Solved Crash During "DIR TO"

[Rikypara 0]

Hi!

I have experienced the same issue, reinstalled Visual C++ and for some flights it was ok, then yesterday it appeared again. 

I notice that on most airports there is no problem using direct to. I have the problem only arriving to Limc from the north and using direct to "Sokbu" and Rigon the IAF. Coming from the south I have never had the problem yet. I used both Navdata Pro and Navigraph. Tried also with default scenery and Simgiants scenery

[keithgiannni 25]

Dave,

 

Very helpful. What is interesting is I have downloaded the files directly from the MS site, yet some of your seems more recent than mine even though I have downloaded in the last 48 hours! For example 2005/8/10 had at least one newer file it seems, yet I thought they should be the same as we are downloading from the same source. It would be helpful to establish what are the latest versions. At least then we can all be assured of having the same files when trying to tackle the problem.

 

Best,

 

Keith 

[DaveCT2003 2553]

Keith,

 

Very good observation!  Thank you!

 

I'm not sure if it matters, but I'm running Windows 10.  Other than that, I'm afraid I don't know why there would be a difference!

 

All I can tell you is that I don't have any issues with the Airbus or any other addon that I have (far too many of any sane person).

 

 

 

[DaveCT2003 2553]

ALL,

 

The other evening I spoke with the developer (Joshua) about this problem and he later wrote the post here in this forum (see above).

 

The information you are providing is appreciated and will be valuable should this problem exist after the next A320 series update.  Until then I'm continuing to try and replicate the problem, but as Joshua said, none of the development team and none of my test team (8 people with 8 different systems/installations) have been able to re-create a Direct To induced CTD. Twice I've actually injected Dir To's every 5 minutes for a 2.5 hour flights without any problems whatsoever.

 

Hopefully this will not be an issue with the next update as Joshua said the FMGS has been changed (fingers crossed).  

 

Hang in there, and thank you so VERY MUCH for your time and effort in helping to track this down! It's really great to see such devoted and involved customers who want to help - every one of you are AEROSOFT ACES in my book!

 

Thanks again.  Any of you guys want to make a flight with me on VATSIM, just let me know!

 

Dave

 

[AUA1377 0]

I have exactly the same issue - CTD because of FMGS.DLL and I think it occurs when inserting DIRECT TO's.

 

(Already started a thread in German forum about it this summer)

 

One specific case when the FMGS CTD occured:

Flying from EDDM to LOWW, recevied a DIRECT TO SNU (=Sollenau VOR / waypoint near Vienna) by ATC yet during climb to CRZ ALT. Note: SNU was not part of my flightplan.

Route: KIRDI UL174 VELOM DCT MASUR

SID@EDDM: KIRDI1S out of runway 26L

CRZ ALT: FL270

Shortly after reaching CRZ ALT (and of course already proceeding direct to new waypoint SNU) my FSX crashed (cause FMGS.DLL).

 

Would be really great if you could fix this annoying bug...

 

[Rikypara 0]

Thank you very much Dave and all the team for your help. So we wait for the next release, when do you think will be released approximately?

 

just a question for the developers, in your test flight trying to replicate the issue do you have just Navdata Pro or a secondary database with Navigraph too?

[DaveCT2003 2553]

34 minutes ago, Rikypara said:

Thank you very much Dave and all the team for your help. So we wait for the next release, when do you think will be released approximately?

 

just a question for the developers, in your test flight trying to replicate the issue do you have just Navdata Pro or a secondary database with Navigraph too?

 

It's been tested with both Navigraph and NavDataPro, under several different cycles.

 

An FMGS CTD is quite often related to a mathematical error of some type, but we can't say that for sure in this case because nobody on the dev or test team can replicate this.  This is one of those bugs that is incredibly hard to pin down.  I know that it's been tested on at least 12 different systems, all with uncoordinated Windows updates. 

 

Anyway, we're just (if not more) frustrated with this one than customer may be, and we've spent a LOT of hours trying to recreate this with any luck.

 

I want to say again that you guys are champs for helping out with this and being very understanding that if we can't find it, we can't fix it.

 

One thing... how many of you guys routinely install Windows updates?  I'm not implying it's good or bad, I'm just curious.

 

 

[Chok 49]

1 minute ago, DaveCT2003 said:

 

It's been tested with both Navigraph and NavDataPro, under several different cycles.

 

An FMGS CTD is quite often related to a mathematical error of some type, but we can't say that for sure in this case because nobody on the dev or test team can replicate this.  This is one of those bugs that is incredibly hard to pin down.  I know that it's been tested on at least 12 different systems, all with uncoordinated Windows updates. 

