Every Game Crashes

[Cavalry Canuck]

Good Day Folks,

 

 

Sorry if this seems long, just trying to include as much info as possible. Been having an increasingly infuriating issue since July. Games I used to play on the regular are crashing on a very frequent basis now.

 

 

Mobo - MSI Z87-GD65 Gaming

CPU - 4790k

Ram - 2 x 8GB & 2 x 4GB G.Skillz RipZawz X. 9-9-9-24 (2). I fucked around with the settings on them ages ago and dropped them to 7-9-8-22 (1) and the system was stable until now

Video Card - MSI GTX 1080 TI Gaming X

 

I upgraded from a 4670k and MSI GTX 770 2GB back in April/May timeframe. After driver updates, things worked great until end of  July. Prior to this upgrade, never had issues with the system since it was first assembled in 2014.

 

As I work away from home, I will be gone for 2-3 weeks at a time, and back for 1-2. Usually there are various updates waiting for me when I get back.

 

 

As to my current problem. It started with World of Tanks back in July when I got home from work. There were updates for Windows, my 1080 (both GPU drivers and some new GeForce Experience Bullshit), and updates for various games I played, including World of Tanks. There were no error codes or anything. It would just randomly crash to desktop or BSoD. The BSoD error codes usually consisted of Kernel errors, Access violations, etc. Contacted windows support, they blamed graphics card drivers, had me install new drivers for a 1080 in place of the drivers I had for the1080 Ti. Things improved temporarily for 3-4 days, then shit the bed again.

 

 

After three weeks, the next time I got home from work, towards the end of August, rather than playing WoT, I jumped on the World of Warcraft Classic. It ran wonderfully. I had my Tauren, Bullphat, up to Level 25 in 6 days. Was a great time. Didn't even check WoT for functionality as I was having such a great time with my animated, meat eating cow.

 

Now I am back at home, been so since 20 September. More updates for everything. World of Warcraft started acting up on me. On the occasions it didn't BSoD, it threw memory read or access violations at me when the game crashed to desktop. The only address I can remember the error reports citing is "0xffffffffffffffff" for both read and write violations. I hope it is obvious why it has been an easy one to remember. During thing time, the GeForce Experience will start spamming random popups asking if I want to record my game, record audio, stream it, etc. It's incredibly annoying. I set the options in GeForce Experience to not throw those popups, but every time WoW crashed, the options in settings reverted to those which allowed for those annoying popups. These crashed occurred every 30 min to 2 hrs or so. Enough to not be detrimental, but certainly annoying. In my frustration I deleted and reinstalled all my drivers, hoping it would help, and ran the error and repair scan feature on WoW. Neither fixed the problem. In fact, there was a time where Nvidea didn't recognise the presence of my 1080 Ti, despite it outputting video to only the monitor. After multiple hard resets it finally acknowledged it's existence and began displaying video to both the monitor and the tv, and thereafter let me install drivers

 

In the meantime, inexplicably, World of Tanks started working flawlessly again. Seeing as how there is an On Track to the Badger event, I've been eagerly playing it whenever WoW pissed me off too much.

 

Anyway, Warcraft updated the Classic client. Things went downhill/off a cliff immediately thereafter. World of Warcraft will do the usual CTD/BSoD after only 5-20 minute duration of play. World of Tanks will CTD/BSoD after approximately an hour of play. All other games, previously unaffected, will CTD/BSoD at random intervals, regardless of developer. Looking at things like Sim City 4, Age of Empires II and III, Diablo II, Star Trek Online, World of Warships, Hearts of Iron III, Skyrim, Kerbal Space Program, etc. They have all began to experience the same issues where they didn't before, yet only WoW is able to cite either memory read/write errors when it CTDs.

 

 

After numerous contacts to Windows support, as well as multiple tickets opened up with WarGaming, Blizzard, and NVidea, I am tired of there "professional" support. I can only uninstall and reinstall the same drivers over and over so many times. I'm basically posting this as a plea asking for any options to examine I haven't thought of prior to wiping my system, destroying my old SSDs and replacing them with new ones, and doing a fresh install with everything.

 

Thanks for your time, Matthew

 

[dgsddfgdfhgs]

looks like some hardware is dying, suspecting rams / ram slot first , do some mem test

meantime you can install windows on a new partition 

[Deetex]

Agreed with above, this sounds a bit like a Memory issue.

 

As stated above, try running a diagnostic tool, following that I recommend reseating the sticks, you never know if a wayward speck of dust, fluff or debris got in between the contacts and started oxidizing said contact. -Should be gold on gold negating that risk, but you never know.

 

Subsequently, after the next bluescreen strikes, try running WhoCrashed and see if that sheds any light on anything.

[Cavalry Canuck]

16 hours ago, Deetex said:

Agreed with above, this sounds a bit like a Memory issue.

 

As stated above, try running a diagnostic tool, following that I recommend reseating the sticks, you never know if a wayward speck of dust, fluff or debris got in between the contacts and started oxidizing said contact. -Should be gold on gold negating that risk, but you never know.

 

Subsequently, after the next bluescreen strikes, try running WhoCrashed and see if that sheds any light on anything.

Ran the Windows memory diagnostic. It came back as a fail, but it did not say what the specific failure was.

 

I removed and re-seated the sticks, and made sure they were all clean. MOBO connectors appeared to be clean as well. For shits and giggles I changed the order the sticks were in the MOBO to see if that would make any difference. While I had them pulled, I noticed that the heat spreaders were very loose, and the thermal pads were no longer firmly against the memory modules. Could my issues be caused by overheating the sticks?

[Deetex]

On 9/28/2019 at 2:41 AM, Cavalry Canuck said:

Ran the Windows memory diagnostic. It came back as a fail, but it did not say what the specific failure was.

 

I removed and re-seated the sticks, and made sure they were all clean. MOBO connectors appeared to be clean as well. For shits and giggles I changed the order the sticks were in the MOBO to see if that would make any difference. While I had them pulled, I noticed that the heat spreaders were very loose, and the thermal pads were no longer firmly against the memory modules. Could my issues be caused by overheating the sticks?

 

Okay so it seems we're looking at a definitive memory issue, next is isolating which stick, set or setting is at fault.

 

Overheating ram is not impossible, but unlikely. 

 

Did reseating help at all?

 

How quickly did the memory diag throw up an error? 

What happens when you relax the timings a bit more?

 

Let's see, you stated to have combined a 2x8 and a 2x4 sets of ram, if the diagnostic tool couldn't specify which specific DIMM is at fault, and if the issues persist, try limiting to one set, run the diagnostics(or just use it until fail) again and if so, pull one DIMM and repeat.

 

Technically speaking, once you've isolated the fault to any specific set(assuming they were bought as a set), you could issue an RMA request, again assuming warranty still applies.