Blue Screen on Windows 8.1 mostly IRQL_NOT_LESS_OR_EQUAL

[ErikCJ]

Hi everyone, I'm new here and I don't know where to post, so I post this thread under Windows 8, but maybe it should be under Drivers/Motherboard. So I'm getting frustrated with blue screens that are occurring on my computer and it seems I ran out of solutions or maybe I'm not such an expert to solve it. So let's better describe my problem. Blue screens happen randomly, it could be 5 mins after startup, it could be 8 hours or more(record was 3 days). Blue screens have many different bug checks, most of them are IRQL_NOT_LESS_OR_EQUAL. How it all started - my old motherboard died, computer didn't boot, no beep sounds after breadboarding and no beeps after removing all devices from motherboard, so I bought new one and new processor, all other hardware I didn't change I just adjusted motherboard to it. After installing new motherboard the problems occurred. I tried many solutions, one of them was Memtest x86 and I tried to find some errors on RAM, but there were none. I did this 3 times and ran through all passes with 0 errors. Then I tired driver verifier and found that drivers from Saitek(I'm using RAT 7 mouse) were causing blue screens on startup so I thought maybe this was the problem. I uninstalled the drivers and removed mouse, plugged in another wireless mouse and waited if the blue screen would occur and it did later on. Now I'm using it back and I'm certain that mouse is the part of the problem and maybe there are more devices causing BSOD. I used bluescreenviewer to see what service/program crashed and it was mainly ntoskrnl.exe. I then updated drivers for Intel network adapter, graphics, downloaded all drivers from official site of motherboard and updated Saitek drivers and still no luck. I also noticed something strange, my computer needs a long time to POST and the beep sometimes occurs after 20 seconds. I didn't have any other PSU to try if that's the problem, and the old PSU is 5-6 years old. I also tried solution posted on this forum which is focused on IRQL_NOT_LESS_OR_EQUAL and I didn't work, I also used many cleaners so I lost many dump files, but I still have more recent ones. The problem still persist if I format computer, so my closest bet is that the new motherboard is not working properly or maybe it's the PSU, but I doubt it. I also tried running in safe mode and had no bluescreens which indicates that there's maybe driver problem, I don't know anymore what it could be. I would be gratefully thankful if someone would take some time and tried to solve my problem.

My rig:
Motherboard: Gigabyte GA-H97-D3H (new)
Processor: Intel Xeon E3 1231 v3 (new)
Graphics: AMD Radeon 7970
Ram: DDR3 1333 MHz
SSD: Crucial BX100 250 GB
PSU: Chieftec APS-650C

So yeah maybe I forgot something, just ask me and if you have any solutions I'll gladly try them out, otherwise I'm thinking of returning motherboard and processor, maybe it's something wrong with new hardware.
Here are also dump files from blue screens:
Dump files

 

[ErikCJ]

After installing driver for audio device I've waited what will happen. And after 3 hours BSOD came. Does this suggest something is wrong with Realtek driver? I downloaded it directly from Gigabyte official site for my motherboard. And here's the link to BSOD. If I don't install Realtek drivers/program I can't use my front panel of audiojacks.

Edit: Now I've downloaded latest drivers from Realtek and will wait what will happen next. And I was also thinking, could be that front jacks are causing BSOD? I have headset plugged in most of the time.

Edit No.2: I now checked case, and I had AC'97 plugged in and had High Definition Audio Drivers. I don't know if AC'97 is compatibile with HDA drivers, but could this be the problem? I now plugged in HDA and I'm waiting for any changes.

 

[johnbl]

generally, audio driver conflicts are in the DMA channel used. Two or more audio devices using the same DMA channel.
in my experience people often have audio sources on their machine that don't actually have speakers connected to them and they don't realize that the system still sends signals to the device. You might look in your windows control panel hardware for devices like your video card that have High definition sound support and disable them if you don't get sound to your speakers in your monitor VIA a displayport or HDMI port. (just disable the driver in control panel, if you remove it it will be reinstalled by plug and play if you don't use the motherboard sound you and disable it in BIOS and the driver should not load in windows) The debugging code for the audio will not be in a minidump file, you have to have a kernel dump and even then it is not too helpful. I have never had to debug it much, most people just disable devices they don't want, reset the BIOS to defaults so the BIOS will scan and report the hardware change and resolve the DMA conflicts, and had off the settings to windows.

also, in general the last audio driver installed "wins" in terms of conflicts. Just means it might be ok until you update a different audio source like a graphics card.