 

Anyway, we're just (if not more) frustrated with this one than customer may be, and we've spent a LOT of hours trying to recreate this with any luck.

 

I want to say again that you guys are champs for helping out with this and being very understanding that if we can't find it, we can't fix it.

 

One thing... how many of you guys routinely install Windows updates?  I'm not implying it's good or bad, I'm just curious.

 

 

 

I run windows update on a monthly basis, the last on being 3 days ago.

 

As of today i did three flights after reinstalling all my C++. In those flights i went DIR TO every single point in my flight plan and i did not get a CTD. i'll report if this continues to work. (hopefully!) 

[Rikypara 0]

2 minutes ago, DaveCT2003 said:

 

It's been tested with both Navigraph and NavDataPro, under several different cycles.

 

An FMGS CTD is quite often related to a mathematical error of some type, but we can't say that for sure in this case because nobody on the dev or test team can replicate this.  This is one of those bugs that is incredibly hard to pin down.  I know that it's been tested on at least 12 different systems, all with uncoordinated Windows updates. 

 

Anyway, we're just (if not more) frustrated with this one than customer may be, and we've spent a LOT of hours trying to recreate this with any luck.

 

I want to say again that you guys are champs for helping out with this and being very understanding that if we can't find it, we can't fix it.

 

One thing... how many of you guys routinely install Windows updates?  I'm not implying it's good or bad, I'm just curious.

 

 

I usually install Windows update... As they are available I do.

[Francescom 4]

Quote

One thing... how many of you guys routinely install Windows updates?  I'm not implying it's good or bad, I'm just curious.

 

Hi to all,

I run Windows Update yesterday and I had 71 Updated to be done, nothing related to C++ runtime. I tried a flight just after the updating and no CTD happened (but I think It is only a case).

 

I don't know why, it is only a sensation, but as explained by Dave I think that the problem could be a math problem that happened in some cases where the calculation of the turn Radius is complicated. I say this because yesterday I performed the same indentical route without problem but just because the Turn radius probably was ok. 

It is strange that, depending of the position where DIR TO is applied or not (in the same route), the system can crash or not. The only thing that is changed is the trajectory calculated by FMGS to acquire new Waypoint...

 

I will continue to test it..

[keithgiannni 25]

Hi,

 

Managed to force another CTD DIR TO on the EGLL to LFPO route as I described above. It sometimes works when selecting direct to Bobsa, then to fly direct to Odilo, then selecting Vasol and then P0615 in quick succession. Dump file attached for anyone's info. I have also attached what I believe to be the most up to date C+++ files. As I mentioned I noticed Dave had some newer ones than mine. Usual health warning here and ensure you check and backup etc. Please let me know if anyone finds newer files.

 

This may sound silly so forgive me. However, when I have incurred a CTD it is normally when I select the DIR button, then select the waypoint then activate the direct routing in very quick succession (within a second), if I get a CTD, I get it then. For three flights I selected DIR, then the waypoint then waited for at least five seconds before activating the route. All good. Of course, may simply be a coincidence. I mention this as seeing FrancesCom comments, (which I think has some substance), maybe slow fingers rather than fast may help?!

 

Best

 

Keith

 

DUMP2.zip

CPLUS.zip

[DaveCT2003 2553]

Thanks guys,

 

I'm not sure that the speed of selecting buttons would cause a CTD, but it's still a great observation and I thank you very much for making it. It shows you're in the game!

 

It certainly is possible for a turn greater (or lesser) than a certain number of degrees to cause this problem, so if you guys can monitor that before selecting Dir To I'd really appreciate it.  I'll try it as well.

 

Thanks again guys!

 

Only for the guys who has posted about this problem... the team I manage are seasoned Beta Testers for a number of developers including Majestic, Aerosoft, TFDi, SunSkyJet, REX Game Studios, and other developers. Asa way to thank you for your support and understanding, I'd like to invite any of you who posted about this problem to join us for a VATSIM flight sometime.  Just PM me and I'll send you the details.

 

[keithgiannni 25]

Sadly another CTD a short while ago. ESGG to EGLL. DETNA1M RW03 Vatsim. Through 3000' cleared direct DETNA. Turn of approximately left 40 degrees. Within 60 seconds of being on track, CTD. Dump file attached. Are the files still of any value? I think the Airbus will be parked in the hangar soon. Getting this every other flight now. C++ files reinstalled yet again. I cannot do anymore.

