Anyway, I wil try to reproduce it here today.

Pete

 

[Pete Dowson]

17 hours ago, pcssundahl said:

To recreate what I was having problems with. 

After installing FSS Server and letting all 3 servers past your firewall and installing the apps from App Store (instruction for promo code will be in email  

I deleted all files in module folder. Run fsuipc5 installer then register. All should be fine at that point. 

Start sim and confirm that autopilot and engine gen switches are working. Then I set my axis assignments and calibrate and filter options. Within 5 minutes after that I start getting the write issues.

I've tried all this here, and then made loads of changes through the FSS MCP/EFIS and the FSS 737 OVHD, but I cannot provoke any problems. Everything carries on working fine.

In your calibrations, if the "filter" option essential to generate the problem? Are you assigning axes "direct to FSUIPC" or via the normal FS Axis controls, "Axis xxxxx Set"? Maybe you posted your INI here somewhere -- I'll take a look.

Meanwhile, just in case it is magically "cured" by no specific action on my part, please try this version, which is exactly the same as the so-far unreleased build I'm using:

FSUIPC5101e_TEST.zip

Please also enable ipcWrite logging, and Event logging (both events and axis events), so I can see if I can spot a relationship between inputs and the start of the problem. No need to provide the whole log, as it will be very large I expect, just the 50 or so lines before and after the first error.

Oh, and please show me the INI file too so I can compare settings.

Thanks

Pete

 

[Pete Dowson]

18 hours ago, Scotfleiger said:

Believe it or not I have never written a LUA plug-in. I work mainly with the Delphi Pascal code, Günter does the LUA.

Ah, shame. Maybe I need to install and run LINDA, then. Could you email me whatever I need, with instructions on how to repro the problem in the very simplest way? petedowson@btconnect.com.

As you'll see from the previous message in this thread, I have tried tp repro it using pcssundahl's methods, without success so far.

This problem is the only thing holding back the full release of FSUIPC 5.102, with joystick scanning fixed (including the crash with no real joysticks), PFCcom and PFChid support, and Active Sky WxRadar facilities working correctly.

Until I can reproduce your problems here I can see no way to fix them, or otherwise inform L-M of a Simconnect problem, if that is what it is. Currently I can see no place where I can realistically add enough logging to allow me to diagnose it on your systems, though I'm still looking at that possibility.

Pete

 

[Scotfleiger]

Hi Pete, running LINDA 3.0.0 would only be useful if you have the Saitek Multi-Panel. It is the functions/defined switches/buttons on this panel and associated offsets that are causing the problem. I have just tried 5101e (you posted above) and have managed to break it, but not as bad as before on 0x0810 broken. Interestingly, it is only the offset or offsets that once broken, will keep reporting Exception 20 DATA ERROR when any offset is set there after. The linked was run without your debug lines:

FSUIPC5.log 

I will look at providing you with a LUA plug-in to drive the offset writes.

[Pete Dowson]

30 minutes ago, Scotfleiger said:

I will look at providing you with a LUA plug-in to drive the offset writes.

Okay, thank you.

BTW are all these:

   297562 Exception 20 "DATA_ERROR", Ref 3157, Index param 1 on write SetData for "AUTOPILOT THROTTLE ARM"
   297562 Exception 20 "DATA_ERROR", Ref 3158, Index param 1 on write SetData for "AUTOPILOT THROTTLE ARM"
   297562 Exception 20 "DATA_ERROR", Ref 3159, Index param 1 on write SetData for "AUTOPILOT THROTTLE ARM"
   297562 Exception 20 "DATA_ERROR", Ref 3160, Index param 1 on write SetData for "AUTOPILOT THROTTLE ARM"
   297609 Exception 20 "DATA_ERROR", Ref 3162, Index param 1 on write SetData for "AUTOPILOT THROTTLE ARM"
   297609 Exception 20 "DATA_ERROR", Ref 3163, Index param 1 on write SetData for "AUTOPILOT THROTTLE ARM"
   297625 Exception 20 "DATA_ERROR", Ref 3165, Index param 1 on write SetData for "AUTOPILOT THROTTLE ARM"

(etc etc) a result of ONE offset write, or is this being repeated in LINDA -- for instance, if it sees the state has not changed correctly?

Do you have axes assigned and calibrated? Direct to FSUIPC or using FS controls?

Pete

 

[Scotfleiger]

Hi Pete

Yes, the Throttle Arm offset x0810 triggered the fault and all the repeats occur when any other button is pressed and the associated offset write to.

I have my throttle, prop and mixture and rudder axes assigned but not calibrated in FSUIPC5.

I am about to test the plugin and with advise shortly.