 

[ErikCJ]

I can't believe how drivers can screw up everything. This is almost stupid how many problems does this computer have. I now recieved BSOD again with bugcheck 0x139 after 2 days running without it. It's getting on my nerves I don't even know what to do anymore. I have speakers that I use and they are plugged in back panel of computer and I also use headset which are plugged in front panel. I had this combo before replacing mobo and processor and EVERYTHING worked just fine, why shouldn't it work now. I just can't see the problem anymore it's ridiculous and hopeless. I won't bother posting any links to BSOD like you said, you can't do anything I guess I'm stuck with this pile of cr*p of a computer.

 

[johnbl]

all you can do if you can not figure out the cause of the problem is
start cmd.exe as an admin, then run
verifier.exe /standard /all

if windows finds a bad driver it will do a bugcheck and name the driver in the memory dump.
Note: use verifier.exe /reset
to turn off the error checking when done testing.

you would also want to change the memory dump type from minidump to a kernel memory dump or the debug info will be removed from the memory.dmp file.
https://www.sophos.com/en-us/support/knowledgebase/111474.aspx.

basically, your BIOS will setup the logic for your system, the drivers provide a interface between windows and the actual hardware. Most often Bugs will be found in OEM custom drivers, it is simply because the fixes are not given to microsoft and pushed out to the machines during windows update. The OEM expects you to get the driver from the motherboard vendor or scrounge them from various websites. The end effect is that most of the problems are in the BIOS and the OEM drivers and most often just require a update.

The problem you have can be debugged only with a kernel memory dump. Even then you might find that you will want to start removing some of the OEM drivers for special devices. Often for buggy OEM drivers the order of the install actually determines if you see bugs or not. (the cause of the bug can be shown in the kernel or full memory dumps)

also, certain processors have a known bug list for the electronics, these are fixed via patches to the BIOS. If the motherboard is BIOS is not patched, then there is a driver that is loaded in windows that does a fix for windows. You don't get this updated driver unless you update the CPU chipset drivers from the vendor of the CPU. (intel in your case)
I don't know if it applies in your case, I would have to dig into the known bugs for the CPU or chipset.
Also, the motherboard vendor will also make bugs in the motherboard logic, only they will know if that is the case.
(most people never figure out problems or figure it is a manufacture defect and get a new board)
if it is a motherboard logic defect the vendor has to make a custom driver or BIOS update to work around the problem.

Note: The best chance for success on most systems is to update the BIOS to current, reset the BIOS to defaults. Most testing will have been done with the default settings.

I can't believe how drivers can screw up everything. This is almost stupid how many problems does this computer have. I now recieved BSOD again with bugcheck 0x139 after 2 days running without it. It's getting on my nerves I don't even know what to do anymore. I have speakers that I use and they are plugged in back panel of computer and I also use headset which are plugged in front panel. I had this combo before replacing mobo and processor and EVERYTHING worked just fine, why shouldn't it work now. I just can't see the problem anymore it's ridiculous and hopeless. I won't bother posting any links to BSOD like you said, you can't do anything I guess I'm stuck with this pile of cr*p of a computer.

 

[ErikCJ]

There's one problem I have with driver verifier: If I check drivers for ATI then computer won't even start I will only get black screen. I'll try it without ATI but how long should I have it enabled? Till the blue screen happens?

Note: I already had kernel memory dump enabled and I already updated BIOS to latest version and set "Load optimized settings".

 

[johnbl]

verifier.exe should be able to work on the ATI drivers without causing a black screen.
it can change the timing of the code execution of the drivers which can lead to deadlocks in poorly written code.

You might totally remove the ATI drivers including the ATI hd sound driver for the card and run with the generic VGA driver to see if you still get a bugcheck.

