Random BSODs and a hard freeze - just upgraded CPU, GPU, and PSU.

thegreennico

Reputable
Jul 28, 2018
32
1
4,545
Hey all. So as the title reads, I just upgraded a few pieces of hardware in my computer. After the upgrade, I managed to boot my computer with no problems at all, until it started giving me BSODs. I've looked around the web, mainly on this site at several threads, trying as many solutions as I could find, and haven't had much progress at all. Some details:

Running Windows 7.

Hardware changes:
GPU: I went from a GeForce GTX 740 to a GeForce GTX 1080ti.
CPU: Unsure what had previously, as I can't find the box and I don't remember, but if it's important I can switch it back into my machine and bring it up in the Control Panel. New CPU is an AMD Athlon X4 880K with its stock cooling unit.
PSU: Thermaltake TT-500NL2NK-A to an EVGA 700 B.

I installed the GPU drivers through the disc it came with initially, but did a clean install of the most recent drivers off the Nvidia website shortly afterwards. It's been running like a charm in-game, so I don't think there's any problems with that - besides I don't think GPUs could contribute to a BSOD.

I suspect the PSU should be giving enough power to everything, as Nvidia recommends a 600w supply for this card, and I went with a 700w, so I don't think it has to do with power supply.

I just finished running memtestx86 and it came up with 0 errors, so I am doubtful it has anything to do with my RAM card.

I've also downloaded Core Temp to monitor my CPU temperature. On the desktop it tends to hove around 4-10 degrees C,and in gaming, it goes to 30-40 C (tested with roughly max settings Crysis 3 and Titanfall 2), although it's hard to tell exactly as it cools down when I tab out of the game.

As for more detail on the BSODs, I have little. They feel extremely random. One happened while I had a few desktop apps open, and was watching Netflix. Several have occurred while gaming. One occurred while I was attempting to defrag my hard drive, and another occurred while I was AFK (afraid I don't recall what was going on at the time, but I don't think it was anything stressful to the system). Those things said, I'm gaming often on this rig, and have managed to play several hours straight with the CPU maintaining those upper temps the whole time, with little to no performance drops. So, regardless of the load the CPU is taking, it doesn't seem to promote the BSOD. This leads me to believe it's not BSODing because of overheating, but I'm not sure how to be 100% sure of that.

Something to note: while installing my CPU, the cooling unit did slide around on top of the chip a good bit, as I was having trouble fitting the clips over the retention frame hooks. I had applied thermal paste (roughly the size of a BB [as recommended by a guide on this site I believe]), so there shouldn't have been any scraping or damage, just extra spreading of the paste. There was also only one port for the EPS 12V 4-Pin cable on my motherboard, while my PSU has 2 of those cables. I don't know how relevant that is.

I also had one hard freeze, where my whole computer, cursor included, simply froze, and I hard to hard reboot. I believe this occurred while running Titanfall 2, but I'm not positive. It was the only crash other than these BSODs.

To sum up, I'm not really sure what's causing them. My unprofessional opinion is that the CPU got messed up somehow, either by my installation (which I kinda doubt), there was defect in the manufacturing (which I doubt more), or it has something to do with my hard drive. I don't have much evidence for the latter, but I kind just have this feeling it might have something to do with that.

Anyone have any ideas as to what might be causing these BSODS?

Also, sorry for the long post. I want to provide as much detail as I can supply.

Edit: Just minutes after starting this thread, I started updating WHQL drivers off of Nvidia, as well as attempting a defrag my drive once more. Everything was going along fine, then my screen changed, I looked over, and it was another BSOD. Noticed it has the IRQL error - I didn't see what the whole message was but I've noticed other people talking about it and it seems like a common one.
 
