Solved Different BSODs in ntoskrnl.exe, (Codes: 0x3b - 0xc5 - 0x124)

[Arc]

You remove the CMOS battery and re-seat it.

If still it shows abnormal voltages, IMHO you need to apply any other PSU, at least as a test.

If all the pins on the motherboard are occupied for ATX12V and there are extra pins on the connector, that is not any issue.

 

[Slyve]

Alright, i took out the CMOS Battery and put it back it, carefully. It is 3v.

When i rebooted, it showed a CMOS Checksum error, which i assume is normal when you remove the Battery. It only gave me the option to restore the "last known working configuration" or something similar, the screen was only there for a couple of seconds before it continued automatically.

The Voltages in Speccy and PC Wizard are still as weird as before. Only difference is that my CPU is now only showing 2 cores, which is weird.

Edit: Forget about the cores. All 8 are there after a reboot.

I'll probably keep a lookout for a new PSU, then. At the very least a PSU should be Cheaper than a GPU.

 

[Slyve]

If all the pins on the motherboard are occupied for ATX12V and there are extra pins on the connector, that is not any issue.

It's actually the other way around. I have 8 12v connectors on the MB and only 4 on the PSU, so 4 are left empty on the MB.

The Manual shows this, so i think it should be fine, but I don't know for sure. It's a little unclear what the manual says to this. At least to me. But since the System is booting, i assume it's fine, right?

 

[Arc]

If all the pins on the motherboard are occupied for ATX12V and there are extra pins on the connector, that is not any issue.

It's actually the other way around. I have 8 12v connectors on the MB and only 4 on the PSU, so 4 are left empty on the MB.

The Manual shows this, so i think it should be fine, but I don't know for sure. It's a little unclear what the manual says to this. At least to me. But since the System is booting, i assume it's fine, right?

The PSUs provide the maximum so that it can be fit on any board. But the board are different, depending on the form factor and some other specifications, they may have less than the maximum.

At your present condition, this is not the point to be worried of. It is OK and fine.

But the PSU itself may be.

We already have Layback Bear with us, he knows hardware and its timings and voltages much much more than that I do. Please wait for his suggestion as well.

 

[Slyve]

So, after i had the PC opened up the last time, it has only crashed once, since then, it seems to be stable. I'm still within the period where it could get back to crashing at any time, but i spent the last couple of days playing GTA V, which taxed my PC way more than other games i usually play.

When i had it open, i removed the CMOS Battery and all Power Cables (except those that connect the DVD drive and the hard drives) cleaned what little dust there was (i had cleaned it recently, so there was barely any) and switched 2 of my Case Fans for newer ones i recently bought. I also removed the graphics card, because the CMOS Battery is directly under it's fan. Then i carefully put everything back together and booted it up. Crashed a couple of hours later one time.

Next up i reinstalled Open Hardware Monitor, which i hadn't used since i reinstalled my OS. Found out that it has this neat little function where it logs pretty much every sensor data there is (Voltages, Bus Speeds, Load and so on) in real time, which i thought would be perfect to show you guys (since i can't do anything with those Infos on my own) what exactly is happening at the moment of a crash.

Since i started logging, it hasn't crashed once. Personally i assume out of pure spite at this point, but i am willing to consider that maybe i fixed some loose cable or something while the PC was open. Don't know why it crashed once, but if the error doesn't come back, i really don't care.

I still have all the Log Files it created the last couple of days (i never stopped it from logging since it's running all the time anyway - i use it to show me CPU and GPU Temperatures via a windows gadget)

It creates the logs as .csv files, which i assume are excel files which i can't open since i don't own word. OpenOffice screws the format up so all the data is bunched together and virtually indecipherable. Tried another free spreadsheet tool but it can only open the files when there is just a little bit of info in them. After a couple of hours the files tend to get very big (it logs 1 line per second) and at a certain size, the tool can't seem to handle the files anymore.

Since it appears that the error is gone for now, i will try to enjoy the stable times ahead and hope it won't come back, because I'm getting really tired of expecting a BSOD at any moment.

Again, thank you all very much for your time!

 

[Arc]

Very well documented, very nice post. And I am happy to know that it is stable now.

But BSODs are unpredictable. They may come back again one day. And I am sure now that this issue will not be able to fool you anymore. For any assistance if needed, we are here.

 

[Slyve]

Very well documented, very nice post. And I am happy to know that it is stable now.

But BSODs are unpredictable. They may come back again one day. And I am sure now that this issue will not be able to fool you anymore. For any assistance if needed, we are here.