Generally, you would just run with verifier.exe until you bugcheck.

I would go ahead and run it with the ATI drivers either update/ removed or excluded from verification and see what the kernel memory dump shows.

Oh, you can also get memory corruption because of firmware in a SSD, run crystaldiskinfo.exe to check your SSD for its errors and firmware version. 4/14/2015 firmware update:
http://forum.crucial.com/t5/Crucial-SSDs/Feedback-Thread-Firmware-MU02-for-BX100/td-p/167261

There's one problem I have with driver verifier: If I check drivers for ATI then computer won't even start I will only get black screen. I'll try it without ATI but how long should I have it enabled? Till the blue screen happens?

Note: I already had kernel memory dump enabled and I already updated BIOS to latest version and set "Load optimized settings".

 

[ErikCJ]

The BIOS update did work, don't get me wrong. I don't get IRQL_NOT_LESS_OR_EQUAL BSOD's. I did fix processor compatibility and errors that it had. The last BSOD's I got was two KERNEL_SECURITY_CHECK_FAILURE and one REFERENCE_BY_POINTER and they all had ntoskrnl.exe+150ca0 and 0x139 bugcheck both had 15c7e9 in stack address. I know these numbers don't help you but could this indicate the same driver? Well if it's driver then what driver could it be if all are updated. I only started recieving this blue screens when I installed Realtek. But then I realised that I had installed HDA drivers and my front panel used AC'97. So I switched cable and started using HDA as the main cable and thought that would fix the problem and computer worked for 2 days straight. The latest thing I plugged in is Xbox 360 wireless controller for PC and I didn't download any drivers for it the computer recognized it alone. I also installed Skype and Origin and downloaded FIFA 15. Before I wanted to play the blue screen happened (I didn't start game).

Yea and right now I'm running the Windows 8.1 on HDD just in case something is wrong with SSD.

 

[ErikCJ]

There's one driver I never installed from Gigabyte's motherboard official site... I was Intel VGA Driver but I don't think I need it, I don't have VGA enabled (which is on motherboard), but could this cause problems?

Edit: The updates from official site are meant for processor graphics, but I have Xeon which doesn't have graphic interface. I also enabled verifier and suprisingly it's now working with ATI drivers. Well what now... I wait for blue screen to occur?

 

[ErikCJ]

Talking about the devil... another IRQL BSOD happened today. I had driver verifier enabled. Should I post DUMP?

Edit: I once read that IRQL bluescreens are commonly happening because of outdated Network adapter drivers. Mine is from 2014/05/02 (I downloaded it from official site). I'll search for latest ones.

 

[johnbl]

sure, put the memory dump on a server.
for certain errors you have to run verifier and change the memory dump type to kernel memory dump or the debug info will not be saved to the memory dump file.

if verifier shows a error at least it will point to the driver even if the debug info is not saved.

old network drivers even from 2013 should be updated, bugs in older drivers tend to cause problems in virus scanners, online streaming games and nvidia graphics cards (you have a amd card) generally a 2014 driver is pretty good unless it is a driver that tries to rearrange packets for online games. (used to kind of cheat at online games by delaying your position packets in first person shooter games)

Talking about the devil... another IRQL BSOD happened today. I had driver verifier enabled. Should I post DUMP?

Edit: I once read that IRQL bluescreens are commonly happening because of outdated Network adapter drivers. Mine is from 2014/05/02 (I downloaded it from official site). I'll search for latest ones.

 

[ErikCJ]

This one is quite big(~1GB): LINK

But I'm not sure that it happened because of driver verifier. More like another random BSOD.

 

[ErikCJ]

Question: Is it bad that RAM is running with 1330,55 MHz and not 1333 MHz?

Here's the picture that I took and it's from CPU-Z: RAM.jpg
DRAM is only running at 665,1 MHz. Is it dropping or is this normal? And is this affecting CPU which has 1333 MHz memory type support.

 

[johnbl]

the memory dump showed a bad address for a delayed procedure call.
I would update the intel management engine drivers from
http://www.gigabyte.com/products/product-page.aspx?pid=4962#dl
they have new versions dated 2015/05/27
you had a old version installed from 2013.