Solution
Alright, so I ended up replacing my hard drive with a totally clean one. I don't believe it's been used before. Did fresh install of Windows (Win7) on it, and have been using to really good success. I've got two bluescreens and one hard freeze, but that's been over the course of four or five days, which is a huge improvement. I've only got the minidump for one of the BSODs - not sure why the other one didn't get created. I've put it into the Dropbox.

thegreennico

Reputable
Jul 28, 2018
32
1
4,545


So, my RAM card seems to be a PNY 8GBH2X04E99927-15-H. I wasn't able to find it on the PNY website, but there are a few Ebay listings that say that its got 1600mhz. So I suppose I should look to upgrade to at least a 2133mhz card? Also, I'm not sure how I can set those settings you listed, especially without being able to find the piece of hardware on the manufacturers website; if that's even relevant in light of the 1600mhz.
 

maxalge

Champion
Ambassador


no, not worth upgrading the ram unless it is defective


go into the bios and load the XMP profile for the ram, it will load the proper settings




as an aside a athlon 880k is not anywhere near strong enough to properly drive a 1080 ti
 

thegreennico

Reputable
Jul 28, 2018
32
1
4,545
as an aside a athlon 880k is not anywhere near strong enough to properly drive a 1080 ti


Oh wow, really? What would the symptoms of using them together be then? The BSODs? I've seen wonderful in-game performance on nearly max settings in two fairly heavy games, so it doesn't seem likes there's a problem there.

Also, I'll try out your suggestion with the XMP profile.
 
make sure you have update the bios or reset it to defaults after the hardware change. This will update the BIOS database of hardware settings that it sends to windows.

you should then reinstall windows if you have not done so. then install the motherboard vendors current drivers from the motherboard vendors website.

windows will detect the new hardware but it can be blocked from using the hardware settings that the bios said were in use. ie will think you have two sets of hardware, one that the bios said was there and another set that it detected. it can not reuse certain setting and has to use kind of stupid setting on the newly detected hardware. I have seen some really screwy configurations that this has caused. fast hardware devices being blocked by slow devices because they were sharing the same hardware resource allocations.
 

thegreennico

Reputable
Jul 28, 2018
32
1
4,545


I believe I've reset the BIOS now (all I saw was an option to load "Optimized Defaults", and I'm assuming that was it). I'll try out the computer now and see if I get another BSOD. If I do, I'll go back and enable to XMP again (it only went up to 1800mhz, so I'm still not sure it will be enough). If that fails I suppose I'll reinstall Windows. I'd just prefer to try alternatives before having to do that.
 

thegreennico

Reputable
Jul 28, 2018
32
1
4,545
Aaand, another BSOD. After an hour and a half of Crysis 3, running mostly fine (some performance drops here and there), it crashed during a session of Civ 5. I suppose I'll reinstall Windows tonight unless anyone has other suggestions?
 

thegreennico

Reputable
Jul 28, 2018
32
1
4,545
Alright. I reinstalled Windows, update my mobo's drivers, BIOS is up to date, GPU drivers are up to date, and while in the process of deleting Windows.old and reinstalling some programs, I got hit with another blue screen. After a total reinstall of the OS it's gotta be hardware specific right? Insufficient CPU strength or RAM speed? Maybe the hard drive is corrupted? I'm at a loss and feeling more than a little defeated.
 
if the windows 7 service pack was installed then you should provide the minidump files.
ie copy c:\windows\minidump files to a cloud server, share the files for public access and post a link.

make sure the windows 7 was updated with the service packs. current graphics drivers require windows 7 sp 2.
previous versions of windows will bugcheck in windows graphics code if you have updated your oem graphics driver.



 

thegreennico

Reputable
Jul 28, 2018
32
1
4,545


Well I'm afraid installing the SPs is giving me grief now. After trying to install SP2 (and yes, SP1 is already installed) I get an error simply saying that it failed to install. Having difficulty finding a solution to this problem anywhere.
 
remove XtuAcpiDriver.sys ( Intel extreme tuning BIOS interface driver)
it is unexpected since you have a AMD cpu


your gpu graphics driver will be unstable until you get to windows 7 sp 2. You might uninstall it until all of the windows updates have completed. then reboot and install it directly from the nvidia website.

most version of this motherboard sound driver cause gpu drivers to crash until you get a version from last year.
\SystemRoot\system32\drivers\RTKVHD64.sys Tue Apr 22 04:45:50 2014

look here for the update from the chip vendor:
www.realtek.com.tw/downloads/downloadsView.aspx?Langid=1&PNid=14&PFid=24&Level=4&Conn=3&DownTypeID=3&GetDown=false

