The infamous CTD *crash to desktop*

pcnthuziast

Posts: 1,516   +1,341
Perhaps the thread title is a little dramatic, but this phenomenon has always baffled me. I am running Windows 10 pro 64 bit and have experienced CTD's with a number of games over the last couple years. Rather than assume it was ever any particular game, I considered that it seems more like a general system issue and about creating the right conditions to give rise to it happening. Was using older video drivers for awhile and recently updated to the latest nvidia for my 1080 and now Sniper Elite 4 has had it happen 3x now, very randomly. I finally may have a clue. While playing I had msi afterburner open and after the game crashed to desktop, I paused the monitor and reviewed all the information. One thing immediately stood out to me. The cpu speed was locked into intel turbo boost (all core sync on asus board) for my non oc'ed 6700k and it appears as though the moment of this event the cpu speed dove instantly to the base speed step frequency of 800mhz. The reason I find this odd is because speed step usually settles gradually rather than instantly dropping. Like if I had exited the game normally, the cpu speed probably would have taken awhile to get down to 800mhz and probably only if I didn't touch the mouse and keyboard for a solid minute after it was closed. Temps appear fine and hwinfo shows no irregularities with the psu voltages. Any ideas?

~Update~
Just happened again and I noticed another detail. Just before the cpu usage dove from turbo boost of 4.2 to speed step base of 800, the pagefile usage dipped dramatically. It looks to have been sitting around 15gb and dropped to about half a moment before the cpu speed dropped. I have 32gb of ram (16x2).
 
Last edited:
Speculation - as to how a CTD manifests. I guess it just whacks the open window so there is nothing there (KA_BLAM) and reverts to the last knowable state which is the desktop. I think the stats you are looking at reflect this. If the active window ceases to exist, there is nothing to be 'active' about. The OS does this to protect the integrity of system memory and the resulting integrity of the files. This is a good thing. It leaves no tidy on screen error notice. This is a bad thing.

Most likely causes?

Driver issue... use DDU (https://www.techspot.com/drivers/driver/file/information/17830/) to remove all vestiges of the old drivers. Download fresh certified drivers (https://en.wikipedia.org/wiki/WHQL_Testing) correct for OS version.

Possibly heat..so open case & clean out the dust (I take my machines outside and use canned air - sometimes I need a face mask). https://www.computerhope.com/cleaning.htm

Also possibly RAM issue...but involves opening case...remove and reinsert - ground yourself while doing this. https://www.computerhope.com/issues/ch000423.htm

Worth checking 'Reliability History' (enter in search box & check what it knows in 'details'). Click on red circled 'x' and learn more details.

My CTDs of recent in Skyrim is caused by Skyrim (well known for it). I live with it.
 
Thanks for the reply. Never looked at Reliability History before. It shows a handful of error code 141 in the last couple days that says it is hardware. My PC is always on and I don't recall experiencing any trouble at the date/times it mentions those errors though.

~Update!~ Ok so those are the result of restarts while adjusting bios settings. Since I almost never power my pc down and barely even reboot, I have noticed sometimes a legit reboot doesn't get record properly by windows for some reason so when it gets back into Windows and Windows checks whether the last shutdown was expected ot not, it comes back as no and generates that error code.

As for your other suggestions, temps are fine for sure, drivers are up to date and I always use DDU.
 
Last edited:
What we are I hope able to see in the Reliability History is a detailed report like:

Description
Faulting Application Path: D:\STEAM\steamapps\common\Enderal\TESV.exe

Problem signature
Problem Event Name: APPCRASH
Application Name: TESV.exe
Application Version: 1.9.32.0
Application Timestamp: 51437ce5
Fault Module Name: KERNEL32.DLL
Fault Module Version: 10.0.17134.556
Fault Module Timestamp: b65166a3
Exception Code: c0000005
Exception Offset: 000170fb
OS Version: 10.0.17134.2.0.0.256.48
Locale ID: 1033
Additional Information 1: 2beb
Additional Information 2: 2beba6fb4680d73a8c78ca7c24ccdb46
Additional Information 3: 6692
Additional Information 4: 66926b94ed01e6a3828d72ce41ac00ff

Extra information about the problem
Bucket ID: 81338f51eb54b3c33e91594d31a1194b (2202639880840878411)

The 41 error is nonsense - 'you had a crash'. We should find something before that. Click on 'view technical details' for the 'critical event'.
 
No, what I'm saying is that the 141 errors are completely separate events that are nowhere near the times when I was getting CTD in the game. The CTD events are like you would expect, there is no trace of info anywhere in windows logs that even acknowledge an event of any kind took place. That is what a CTD is, literally a phantom event.
 
Oops, I misunderstood. I was thinking of the silly error Microsoft throws when there is an unexpected shutdown. I do not know what an Error Code 141 is, but its meaning may be determined by which operating system or computer system it occurs in. If it were a Dell, I might want to test HDD. Unsure.

I think you want to track it down - whether it is HDD or something else, it may be the source of the instability.
 
Back