I will try to take another look at the memory dump, but I have to run errands with my wife in a minute

Question: Is it bad that RAM is running with 1330,55 MHz and not 1333 MHz?

Here's the picture that I took and it's from CPU-Z: RAM.jpg
DRAM is only running at 665,1 MHz. Is it dropping or is this normal? And is this affecting CPU which has 1333 MHz memory type support.

 

[ErikCJ]

Alright then here is what I got for you: I again reinstalled everything on SSD. I disconnected all the unnecessary devices from computer and I'll slowly connect them back and I'll see what happens. For now I only installed 3 drivers: audio, network and chipset. I didn't yet update AMD drivers and I didn't get any blue screen for like 5 days so I think you were right. The problem is with audio device(HDMI) that causes random blue screens but mostly Rezultati
SYSTEM_SERVICE_EXCEPTION blue screen. I'm 80% sure it's this and you said it yourself although I didn't test it because I wanna make sure every device is working properly. Yesterday I plugged in front panel with audio jacks and today I got blue screen Bug Check 0x133 DPC_WATCHDOG_VIOLATION and I believe it's related to procedure calls as I read about it. I didn't yet update Intel Managment Engine Interface (I'll do it right away) and this is probably the cause of the latest blue screen. If you think that something else could also do this, then you can just tell me. I believe we are on good path right now, but I guess you never know.

 

[ErikCJ]

After trying devices out, I didn't get any blue screens. Then I decided to start up Minecraft and while playing for some time I got blue screen. Then I updated ATI drivers because that was the only thing I didn't update and thought that was the culprit because of some blue screens that indicated video problem. I also disabled AMD High Definition Audio Device. I then went back testing and for purpose of it I started Minecraft again and waited. At the end nothing happened so I closed it down and then it did happen - blue screen appeared and it was old friend IRQL_NOT_LESS_OR_EQUAL. I thought there is something connected to Minecraft and ATI driver or maybe PSU. So I enabled driver verifier and went back in Minecraft. After aprox. 10 min I got DRIVER_IRQL_NOT_LESS_OR_EQUAL blue screen. Could you please check if you can find anything? Here's the link: MEMORY.dmp

I also have 2 other dump files of IRQL_NOT_LESS_OR_EQUAL but they didn't have DRIVER infront and if you want to I can post those too.

 

[johnbl]

is your gpu factory overclocked?
this bug could be timing related, verifier flags will expose timing related bugs so it could be artifact of the verifier tool. IE if you turn off verifier you may not have the problem.
(or not, if you turn off verifier you can not debug the problem.) All, i can suggest is reset the BIOS to defaults, maybe disable hardware in BIOS that you are not using)
confirm your AMD Audio Bus Lower Filter (amdkmafd.sys) is the version that came with your graphics drivers:
\SystemRoot\system32\DRIVERS\atikmdag.sys Mon Aug 03 20:28:52 2015
\SystemRoot\system32\DRIVERS\atikmpag.sys Mon Aug 03 18:42:28 2015

i could not read the date and time stamp on the ati audio driver to see if it came with the same build. run cmd.exe as an admin,
then
cd c:\
dir /s amdkmafd.sys

and confirm that the file and date stamp is the same date as the other graphics files.


---------------
on CPU 0 direct x code running doing some cleanup of memory (this generated the bugcheck)

on CPU 1 the memory manager was getting ready to shut the system down.
it had called nt!MiTrimAllSystemPagableMemory+0x286
to unmap all of the pagefile mappings of system memory. (I think this exposed a bug in the graphics driver, I would think they system thought the memory was not being used. when in fact the directx was still using it)

CPU 4 directx was trying to access video DMA buffers.

rest of the CPUs were idle.


I would think that this is going to be a bug related to a graphic driver. both directx calls generated a page fault, the system call on processor 1 nt!MiTrimAllSystemPagableMemory+0x286
should not remove the page mapping if the memory still had a reference count to show that it was still in use.

bug in directx, or the ati driver or a timing issue relating to your graphics card.
Maybe the direct memory access to the graphics took too long to complete.