\SystemRoot\system32\DRIVERS\Rt64win7.sys Mon Mar 17 19:27:09 2014
this version of your network driver will crash current version of virus scanners and streaming software. get a update from here:
www.realtek.com.tw/downloads/downloadsView.aspx?Langid=1&PNid=13&PFid=5&Level=5&Conn=4&DownTypeID=3&GetDown=false



 

thegreennico

Reputable
Jul 28, 2018
32
1
4,545


Alright, I believe I've got the Service Pack installed now.I downloaded it from the Mircosoft download catalog, rather than the standard Mircosoft website and it worked. I've downloaded the drivers you linked me and deleted XtuAcpiDriver.sys from System32.

Prior to doing these though, I got a separate bluescreen. I uploaded the dump file to that same link I already provided; this one is dated as 7/31/2018 at roughly 3:50 P.M. It showed up shortly after I replugged all the wires on my motherboard. A common suggestion was reading was to unplug and replug all the hardware to make sure there were no loose connections. I'm quite confident I didn't harm any of the components while doing so, so I don't believe that would be the cause of the new BSOD.

I'll test out the computer now I've followed your most recent suggestions and report back and update the minidump if anything goes awry.
 

thegreennico

Reputable
Jul 28, 2018
32
1
4,545
Alright, so I thought it was fixed. It made it through all of last night with one hard freeze (I feel like this is a separate problem), and from about 9:00 this morning to 4:30 or so in the afternoon today. Got SYSTEM_SERVICE_EXCEPTION error. Here's the whole minidump folder: https://www.dropbox.com/sh/8fqf8ma7jmi9yef/AADv7B7gmpqDlbDSOUUyxW4Aa?dl=0

This latest one is from 8/1/2018.

Edit: Another one. I'm going to continuously update the Dropbox as they happen. This one was a ATTEMPTED_EXECUTE_OF_NOEXECUTE_MEMORY.
 
system bugchecked when the CPU started running some random memory location as code.

I would suspect this driver \C:\Windows\gdrv.sys Thu Nov 30 21:40:53 2017
Gigabyte Easy Saver - mobo power utility driver it most likely does not match your BIOS version
(I would remove it to see if your problem goes away, certain driver need to match the BIOS version)


I would also remove these two overclock driver (best not to have overclock drivers tweaking voltages when you are attempting to isolated problems)
Gigabyte Easy boost related driver GPCIDrv64.sys
C:\Program Files\AMD\ATI.ACE\Fuel\amd64\AODDriver2.sys Tue Feb 11 03:06:52 2014

machine info:
Manufacturer Gigabyte Technology Co., Ltd.
Product F2A68HM-H
BIOS Version F1
BIOS Starting Address Segment f000
BIOS Release Date 12/02/2014



 

thegreennico

Reputable
Jul 28, 2018
32
1
4,545


Well, there have been a few more BSODs; Dropbox minidump is updated. That said, the computer's feeling a lot more stable overall.
 
this bugcheck is likely going to be a bug in:
NVIDIA Share.exe
you might disable the service if you are not using it. (or look for a update)
https://nvidia.custhelp.com/app/answers/detail/a_id/4228/~/disabling-the-geforce-experience-share-in-game-overlay



looks like the program made a call to the nvidia graphics driver, then windows directx kernel got a interrupt, the direct x tried to work on it but the memory address it had was bogus.


I would uninstall the tool if you are not actually using it. I would also make sure you have plenty of pagefile space on your c:\ drive (virtual memory)
tools that are doing screen captures can run out of memory.

if you change your memory dump type to kernel, I can dump out the virtual memory stats. (or just make the virtual memory space larger)

(make sure all of the windows 7 updates have been applied)
 

thegreennico

Reputable
Jul 28, 2018
32
1
4,545


Well, I wasn't able to find NVIDIAShare.exe anywhere. I did however disable the in-game Nvidia Overlay, which seems to be how to turn off the screen capture program from Nvidia. Not sure if there's a way I can completely uninstall it. I've upped my virtual memory from a little over 8,000mb to 12,000mb. Also updated minidump with two new crashes.