 

Best,

 

Keith

DUMP3.zip

[bpcw001 28]

I am getting these DIR TO crashes as well. I have a couple of VC++ redistributable versions installed. Should one only have the latest versions for each (2005, 2008 etc) and deinstall the older versions or is it ok to have several versions for example for the VC++ x86 2005 redist package?

 

I have to say that I did not have this dir to crash problem on my old computer running Win 7 pro x64 on an i7 core extreme 990X but it occurs now on a new build based on an i7-4790k with Win 10 pro x64 and everything freshly and cleanly installed from scratch. 

I don't know if the MCDU makes any use of the newer features of the newer processor and hence causes these problems .

 

The new box runs 4h flights with the A319/320 just fine when not touching the dir to function.

[DaveCT2003 2553]

14 minutes ago, bpcw001 said:

Should one only have the latest versions for each (2005, 2008 etc) and deinstall the older versions or is it ok to have several versions for example for the VC++ x86 2005 redist package?

 

Thanks so very much for the information!

 

It's fine to have different versions, please don't uninstall any of them as some software uses one version, other software uses another version, etc.

 

I run the same processor as you do, so it's not processor related. I also use Windows 10/64.

 

Thanks again!

 

[DaveCT2003 2553]

8 hours ago, keithgiannni said:

Sadly another CTD a short while ago.

 

Keith, file collection may be of value long term, especially if the problem continues after the new build. Until then, we just don't know but your effort is VERY appreciated.  These types of problems are just so hard to track down, so the info may very well make the difference in figuring it out IF the problem persists through the new build.

 

Also, thank you so very much for the degrees of turn for the Dir To. This is very import in determining if the problem is mathematical or not.

 

[DaveCT2003 2553]

 

8 hours ago, keithgiannni said:

I think the Airbus will be parked in the hangar soon. Getting this every other flight now. C++ files reinstalled yet again. I cannot do anymore.

 

Yeah, I understand how you feel, and my hearts with you on this one.  Thank you for all your effort Keith.  It means a lot!

 

 

[bpcw001 28]

Hello,

just two more things from my side:

 

First:

I don't know if it is related to the DIR TO problem or not, but I also get occasional crashes in FMGS.DLL (although not has frequent as when using DIR TO) in situations like these:

1. The aircraft is following the flight plan and is inbound to waypoint "X" but has not reached it yet. Now, ATC tells you: "after X proceed direct Y", with "Y" being a waypoint part of the route further down the flight plan.

2. Modify the flight plan so that you insert wpt "Y" directly after wpt "X" by typing the name of fix "Y" and LSK it below fix "X".

3. You get the according yellow-colored modified flight plan

4. Press execute => CTD due to FMGS.DLL

 

Second (subjective impressions, so take with a grain of salt):

Using DIR TO usually appears to work for me in situations when there are a couple of waypoints in a rather straight line ahead of the aircraft, and I use DIR-TO to jump over a couple of those waypoints to a waypoint which can be reached by more or less keeping the heading/track that is currently being flown anyway, with minor adjustments.

I have seen the DIR TO crashes appear mostly when the DIR TO required a bigger change in heading/track (talking about more than 10-15° here), being to the right or left did not matter.

 

Also, I am not sure if using DIR TO to a waypoint that is located on a different airway from the one that the aircraft is currently on makes a difference here. Maybe something to look into.

Yesterday, I got "shot down" by the DIR TO CTD when doing a DIR TO waypoint BALSI(UN852) after having passed VEROX(UN869) on the following route:

EDDH IDEKO Y900 TIMEN UL126 WRB UN850 NATOR UN869 MILPA UN852 VERSO UL129 LUNIK LEPA

 

 

 

[keithgiannni 25]

Dave,

 

Your enthusiasm and support is inspiring!! Bus cleaned and out of hangar. Logging started. Berlin for late lunch in 60 minutes!!

 

Keith

[Francescom 4]

4 hours ago, bpcw001 said:

I don't know if it is related to the DIR TO problem or not, but I also get occasional crashes in FMGS.DLL (although not has frequent as when using DIR TO) in situations like these:

1. The aircraft is following the flight plan and is inbound to waypoint "X" but has not reached it yet. Now, ATC tells you: "after X proceed direct Y", with "Y" being a waypoint part of the route further down the flight plan.

2. Modify the flight plan so that you insert wpt "Y" directly after wpt "X" by typing the name of fix "Y" and LSK it below fix "X".