Not even 24 hours after i dared to think "Maybe it won't crash again" it crashed again. This means that my PC just straight up hates me. There is no technical defect, this is just a personal vendetta. I don't even know why, i always treated it nicely. Never pushed it too hard, always made sure it's clean and comfortable...

But on a more serious note, i think i will try to switch the PSU, if i can find one. Don't want to buy a new one until i know it actually is the PSU.

I'll also attach the log file of Open Hardware Monitor. Maybe the Data in there can help a little.

The File is formatted very weird. I got it to finally show correctly by telling OpenOffice to separate it by commas and not with spaces, otherwise it gets screwed up and you can't see anything. The Problem here is probably that i have no experience with spreadsheets of any kind. (See the attached Screenshot.)

If you or Layback Bear can find anything in that File that could help, I'd be very grateful. If there isn't anything in there, I'll just hope i can find a PSU to test it. Otherwise i will try the other GPU and work my way forward from there.

I'll also attach the dump file from today, better to have it and not need it than to need it and not have it.

 

[Arc]

Can you please confirm the motherboard's revision version? It would be found here as the image:

 

[Slyve]

Can you please confirm the motherboard's revision version? It would be found here as the image:

I checked on the Box, it's the Rev. 5.0, so i used the correct drivers from the Homepage, although i currently have the more recent ones directly from AMD installed. The Crashes happened with both driver versions, though.

 

[Arc]

Nice info.

As you stated in post #33 .... you have done almost everything possible.

Now, is there any other hardware or devices attached there with the computer? Any USB or PCI devices? If so, disconnect them. See how it goes.

 

[Slyve]

Nice info.

As you stated in post #33 .... you have done almost everything possible.

Now, is there any other hardware or devices attached there with the computer? Any USB or PCI devices? If so, disconnect them. See how it goes.

The Only USB devices attached an on are my Mouse and Keyboard. I also have an external USB HDD, but that is almost always off, except when i need it. I also don't have any PCI devices installed. So there isn't really anything i could remove.

Come to think of it, my mouse doesn't always work when i boot up the PC. I'd say about 3 out 4 times, only one of 2 lights on the mouse is actually on and it doesn't work, i have to re-attach it for it to work. I bought a USB Extension cable, so i can re-attach if without crawling beneath my desk all the time. Could it be worth a try to get rid of the extension cable? Or maybe the Mouse? I never considered this because i didn't really think they could cause those kinds of crashes.

 

[Arc]

Can you apply any other keyboard/mouse combo, at least as a test?

 

[Slyve]

Can you apply any other keyboard/mouse combo, at least as a test?

I don't have any here, but i think that should be pretty easy to get. Thankfully, they're a lot cheaper than GPUs and PSUs.

I'll ask around to see if i can borrow some. Or I'll just buy new ones. Wouldn't be the worst 30 euro i ever spent! Since the crashes seem to be very rare right now (4 days without one hasn't happen very often), it may take me a while to post back.

 

[Layback Bear]

Sorry for being gone so long. I have been back and forth to hospitals and doctors.

I'm not understanding post # 43.

What I think is being said is your motherboard has a 8 pin cpu connector in the upper left corner and the motherboard. And the harness connector from the power supply only has 4 wires in the matching connector. Is that correct.
I have never use a Gigabyte motherboard but all motherboards I have used had the cpu connector from the power supply for the cpu connection filled with wires.

Example:
If the mother board has a 8 pin cpu connection the harness from the power supply should also have 8 wires in the cpu harness connector.
If a motherboard has a 8 pin cpu connection and you only have a 4 wires in the harness connector you are not supplying enough power to the motherboard.

Check your power supply harness and manual. You might have a 4 + 4 cpu harness and connectors.

 

[Slyve]

Sorry for being gone so long. I have been back and forth to hospitals and doctors.

I'm not understanding post # 43.

What I think is being said is your motherboard has a 8 pin cpu connector in the upper left corner and the motherboard. And the harness connector from the power supply only has 4 wires in the matching connector. Is that correct.
I have never use a Gigabyte motherboard but all motherboards I have used had the cpu connector from the power supply for the cpu connection filled with wires.

Example:
If the mother board has a 8 pin cpu connection the harness from the power supply should also have 8 wires in the cpu harness connector.
If a motherboard has a 8 pin cpu connection and you only have a 4 wires in the harness connector you are not supplying enough power to the motherboard.

Check your power supply harness and manual. You might have a 4 + 4 cpu harness and connectors.

Yeah, that's correct. I posted a Photo of the connector on the MB on page 4 i think. Should be Post number #41.

