P3D v3 crash but no Event Viewer entry

[Ray Proudfoot 8,980]

Yesterday Concorde CTD at the same point it did previously. This morning I flew the same route using a default aircraft - F35 Raptor. No CTD. So I flew it a second time. Again, no CTD.

This leads me to suspect the problem lies with Concorde but I got short shrift off Lefteris the other day for saying that. I'm flying the same route with Concorde now. I'll get to the suspect location within the hour.

[Ray Proudfoot 8,980]

Concorde CTD at the same location today as yesterday. Message posted on FS Labs Concorde forum. Can’t see anything but a bug in the latest release. But where is a mystery. Hopefully Lefteris will have the time to check things.

[Ray Proudfoot 8,980]

Nothing back from FSL yet but I’ve had a thought about one further test to try and nail the source of the problem. I’m 90% certain it’s related to Concorde so a test flight tomorrow will be without using the INS navigation system. I have a moving map on another computer showing waypoints and headings. It’s easy enough to stay in Heading Hold mode and manually change the heading on the AP.

It’s this notice that has made suspect where the problem might lie...

v1.41 is a small maintenance release that updates the Concorde-X for FSX and P3D v3 to save selected options properly, as well as fix the problem with the freezing INS instruments.

More tomorrow.

[Ray Proudfoot 8,980]

I have just posted this in the FS Labs Concorde forum following another test flight...

I've just flown another test flight LPPT-EGCC with one significant change. I did not load a flight plan into the INSs. I entered my current position and aligned them but for the flight I used TRK HDG adjusting it as required.

I passed the point where the CTD had occured in previous flights and that was encouraging. Descended down to 10,000ft and decelerating to 250kts I was 4nm from MIRSI  - a hold for EGCC. At that point the sim crashed. No entry in Event Viewer.

I remain convinced these problems lie in the INSs especially when the release notes state... "v1.41 is a small maintenance release that updates the Concorde-X for FSX and P3D v3 to save selected options properly, as well as fix the problem with the freezing INS instruments."

The fact that I have flown this exact route twice in a default aircraft without problem leaves me in no doubt that the cause of these CTDs lies in the Concorde code for v1.41. By a process of elimination it cannot be anything else.

I await a reply from FSL or indeed any other Concorde user. Sadly, the forum is extremely quiet so it's hard to say how many people are flying v1.41.

[Ray Proudfoot 8,980]

An update after more test flights today...

Another Concorde pilot flew the LPPT-EGCC route and reported no problems. He installed Concorde and CIVA on a clean system. So I decided to uninstall CIVA after Concorde and then the Concorde installer would prompt me to install CIVA. Did all that and loaded the plan.

All went well until I was 4nm from MIRSI on the approach to EGCC. P3D crashed at exactly the same position as last Thursday. EXACTLY THE SAME POSITION ON A DEFAULT P3D install.

I should say at this point I also flew the route earlier today with the INS turned off. Flew the entire route with no problem. So this definitely points to something in the INS code. Logically it cannot be anything else.

Ive posted in the FS Labs forum and asked if I should raise a ticket. Waiting to hear from Lefteris Kalamaris.

https://forums.flightsimlabs.com/index.php?/topic/24544-concorde-v141-causing-ctds/

[Ray Proudfoot 8,980]

I'm cautiously optimistic I may have fixed the problem. I've completed two flights - LPPT-EGCC and back. Both were fine.

Working on the principle that to resolve an error you should only change one thing and then test I thought about why a fellow Concorde pilot could complete a flight and I couldn't. Both with W10, both with default P3D and both with Concorde v1.41.

But looking at his OS build number it was different to mine. His was 18363.657 whereas mine was 18363.592. I had been holding back on Windows updates because everyone knows they can cause problems. But in this case if he was running with those updates and he had no problems then they were worth trying. So I installed the updates and my build number now matched his and I departed LPPT.

Very nervous approaching the problem waypoint but I passed it okay and the next potential location was approaching MIRSI. But that was okay too. Landed and after a break for lunch went for the return back to Lisbon.

That was fine too so I'm crossing my fingers it's fixed. Only more flights will prove that one way or another. This is the first two flights I've had that haven't resulted in CTDs in a few months. The attached shows the updates I applied today. Looks like one of them fixed it but which?

I'm very grateful for the help and suggestions from two fellow Concorde pilots over at FS Labs.

So Jim, if I don't come back to this topic in a few days you can assume it's fixed. Windows updates fixes a problem!! Wow! 😁

[pedrotrindade 335]

Here you go Ray

 

 

[Ray Proudfoot 8,980]

Thank you very much Pedro. December 1997 and a full length 03 departure but to where? No notes with the video. Congratulations to the cameraman with very good panning.

it only takes 1h 25m to get to Manchester. Add another 1h 15m for a Boeing / Airbus. 😎 So nice to be able to fly her again after three months of trying to find the problem.