3. You get the according yellow-colored modified flight plan

4. Press execute => CTD due to FMGS.DLL

I had one crash during a similar event. I was Inbound TAQ VOR and the ATC gave me a short cut to a Waypoint beyond TAQ VOR. I pushed DIR TO to go to the new waypoint. After that I noticed that in the MCDU FlightPlan there was and FPL Discontinuity. I deleted the Discontinuity and the system crashed with FMGS.dll error. I that events I assigned the fault to the DIR Function because pressed some seconds before..

 

Quote

Second (subjective impressions, so take with a grain of salt):

Using DIR TO usually appears to work for me in situations when there are a couple of waypoints in a rather straight line ahead of the aircraft, and I use DIR-TO to jump over a couple of those waypoints to a waypoint which can be reached by more or less keeping the heading/track that is currently being flown anyway, with minor adjustments.

I have seen the DIR TO crashes appear mostly when the DIR TO required a bigger change in heading/track (talking about more than 10-15° here), being to the right or left did not matter.

 

It is my same impression...Generally I noticed that the crash happens when there is a large turn or a complicated Calculation of trajectory to get the new Waypoint.

 

Quote

so, I am not sure if using DIR TO to a waypoint that is located on a different airway from the one that the aircraft is currently on makes a difference here. Maybe something to look into.

Yesterday, I got "shot down" by the DIR TO CTD when doing a DIR TO waypoint BALSI(UN852) after having passed VEROX(UN869) on the following route:

EDDH IDEKO Y900 TIMEN UL126 WRB UN850 NATOR UN869 MILPA UN852 VERSO UL129 LUNIK LEPA

 

I tested also this option. I Jumped from UNITA on UM858 to BTA VOR or from TIDKA on UL50 to BETEN on UM729 directly for the moment without problem. 

[DaveCT2003 2553]

8 hours ago, keithgiannni said:

Dave,

 

Your enthusiasm and support is inspiring!! Bus cleaned and out of hangar. Logging started. Berlin for late lunch in 60 minutes!!

 

Keith

 

Thank you Keith, I can't tell you how much that means to me!

 

Thank you again my friend!

 

[DaveCT2003 2553]

12 hours ago, bpcw001 said:

Hello,

just two more things from my side:

 

Thank you so very much for the additional information.  I'm going to try and replicate this in a flight this afternoon and I'll let you guys know how it went.

 

Thank you again, so very much!

 

 

[DaveCT2003 2553]

My friends,

 

After consultation with the development team and another Beta Tester, we decided that we don't need any more log files. If you will, please disable the log files in the Airbus Configurator(s), which should also help give you a bit better frames rates or less stutters (for those who may need them).

 

The information you guys have provided to date is so very, very much appreciated!  Really super stars for helping with this, and I'm not the only one on Team Aerosoft who appreciates your time and effort! While we haven't found the problem yet, we've ruled out a number of things and that is almost as important!

 

By all means, please stay in touch and provide any more observations you have, and I will do the same.

 

Remember to PM me if you'd like to join me for a group flight.  By the way, how many of you guys fly offline (have never flown  on VATSIM or IVAO)?  I ask, because I used to teach new VATSIM pilots (I covered EVERYTHING) using Team Viewer and some excellent training materials, and as appreciation for what you guys have done here, I'd be happy to do that with just you guys IF you were interested.We can cover everything from tools to make flight planning easier, a fast way to learn and master ATC communications, how to know where to fly, how to fly Connected Flight Deck / Shared Cockpit on VATSIM, and many other things.  Just let me know!

 

Happy flights my friends!

 

 

[keithgiannni 25]

Ok, here's a way around the problem that has worked for me so far. Once airborne choose a point you wish to fly direct to (or one provided by ATC). Disengage navigation and turn the aircraft manually towards the waypoint using the ND display to guide. (It needs only to be approximate but as close as you can get I found ideal.) Once established, start deleting those waypoints above the desired one. You can  either do that individually clearing each discontinuity  or collectively by deleting them all at once. Once the desired waypoint is at the top of the FMC , thus the next point, clear any discontinuity, engage navigation and the aircraft will adjust to that course.  This new waypoint will now be in a straight line to you.

 

Now the heart beat raising option, but it is not needed, is to select DIR button and select the waypoint and engage!! I have found this works particularly on direct to points that involve those wider turns that have previously caused a CTD.

 

Best,

 

Keith

 

[DaveCT2003 2553]

Thanks for that Kevin!

 

I recommend anyone suffering from this darned problem give Kevin's work around a try.  I know we all wan to see this fixed, and the development team is aware of how serious it is. But until then, I think Kevin's idea will work really well.

 

Thanks again to everyone (I know I say that a lot, but I really do mean it guys!).