I'll check the PSUs Manual, if i can find it online (i don't have the manual or the box anymore) but I'm pretty sure there is only this one 4 pin connector. I looked several times at all the other connectors. I'll open the PC up, check the exact model of the PSU and post back. Thanks for the Input!

 

[Slyve]

I assume the problem is solved now.

This is my PSU: Leises Netzteil SYSTEM POWER 7 | 700W be quiet! Silent PSU & Cooling for your PC

As you can see, it has 2 separate 12v connectors. 1 P4 and 1 P8. Guess which genius had the p4 connected to the MB that required a P8? Right, this beacon of human brilliance.

I always thought "Hey, the P4 is already connected, keep a look out for another 4 pin you can connect to there as well". I never even noticed the P8 before just now. It's pretty amazing that the System ran at all.

I cannot even begin to tell you how utterly embarrassed i am. Right now a part of me is hoping that the Crashes continue so the reason for all of this was NOT just me being denser than... i don't know, something really frickin' dense. But i doubt it, this seems to obvious now.

I cannot thank you enough, both of you but especially Layback Bear for making me look at the connector thing again. I was close to just spending about 100 Bucks on a new PSU and just stop eating for a week just to get this problem solved.

I still feel embarrassed as heck. I think i will rename my PC to "Layback Arc" in honor of you two and so future generations can remember never to be as stupid as i was all this time...

 

[Layback Bear]

Thank you for hanging in there. Some times you must go back to basics and go from there.

Let us know how things go.

 

[Slyve]

Well, apparently, the 12v connector thing wasn't the cause after all. Crashes still occur.

I will try to switch the GPU, since i have the GTX 275 still around. Apart from that, i will continue to keep a look out for other spare parts to test it. First i will try to get my hands on a PSU and a CPU, because these 3 things seem to be the most likely candidates. Would actually be great if it is the CPU that's the cause, since that one is still new and still has warranty.

I will let you know if anything changes.

 

[Slyve]

I am now sporting a pretty old GTX 275 with some very nice features, for example: 55°C Idle Temperature! Hooray.

Well, I will keep it in and try to find games that won't necessarily melt the rest of my PC (pretty sure melting is bad) and see if the crashes persist.

I also managed to get my hands on a old Quadcore CPU (I think it's an Athlon II x4, don't know which Model, though), which should work. So if the crashes persist with the GPU, I'll put the other GPU back in and switch out the CPU. If that doesn't solve it, only the PSU or the MB remain.

The best thing that it could be is the CPU or the MB. The CPU still has warranty, so i could get a new one. The MB would be the second best option because i was planning on upgrading it anyway. Although i would probably have waited a couple of months. If it is the GPU i will first hit myself for buying second hand without warranty to save 30 bucks and then I'm gonna look for a replacement.

PSU would be... meh, i don't know, it is pretty old and it may just be old age that is killing the PSU. It's about 5 or 6 years old now. Not sure how old that is for a PSU.

In any case, it feels good to finally try something out. I could have done without the moment where i thought i had solved the problem. Next time i will wait with the celebrations and make sure that i actually got the bad guy before popping the champagne.

 

[Slyve]

Well, i aggressively tried to make my PC crash while playing games as much as i possibly could (i even called in sick on Wednesday, mostly because i really didn't feel like doing anything other than breathing that day, but the PC thing was a nice excuse. But if someone asks, i was dying that day) And no matter how much i play, it won't crash with the older GPU.

I gave the 7870 to a friend and asked him to see if it works for him, and so far, no crash there either. So i am assuming it is some kind of hardware conflict with something else in my PC.

It is still very soon and i will continue to watch out for crashes, but my gut was pointing the finger at the GPU for a while. I had some other BSOD issues, but the system began this whole BSOD torture circus somewhere around the time i got that GPU. Although i switched EVERYTHING except the memory since then, the crashes remained.

To end this, i ordered a GTX 960 (i read that a lot of people having trouble with AMD GPUs in general because the drivers are supposed to be very unstable). So i looked for a replacement from Nvidia which is comparable to the 7870 and costs around 200 bucks. The GTX is even quite a bit better, according to what i read about it. I paid 208 Euro for one from Gainward. This time directly from Amazon, because i won't make the same mistake twice.

Again, thank you guys for all the help. I'm rather confident that i found the bad guy this time. Although i will keep an eye out. (not that BSODs are easy to miss, anyway...)

Usually a crash occurs mere minutes after i made a new post saying i fixed the problem, but i hope this won't remain true this time. Wish me luck!