(boot into BIOS, reset it to defaults and it will reassign DMA channels and provide the new assignments to windows)

Looks like you have a very current version of the graphics driver.

---------
looked at some internal error logs, the only error I found was in
ProductId: USB Mass Storage Device
DeviceDescriptor: VID 1307 ( USBest Technology Inc.) PID 0165 REV 0100, Class: (0)(0) BcdUsb: 0200

it returned a not_supported error code on some power request.

(not a big deal if the system was shutting down)

 

[ErikCJ]

I don't think it's overclocked and I didn't overclock anything so that should not be the problem I guess. I'll run now without verifier and do some tests.
Here is the result of cd:\
dir /s amdkmafd.sys

Volume in drive C has no label.
Volume Serial Number is DA91-1A92

Directory of c:\AMD\AMD-Catalyst-15.7.1-With-DOTNet45-Win8.1-64bit\Packages\Drivers\amdkmafd\WB64A

23.09.2012 01:17 21.160 amdkmafd.sys
1 File(s) 21.160 bytes

Directory of c:\Windows\System32\drivers

23.09.2012 01:17 21.160 amdkmafd.sys
1 File(s) 21.160 bytes

Directory of c:\Windows\System32\DriverStore\FileRepository\amdkmafd.inf_amd64_2b9d5c8009401001

23.09.2012 01:17 21.160 amdkmafd.sys
1 File(s) 21.160 bytes

Total Files Listed:
3 File(s) 63.480 bytes
0 Dir(s) 206.435.950.592 bytes free

It shows date: 23.09.2012 01:17. The driver installer has the same date. I installed graphic drivers yesterday so they are new yes. I also disabled AMD HDA driver like you said before and I'm not getting SYSTEM_SERVICE_EXCEPTION blue screen anymore (I hope I'm not calling the devil right now).
I reset the BIOS too.

I have an USB Storage Device in my front panel and I leave it in computer all the time even when I shut down computer, so maybe that's the case in not_supported bug?

 

[johnbl]

I am thinking the bad link might cause the last two filters from working. This would change timing of the calls and could expose timing problems in the graphics code.

I would reenable the amd high def audio, in the hopes of repairing the link.
there is a command line tool for loading and unloading and displaying mini-filters
it is called fltmc.exe
start cmd.exe as an admin then run
fltmc.exe to see what filters you have loaded.
use fltmc.exe /? and you can see options to load and unload filters.

I would enable the amd filter in device manager, then use the fltmc.exe command to unload it.

I looked at the amd build for 8.1 and it does include the very old 2012 driver.
the windows 10 version contained a new sound driver (30% bigger) and binary from june 2015
some of the versions did not provide the binary at all.

I am not sure how to fix the links in the filter manager list, just guessing that enable of the driver will fix it, but you might have to completely remove the AMD graphics driver and reinstall it again.
I will see if I can find a better way.
-----------------

you have a bad link in your filter driver chain. The GPU graphics card sound support is filter drive is not listed.
I think this is because you have the driver disabled in device manager. I think the bad link might prevent the proper access to the last two filter drivers.

wdfilter
luafv
npsvctrig
BAD link here ( guessing it is where the amdkmafd.sys driver was located and not properly removed because it was disabled)
fileinfo
Wof

-----------
wdfilter=Windows Defender Mini-Filter Driver
luafv=LUA File Virtualization Filter Driver
npsvctrig ="Named pipe service trigger provider"
fileinfo =Microsoft FileInfo Filter Driver (i think it involves prefetching )
Wof= Windows Overlay File System Filter Driver

here is basic info on filter drivers:
https://msdn.microsoft.com/en-us/library/windows/hardware/ff541610(v=vs.85).aspx

I don't think it's overclocked and I didn't overclock anything so that should not be the problem I guess. I'll run now without verifier and do some tests.
Here is the result of cd:\
dir /s amdkmafd.sys

