Getting CTDs again…

[RobJC 2,069]

So far…

CTDs…

TBM
CJ4 (WT)
DC-6

No CTDs…

Caravan (modded or unmodded)
CRJ
787 (unmodded)
King Air

Edited September 23, 2021 by RobJC

[Tom_L 533]

6 hours ago, RobJC said:

CTDs...

CJ4 (WT)

No CTDs.....

CRJ

 

Yep, can confirm this, but I also have no CTDs with the TBM (modded) and WT G3000. I just had an interesting conversation with Matt Nischann on the WT Discord. They are aware that some of us are having issues (with the CJ4), but as long as there are no reproducable conditions under which these CTDs happen ("when I do this or get in this situation, then I always crash" - and I assume there's other factors like our individual configuration involved), there seems to be only a slim chance to fix these issues. Even if these conditions are identified it would take someone to "hit it while debugging, and that it's actually the issue causing the crash, and if you do, that you can actually determine the cause from that one instance". That's why I have hangared my CJ4 with a heavy heart until SU6, 7, 8........

[RobJC 2,069]

This is what I was told by a developer working on the NXi mod...

Well. I think its in general a timing issue that needs to be looked at inside the engine. And it was two possible things:
- Initializing the Bing maps
- Initializing the intercept for autopilot events

The instrument will call the sim like "hey i want a map" and the sim creates the map instance and comes back.
Although during flight load, the instruments are refreshed 2-3 times (for whatever reason).

So I think what is happening is, that the sim calls back for a request that isn't alive anymore because the instrument was refreshed.

These issues mostly started when multi-threading was added to SU5 to improve performance. In some cases based on the components in your PC, you don't get a crash because things come together at the right time. But in other cases things don't come together, and then you get a CTD because the request isn't alive anymore. He was able to fix the NXi CTD that was impacting some (including myself) most likely by making another call or verifying the object was still alive. Now keep in mind this is just one aspect of a very large project. These type problems could be littered throughout the code, which is why Asobo needs to fix the engine. 

[fogboundturtle 2,708]

Just now, RobJC said:

This is what I was told by a developer working on the NXi mod...

Well. I think its in general a timing issue that needs to be looked at inside the engine. And it was two possible things:
- Initializing the Bing maps
- Initializing the intercept for autopilot events

The instrument will call the sim like "hey i want a map" and the sim creates the map instance and comes back.
Although during flight load, the instruments are refreshed 2-3 times (for whatever reason).

So I think what is happening is, that the sim calls back for a request that isn't alive anymore because the instrument was refreshed.

These issues mostly started when multi-threading was added to SU5 to improve performance. In some cases based on the components in your PC, you don't get a crash because things come together at the right time. But in other cases things don't come together, and then you get a CTD because the request isn't alive anymore. He was able to fix the NXi CTD that was impacting some (including myself) most likely by making another call or verifying the object was still alive. Now keep in mind this is just one aspect of a very large project. These type problems could be littered throughout the code, which is why Asobo needs to fix the engine. 

Do you have any empirical data to prove your hypothesis ? The Developer mode can give you access to a console and a tons of data.

[RobJC 2,069]

Just now, fogboundturtle said:

Do you have any empirical data to prove your hypothesis ? The Developer mode can give you access to a console and a tons of data.

The developer working on the NXi mod confirmed this first hand, made the fix and that particular problem is not an issue anymore. That is all the data I need, but feel free to knock yourself out with developer mode. 

[fogboundturtle 2,708]

9 minutes ago, RobJC said:

The developer working on the NXi mod confirmed this first hand, made the fix and that particular problem is not an issue anymore. That is all the data I need, but feel free to knock yourself out with developer mode. 

If Matt says so, than it's good enough for me. I just wanted to understand how you came to this conclusion.

[Bunchy 227]

15 hours ago, Tom_L said:

Yep, can confirm this, but I also have no CTDs with the TBM (modded) and WT G3000. I just had an interesting conversation with Matt Nischann on the WT Discord. They are aware that some of us are having issues (with the CJ4), but as long as there are no reproducable conditions under which these CTDs happen ("when I do this or get in this situation, then I always crash" - and I assume there's other factors like our individual configuration involved), there seems to be only a slim chance to fix these issues. Even if these conditions are identified it would take someone to "hit it while debugging, and that it's actually the issue causing the crash, and if you do, that you can actually determine the cause from that one instance". That's why I have hangared my CJ4 with a heavy heart until SU6, 7, 8........