[IanHarrison 522]

I think it is still the most beautiful passenger aeroplane ever designed.

[Ray Proudfoot 8,980]

Sadly and very frustratingly, I was premature in hoping these CTDs had been cured by Windows updates.

Another CTD this morning on a different route. Oslo to Heathrow. I feel I have no alternative now but to try an earlier version of Concorde.

[Ray Proudfoot 8,980]

I installed v1.39 and tried to assign key combos to opening certain sub-panels. It caused P3D to crash but this time there was an entry in Event Viewer. The CTD was caused by FSL_Events.dll. So I suspect only a certain combination of Concorde versions will work with Windows 10.

Its not realistic to expect FSL to keep things up to date with each W10 update so reluctantly I have decided it is not compatible with 10. Some users have locked their systems from updates to prevent problems but that’s not something I’m prepared to do.

Hopefully FSL will release a 64-bit version at some point for either P3D v5 or MFS. Back to subsonic flights now. 😪

[Ray Proudfoot 8,980]

I’m afraid I’m like a dog with a bone when I have a problem like this. Logic says leave it and fly P3Dv4. But when you’ve flown this aircraft for as long as I have it’s bloody annoying when you get told “you’re the only one with this problem. Reformat your SSD and start again.” 😠

If the problem is a corrupt system file as has been suggested how come I can run sfc /scan now and it returns zero errors?

If there is a corrupt file (which there isn’t) why does this CTD manifest itself well into a flight and never at the start? On a flight from Oslo to Heathrow it occurs at one of two places less than 20nm apart off the Essex coast. The only addon stuff is FTX Global and Aerosoft Heathrow Xtended.

If I restart P3D and load the last AutoSave flight it is fine for the rest of the flight including flying over London.

My analysis of the problem in layman’s terms is something is being written to a file as the flight progresses and eventually causes an overflow resulting in the CTD. When the flight is reloaded whatever was being stored is now empty which is why I can fly over the same point and not get the CTD. Unscientific but it makes sense to me.

I have submitted evidence to FS Labs of a system file check with no errors. I also ran Resource Monitor and selected P3D. All the resources it uses were listed with nothing untoward. I don’t know if I can use that tool for any other analysis. Process Monitor grabs everything but analysing it needs an expert.

The ball is in FS Labs court. Let’s see what they say.

[Ted Striker 338]

Ray, I had crashes with P3Dv3 that I could not find the source of. I was spending more time with Process Monitor, Process Explorer, etc. trying to solve the mystery than it would take to just reinstall everything so I threw out the bone and uninstalled everything including P3Dv3. I then reinstalled P3Dv3 and addons one by one with testing after each install. I have not found the source of the problem and I don't have all the addons back in but the problem is gone. I would reinstall the Concorde and test for you but I am on windows 7. I see no reason to switch to windows 10 until MSFS is released.

I noticed that you still have some other addons besides the Concorde. I suggest you clean house and then reinstall P3Dv3 and just the Concorde, no other addons including FSL spotlights. If that works, install your other addons one at a time and test. The good news is that you say others are successfully using the Concorde with windows 10.

Ted

[Ray Proudfoot 8,980]

Hi Ted,

I've already uninstalled P3D v3 and performed a clean install. Likewise with Concorde. I flew a route with a virgin system bar SODE and Couatl for GSX. I was hopeful it was a SODE/GSX problem but I still had the CTDs.

I'm now wondering if when I uninstalled P3D and reinstalled it again I should have run the SimConnect installers again even though they were never uninstalled. AS16 and other moving map software connects okay so it's doubtful it will fix it but I have few options left.

Others say they have no CTDs but you don't know what if any software they're using as well as P3Dv3. In all my years of simming I've never had a problem like this. The simplest solution would be to revert to the previous build of Concorde but no-one has it and FS Labs only support the lastest version. I'm between a rock and a hard place. 😕

The only successful flights I have had are with the INSs turned off using TRK HDG to navigate. That points hte finger at the Concorde code but I'm bashing my head against a wall there trying to convince FSL. I'll reinstall simconnect again tomorrow.

[Ray Proudfoot 8,980]

I’m happy to report the cause has been found and fixed. Last Friday Andrew Wilson at FSL recommended I uninstall GSX and SODE. Instead of uninstalling them I foolishly just disabled the two processes from running in the EXE.XML thinking it was the same.

But the CTDs persisted. Lefteris reminded me this morning that it was almost certainly scenery related as the CTDs happened at specific locations. So today I uninstalled both and a message appeared twice during the GSX uninstall stating default files would be restored.

Clearly that was the critical thing because two flights since were error free. Two benefits to this approach rather than the wipe and reinstall option. 1) loads of time saved and 2) the problem application has been identified.

A salutary lesson even to this old hand. I was barking up the wrong tree because of a coincidence. So even when you install a clean P3D do not forget any add-ons as those could be the cause.