Volume in drive C has no label.
Volume Serial Number is DA91-1A92

Directory of c:\AMD\AMD-Catalyst-15.7.1-With-DOTNet45-Win8.1-64bit\Packages\Drivers\amdkmafd\WB64A

23.09.2012 01:17 21.160 amdkmafd.sys
1 File(s) 21.160 bytes

Directory of c:\Windows\System32\drivers

23.09.2012 01:17 21.160 amdkmafd.sys
1 File(s) 21.160 bytes

Directory of c:\Windows\System32\DriverStore\FileRepository\amdkmafd.inf_amd64_2b9d5c8009401001

23.09.2012 01:17 21.160 amdkmafd.sys
1 File(s) 21.160 bytes

Total Files Listed:
3 File(s) 63.480 bytes
0 Dir(s) 206.435.950.592 bytes free

It shows date: 23.09.2012 01:17. The driver installer has the same date. I installed graphic drivers yesterday so they are new yes. I also disabled AMD HDA driver like you said before and I'm not getting SYSTEM_SERVICE_EXCEPTION blue screen anymore (I hope I'm not calling the devil right now).
I reset the BIOS too.

I have an USB Storage Device in my front panel and I leave it in computer all the time even when I shut down computer, so maybe that's the case in not_supported bug?

 

[ErikCJ]

I got SYSTEM_SERVICE_EXCEPTION in 5-7 minutes after enabling AMD HDA back. I think this won't work. Computer is running fine otherwise as long as I don't start Minecraft and then close it down. When I close Minecraft Launcher it sometimes crashes (if I run it more than half hour) although I tried this only once now that I'm not using verifier. I let computer to run and no problems. I think something is CPU related but I'm not sure. I also got IRQL_NOT_LESS_OR_EQUAL when I tried to start CS GO like 2 weeks ago. My friend told me that it's the Source engine problem and I had to disable Multicore Rendering in-game and then it worked. I then thought alright It got to be game related problem. But then I remembered that I could run the game without disabling anything before the BIOS update. I don't know what's happening anymore. Is this Xeon really so complicated? I know that is processor meant for servers, but for this price I thought it would be nice and I didn't need that graphic interface anyway. But now only problems occur and I hope it's not CPU related, although I think it's the main problem but I'm no expert on this.

 

[ErikCJ]

Wow it seems that after every game I quit, computer either BSOD's or restarts without BSOD. Wierd. What the hell is going on?

 

[ErikCJ]

I think there's something wrong with hardware. Whenever I watch a video or play video games, computer restarts. If I stress test CPU, nothing happens, if I stress test GPU nothing happens. If I benchmark the computer with Heaven | UNIGINE: real-time 3D engine, it restarts. If there is a driver problem then I can't believe it's so hard to repair a computer - If I've would have decided to format the PC, then I would need to do all the fixes again? Seems very wierd to me. I think I'm gonna give it to a computer service, because I'm done with it and don't have nerves anymore or any spare parts that could try out. I'll report back when all works fine.

 

[ErikCJ]

Okay I found the source of the problem. I have RAM of 3 sticks (Kit of 3) and so I was playing around with them and decided to remove the one that was working in single channel. From now on I don't have any problems or blue screens for about a week now. I think it's solved. But my last question is: What was wrong here? Does my motherboard only support dual channel? Or is the processor that is strict with dual channel? I wanna know, because I was going to order new RAM anyway and I want to know what was wrong so I can buy proper RAM.

 

[johnbl]

configuration of the memory and its defaults is handled by programming in the BIOS of the machine. Some BIOS versions are better than others but most can mess up at setting up default values when Mixed memory modules are used. Sometimes you can set them manually, but often the BIOS only lets you set a few of the timing values. There are like 12 or 14 timing values often the BIOS only exposes 4 of them. BIOS updates may change the defaults to get the machines to work better on different memory modules.
I is always a good practice to run memtest to check out your memory system. Generally in a bugcheck with hardware memory timings problem you will get a bugcheck in memory management and a sub error code of 0xc0000005
running memtest on its own boot image eliminates windows and windows drivers as a potential cause of problems, if memtest fails it will be the software in BIOS, the logic of the motherboard or a actual problem with your hardware (RAM). The BIOS also loads microcode to work around known bugs in various CPUs. Basically all CPUs will have a list of known bugs and microcode updated to "fix" or avoid them. The intel cpu chipset also updates a driver that will reload new microcode for your CPU in case you don't update the BIOS or the motherboard vendor fails to put out a BIOS update.