If it gives Matt a clue, I have repeatble CTD's with the CJ4 (no other mods) on a KPBI to KOXC flight. The CTD's happen on airway Y299 at and around duunk waypoint area. 

 

KPBI/28R SLIDZ2 LENDS AR16 SNABS Y299 JRDAN Y291 SAGGY DCT DIW J174
ORF V139 BRIGS DCT DPK DCT KOXC/36

 

I have also had many CTD's of a random nature on many different flights that have not been repeated when repeating the same flights.

I would love to share this info on their Discord but I can only join as a guest. I can't login with any account I have already claimed. I hate Discord. Unnecessarily word not allowed!

 

Stu

 

[spitzer45 99]

Just crashed today with fbw 320. Cancelled cpu overclock, then it was good. 

[Hernan1 5]

Good for you Spitzer45, but how about the rest of us who simply play this game? What is a CPU Overclock? Is tthis a secret that only the "elite" MSFS users know about? How about sharing your solution with the rest of us.

 

 

[Gate99 246]

27 minutes ago, Hernan1 said:

 What is a CPU Overclock? 

 

 

Google is your friend 🙂

[vrdubin6 184]

9 hours ago, Bunchy said:

If it gives Matt a clue, I have repeatble CTD's with the CJ4 (no other mods) on a KPBI to KOXC flight. The CTD's happen on airway Y299 at and around duunk waypoint area. 

 

KPBI/28R SLIDZ2 LENDS AR16 SNABS Y299 JRDAN Y291 SAGGY DCT DIW J174
ORF V139 BRIGS DCT DPK DCT KOXC/36

 

I have also had many CTD's of a random nature on many different flights that have not been repeated when repeating the same flights.

I would love to share this info on their Discord but I can only join as a guest. I can't login with any account I have already claimed. I hate Discord. Unnecessarily word not allowed!

 

Stu

 

Similar situation for me. Trying to do a flight form KVGT to KASE the other day and I had 3 crashes in a row pretty much at the exact same spot from LAS MMM via V394. I just gave up and flew KJAC to KASE with no issues.

Sometimes I just change the scenario a bit, something as small as changing which parking spot I'm at, and it seems to fix the issue. Yesterday I was trying to fly out of KASE and I got a CTD about 3-4 times while setting up the FMS. Changed my parking spot and had no issue.

[Noel 4,502]

1 hour ago, Hernan1 said:

What is a CPU Overclock? Is tthis a secret that only the "elite" MSFS users know about? How about sharing your solution with the rest of us.

🤣

[Mikeingreen 139]

I am more than pleased to say that after changing out my PC3600 DDR4 to a set that says it is Ryzen ready, I have now went all this week without a CTD. I have also been adding things back to the Community folder and things are still working fine with no CTD's, which I had been plagued by ever since the Beta for SU5. It's nice to be able to just FLY without issues.

[MikeyOnTheFlightDeck 193]

Strangely started getting CTDs again using FBW experimental. I know it’s buggy but previously ran fine. I did install a couple of mods recently so will delete these and do a clean install of FBW to see if it makes a difference. (One CTD happened when I went to print out the METAR which I believe is a known fault). The other CTD was when I replugged my Xbox controller in which haunts me a little bit because it takes me back to the bugs in August 2020!

Edited September 25, 2021 by MikeyOnTheFlightDeck

[Bunchy 227]

7 hours ago, Mikeingreen said:

I am more than pleased to say that after changing out my PC3600 DDR4 to a set that says it is Ryzen ready, I have now went all this week without a CTD. I have also been adding things back to the Community folder and things are still working fine with no CTD's, which I had been plagued by ever since the Beta for SU5. It's nice to be able to just FLY without issues.

I'm not sure having to replace RAM after a sim update is very sustainable! 

Pretty sure your original ram was good. If problems started after SU5 it is likely to be the software. SOOOO many people having CTD's after SU5...... It can't be everyones hardware is suddenly broken.

For me SU5 CTD's were calmed a little after the hotfix, but they still remain after WU6 and its hotfix. But only with the CJ4 WT.