okay.. I was not 100% exact enough here. That folder can you remove IF you NOT installed the packages in default location, which is in steam in %appData%\MicroFlig..\packages. ( I installed that at a different drive, therefore I forgot for a moment that fact ). All other folders /files are created at runtime ( or at first start, or while settings change like the UserCfg for setting ) and can so contain files which possible cause issues, like we know from the rolling.cache or all the cache,scenery*, simobjects, etc. files. If you not do that in that way, you not realy done a âfull cleanâ. Most users only re-install the game, but not clean former this folder and so all the files left.
So more exact: to avoid that you have to redownload the whole content, you can remove all of that files/folders within , except the âpackagesâ if the default location was not changed.
okay, thanks.. so I assume we can exclude a relation to multi-monitor systems for now.
i was following some reports for users reinstalled the whole windows, & got the error afterwards, so the leftovers are not a big deal. thats why i believe its within the internal sim files, conflic of some sort of, or something like that.
Yes, only in flightsim. And the more I read the different posts the more I am convinced that the issues is purely hardware. That explains the randomness of it, the inability for Asobo to figure out whatâs going on, and also the suggested workaround. Like underclocking, which it works to the extent that you reduce the stress/heat on the system (and the card bending in my case). As I wrote above, after installing the support I have everything overclock (cpu, mem, and gpu) and no more crashes with LOD 350 and details 200 in VR.
I got a pleasant surprise today doing some more testing.
After i had a crash using the A320 with my old settings I reduced the core frequency and memory frequency by 100MHz more and got also the A320 stable.
But with -400MHz I am running my GPU now at 878MHz instead of 1278MHz and paid with a significantly lower FPS.
Then I increased the core frequency again and to make a long story short I figured out that
the CTD is only impacted by the memory clock and not the core clock!
My new setting is now 0MHz core clock and -500MHz memory clock. I also tested with the A320 and could not get a single CTD. When I tried reducing the memory clock before I have probably not reduced it far enough and experienced a crash. So i am now happy with my frame rates and global rendering settings on a stable system. I finally also increased with MSI afterburn my fan speed to 80% and did reach more than 10C temperature reduction to only 67C max.
If you look at the vids (for example the one I posted in another thread) and listen to the feedback from others you will see there is no heat or stress on the system necessarily when the crashes happen.
I will post it again here so you can see the CPU/GPU/Memory usage as well as Temps. when the sim crashes . No stresses here.
This is 100% repeatable for me in the sim and only the sim. I have also, as have a few others, experienced the crashes stopping and being convinced I/we had solved the problem only for them to return a few days/weeks later. I have an under volt on already btw.
The idea that a âhardware faultâ only happens when using flight sim and nothing else makes zero sense to me - I would put big money on the problem being software - whether sim, drivers or whatever.
I agree with you that it seems odd that it only happens with the sim. I too had experienced it while the overall workload wasnât very high. I was able (still am) to run all the occt tests and yet getting the msg right after while simming. So, thereâs definitely something âuniqueâ with msfs, but then again, by underclocking the hw what are you really achieving other than reducing the stress on the pc? The code doesnât really change. So, in my case I was lucky and found the fix, but it might not be the solution for all.
i agree 100%. your 4 crashes in a row in the video clears it, but the mystery why does it go dark for long sometimes, if it can be consistent in this way? its one of the rarest Species xDxD
In my case, this error always appears if you change the size of the sim window and any window adjacent to it (for example, the navigraph inside the sim or something like that) RTX4090, 13700K
And it has nothing to do with the load in the msfs. Most often on the menu
I solved the problem by deleting everything along the path C:\Users\user\AppData\Local\Packages\Microsoft.FlightSimulator_8wekyb3d8ekybe\LocalCache
except Packages, which threw off all graphics settings
additionally, everything was previously deleted from the Community folder
I discovered this last year whilst I was still using my 3070. A bit of research showed that Reshade can sometimes clash with MSI Afterburner. Removed Reshade and the crashes stopped - until I upgraded to the 4070ti and moving over to DX12 and FG that is.
thank you for your message. I tried and it worked the 1st time. But then CTD again when loading another flight from the main menu. it must be said that I had only deleted the .ISON and .GPUDMP files which were dated the previous day. Apart from Packages, should the following files be deleted: .CCC, .CFG,.OPT,.BIN, .XML, CAMERAS.CFG, Simconnect.xml. And what to do for folders: Sceneryindexes (airports) Simobjects (planes), Facilities (.IDX files), Scenerycache (Empty), PC (empty too). Thank you very much for your suggestions
Hi again.