Okay I found the source of the problem. I have RAM of 3 sticks (Kit of 3) and so I was playing around with them and decided to remove the one that was working in single channel. From now on I don't have any problems or blue screens for about a week now. I think it's solved. But my last question is: What was wrong here? Does my motherboard only support dual channel? Or is the processor that is strict with dual channel? I wanna know, because I was going to order new RAM anyway and I want to know what was wrong so I can buy proper RAM.

 

[ErikCJ]

configuration of the memory and its defaults is handled by programming in the BIOS of the machine. Some BIOS versions are better than others but most can mess up at setting up default values when Mixed memory modules are used. Sometimes you can set them manually, but often the BIOS only lets you set a few of the timing values. There are like 12 or 14 timing values often the BIOS only exposes 4 of them. BIOS updates may change the defaults to get the machines to work better on different memory modules.
I is always a good practice to run memtest to check out your memory system. Generally in a bugcheck with hardware memory timings problem you will get a bugcheck in memory management and a sub error code of 0xc0000005
running memtest on its own boot image eliminates windows and windows drivers as a potential cause of problems, if memtest fails it will be the software in BIOS, the logic of the motherboard or a actual problem with your hardware (RAM). The BIOS also loads microcode to work around known bugs in various CPUs. Basically all CPUs will have a list of known bugs and microcode updated to "fix" or avoid them. The intel cpu chipset also updates a driver that will reload new microcode for your CPU in case you don't update the BIOS or the motherboard vendor fails to put out a BIOS update.

Okay I found the source of the problem. I have RAM of 3 sticks (Kit of 3) and so I was playing around with them and decided to remove the one that was working in single channel. From now on I don't have any problems or blue screens for about a week now. I think it's solved. But my last question is: What was wrong here? Does my motherboard only support dual channel? Or is the processor that is strict with dual channel? I wanna know, because I was going to order new RAM anyway and I want to know what was wrong so I can buy proper RAM.

It could be faulty memory yea, but then I must be really lucky to have removed the faulty one. And I didn't mention on previous post that I ran memtest several times for hours and didn't get any errors. So the new RAM that I'm going to buy isn't going to cause any problems as long as it's compatible right? I was thinking about getting 2x8 GB sticks as I lost my trust in kit of 3 for no reason I guess, but for running in dual channel I suppose it's better to have even number of sticks.

 

[ErikCJ]

I'm here to say thanks for all your help, I'm very grateful and computer is now working properly. You did help me with many problems and some I discovered myself by coincidence. I'll tag solved for one problem, but I can't tag them all so I will leave all problems that occurred in a list.

Problems:
- RAM (This was probably main cause for my problems, but it wasn't the only one. I switched it with new and now the computer is working properly),
- AMD HDMI Audio driver (This was a problem that caused SYSTEM_SERVICE_EXCEPTION BSOD. I disabled the driver and it fixed it),
- AC'97 cable but HDA drivers (I think that was one of the problems to because I got some BSOD's when I installed audio drivers on fresh Windows. Changing to HDA cable fixed it),
- SSD firmware update (I had some BSOD's connected with paging so when I installed new firmware, they disappeared),
- BIOS update (I think this problem was somehow connected with first on this list because I was getting IRQL_NOT_LESS_OR _EQUAL blue screens all the time before the update, but when I updated the BIOS, the bluescreens only occurred after computer was doing some intensive work like video games),
- Long POST time (It isn't a problem, but I'll only mention that after removing card reader, the POST time was shorter)

I think these problems were mainly causing instabilities in my computer and I'm glad we fixed it. Thanks again for you help johnbl, I'll tag one of your posts, but I know there's more fixes in this thread and not just one. If anything happens from this point on, I'll open another thread, because it probably won't be related. Case closed.