As an aside, I also want to thank you for your consistent responses and help. I can't imagine I would've been able to do this myself so it's greatly appreciated!
 
basically the last bugcheck was pool corruption in your storage driver. while running nvidia web helper


I would run memtest86 on its own boot image to confirm your memory timings are ok.
I would run a rootkit scan, then Malwarebytes scan (new malware runs bitcoin mining on the gpu)
I would reinstall the storage driver


then start cmd.exe as an admin then run
verifier.exe /standard /all

this will force the system to look for drivers that have common programming mistakes that can corrupt pool memory.
(pool memory is just memory blocks shared by device drivers)

note: be sure you know how to get into safe mode, if verifier detects a problem during boot it will force a bugcheck and you will have to go into safe mode and run
verifier.exe /reset
this will turn it off and allow you to boot again.
Note: always turn off verifier when you are done testing or your machine will run slowly until you do.

if you get a bugcheck with verifier enable, it should name the driver in the memory dump.



I would also check the bios to see what mode the sata controller is in (generally you will not want Ide mode, you want AHCI mode) if the controller is in ide mode, for windows 7 you have to add a driver before you change the mode in bios. google "how to enable AHCI mode after install on windows 7"

you might also check what port your drive is connected to. Some boards have ports with special custom functions.




 
this bugcheck was a pool corruption while running nvidia web helper
(I would remove it and run a Malwarebytes scan and a rootkit scan)

the pool corruption was in a storage driver, these are often attacked by malware.

notes: pool is just a memory block that gets used by drivers.

if the Malwarebytes scan does not find adware or unwanted programs then I would run memtest86 on its own boot image to confirm your memory subsystem is ok.

I would reinstall the storage drive incase it has been modified (since you have windows 7)
go into bios and see what mode the sata controller is in. Ide mode? AHCI mode?
certain ports on a sata controller can have special meanings best to use a low number port.
many machines have 2 sata controllers, sometimes you can put your drive data cable on the slower sata controller and avoid bugs in the external sata controller driver.

if all else fails you can run cmd.exe as an admin then run
verifier.exe /standard /all
this will check drivers for common programming mistakes and force a bugcheck if problems are detected. the memory dump will name the driver and what problem was found.

note: be sure you know how to get into safe mode so you can turn it off if the machine bugchecks on boot up.
turn off by:
verifier.exe /reset

 

thegreennico

Reputable
Jul 28, 2018
32
1
4,545


Alright, so Malwarebytes came up with 5 Google Chrome related threats and then quarantined them. Memtest86 came up with no errors. I ran a full GMER scan, as well as a quick one, to check for rootkits, and I've put the logs in the Dropbox. There was only file it listed in red: MbamChameleon.sys. It seems like this is a Malwarebytes file. I tried to delete it anyway and was met with several errors.

SATA controller is in IDE mode. My hard drive is in Port 0, while the disc drive is Port 1. Ports 2 & 3 are currently empty. I'll swap it over to AHCI mode.

I'm also not sure about the storage driver. Does my hard drive have a driver I need to update? Or is this something tied to Windows?
 
malware tries to infect the windows storage driver. with windows 8.1 and 10 you can run
dism.exe /online /cleanup-image /restorehealth
it will get a clean copy of modified drivers from a trusted source (microsoft servers)

windows 7 does not have that option, you have to do a repair using a known good disk image.
you can try running cmd.exe as an admin then run
sfc.exe /scannow

but most of the time malware will have modified the hidden backup copies of the files on your machine.
(which is why they added the /online option to the dism.exe command)




 

thegreennico

Reputable
Jul 28, 2018
32
1
4,545


Well I'm afraid I don't have a back up that I'm sure is clean. Would the best course of action be buying a new hard drive and installing Windows onto that?
 

thegreennico

Reputable
Jul 28, 2018
32
1
4,545
Alright, so I ended up replacing my hard drive with a totally clean one. I don't believe it's been used before. Did fresh install of Windows (Win7) on it, and have been using to really good success. I've got two bluescreens and one hard freeze, but that's been over the course of four or five days, which is a huge improvement. I've only got the minidump for one of the BSODs - not sure why the other one didn't get created. I've put it into the Dropbox.
 
Solution