This time I have done some measures using âTech Power Up GPU-Zâ. Will share some pictures from a log and also attach the log. Here are my findings.
1.When crashing, my GPU-clock rise above what I´ve expected reading the specs for ASUS GeForce RTX 2080 Ti ROG Strix OC. 1905 MHz when expected 1665 MHz.
2. The Asus factory OC should occur when âpower is avalibleâ. At the moment when GPU-clock boost occur there is a significant descent in powerconsumption logged for the motherboard. A number that before that was very persistent. So, more power for the GPU, less for the motherboard.
3. Still. I do have a PSU valid for 1050 W. So I should be well eqquipped. At the moment for the crash the percentage for W outage is about 50 percent. Still⌠For some reason. Just some posts (seconds) before that, the outtage was over 100%. I dont have a clue why.
SoâŚmy somewhat overall conclusion is that this problem might occur on âfactory-overclockedâ graphicscards. Excactly why, I can t understand, but I am guessing ASUS (for example) are trying to push their limits by using âsoftâ code" while Asobo is trying to push theirs gathering xbox hardware with PCs.
And somewhere inbetween that lies the problem we haveâŚ
WellâŚDon´t know If this post made any sense at all, but feel free to read my uploads and see if you can see something there that you feel is same for you for example.
as far as i can tell your readings are very normal, for power draw for example, my system uses as an average 120 W just for the gpu on a LAPTOP, 1700 Mhz speed, no OC, use +33% memory than shown in your readings, thats about 10 GB VRAM as minimum (because iam using the sim in the max graphic settings), & still the sim would run perfectly for me, when this issue is sleeping,
when it wakes up, the crashes would be constant, & the ability to use the sim is not long, until this issue goes to sleep again & that can reach weeks, then awakes again, & so onâŚ
there is no clear trigger yet, but for me iam still testing if the cloud save files have any effect on this issue, & waiting for this issue to awake again to see the results. LOL.
for me it looks more like that after your GPU is a while in 100% load, the crash happens and that points not to the game as reason for the crash.
About factory OC we spoke a lot of time in the topic. If you can proof, e.g. in case you it not crash if you disabled the factory OC with setting debug-mode in NCP, then your GPU (system) have an issue.( e.g. have you noticed the hotspot temps ). Dont forget that here also RAM is highly involved, users forget that often.
My former system was with a âASUS 2080TI OCâ and I never had an issue.
Thx for the reply. I also feel my numbers are within reasonable span. Will use the same tool again soon and see if the crash again occur just after the GPU is âturboingâ.
Regarding the cloudsaves, I am looking forward to see what you find out. I am also feeling this problem is due to something that is not 100% related to my hardware. (Why does it sometimes occur while in menus, and some other time you can play without an issue for like 30 minutes).
I hope they plan to attend to this bug when they are releasing their âbug-cleaningâ update. Heard something about a fix like this on "Obsidian Ant"s Youtube channel (I think).
itâs the hotspot temp.. so no, healty max is not 125°C
When 20°K and more difference in hotspot- vs gpu- temp , then it can point to bad tempflow ( thermal paste, or bad ariflow, etc ).
You are below 100°C or near the 100°C with the hotspot temp, so it must not be âtheâ problem, but its just a hint.
PS:
if you not use the Low-Power-Mode, then in menue is a very high GPU load.. In special my 2080TI run at 100% in menu to render that useless background hangar ( near 4K ). That was fixed with the mentioned Low-Power-Mode.
125 is not considered as a âhealthyâ temp, but, as I read it, its a max temperature. So 100 C is well within reasonable temperature.
At the moment I am still at a state that I will say: âPlease, Asobo/MS, fix thisâ as it feels obvious that the solution lies within some of their âoptimisationsâ they made since release dateâŚ
UNtil its solved, I do rest my case on the âbuy-buttonââŚin sim market and on 3rdpartysitesâ
I couldnât fly since 1.30.12.0 becuase of this error.
Now I am flying fine.
This is what I did.
And now, for the last few days, I havenât even needed to reinstall the driver continuously. It has strangely become reliable again since the last clean install of nVidia 528.49 Studio driver.
I am just now completing a 7 hour flight from Switzerland to Northern Canada with zero problems.
