Frozen Screen! A non Windows related, long term problem...

cemster

n00b
Joined
Jul 31, 2012
Messages
50
My System

Asus P9X79 WS
Intel i7-3930K
CORSAIR Vengeance (8x8GB) DDR3 1600
Zotac GTX 680 4GB (which feeds Three Monitors)
and now Asus DUAL-GTX1660TI-O6G-EVO

I've have been using this set up more than 7 years.

I've always had the latest BIOS.


Hello,

This problem was there for over two years... It was annoying but occurred between long periods of time so I was just complaining once in a while and forgetting it the rest of the time.

All of a sudden picture would freeze in my three monitors and whatever was on the screen would stay there until the system reboot.
And The Q code would show "B2" Towards the end, the Q code would never pass the "B2" state and it wouldn't even make to the BIOS.

At first only one of the two operating systems (with separate drives) was suffering from the prıblem. So I was thinking maybe this is a software issue.
But maybe almost a year later my other OS was frozen too. The major difference between these two Win 10 x64 operating systems was internet capabilities, my workstation system did not have internet.

I took my huge and heavy WS twice to the professionals, once for 4 hours, and the last time for 3 days and they couldn't be able to repeat the problem...

In the last pro shop session, I was told that none of the suspects were faulty. Memories were checked OK.

Even though it has a Memory part in it, the CPU, I was told that couldn't be the problem. Because with a faulty CPU, the picture wouldn't freeze instead there would be no picture at all...

The Zotac GTX 680 4GB apparently survived all the intense 3D tests and all. But afterward, when I told them that I had three monitors hooked up to it, I was pointed out that the "three monitors" in the equation, would have changed the integrity of those tests.

So since those monitors heavily depend on the memory of the Graphic card, and all the things I read on the Asus Rog forums "B2" points towards that direction I was %90 sure that the Zotac was the problem.
Hence my new Asus DUAL-GTX1660TI-O6G-EVO...

So I installed the new card, had issues with the driver, it wouldn't install and I had only one monitor registering! But I was at least at ease because nothing was freezing... That ease only lasted a couple of hours. The picture froze and this time "b1" code was shown as the machine stuck in that state...

So as a next attempt I took my memory sticks out of the board. And tried them first one stick (8GB) then two, up to six sticks; 48GB. And it worked without any freeze. Except that I have to install a Windows 10 post-2016!!! This lovely Ausu card is not compatible with any Windows 10 version released prior to 2017... Of course, I learned that the hard way since there was no indication of that on the product page...

So, okay it's all good. I bought a new GPU card for nothing but okay. Nope, not okay! Maybe 40 to 45 days later my single monitor was frozen...

I attempt to reboot 4 times after that initial freeze.
And each time it failed I got a different Q Code.

"B8" was the first one after the first freeze. I rebooted it and it lasted 2 minutes before it was frozen again. This time the Q Code was "01" then it never booted again but I got these lovely Q Codes: "5A" and "E7"

I left it alone. The next day it worked 5 hours with no glitch. Today it has been an hour so far so good. I'd be happy to finish this before sht hits the fan...

I am assuming at this point my Mobo must be very tired and dying. It has weirdness to BIOS as well. just little things.

I don't know but I am so tired and screwed. I cant totally upgrade my machine. And even if my very uneducated hunch comes toı be true then I have no idea how am I gonna find a suitable MOBO. I doN't one those Chinese X99 cards, and I don't even know if those are compatible with my Ram Cpu and GPU either...

So I would love to hear from some of you to guide me.
I know it has been a lengthy post so thank you for reading the whole thing. I'll appreciate your input.


All my Rams GPU and CPU tested. Ram has been tested with some heavy-duty test programs without booting into the system.
My CPU's thermal paste has been renewed.
My temperature control is very unorthodox yet very powerful. I have very low temperatures in my case it is a huge case. and I have 10 HDDs in the case so I care about the airflow and the heat.
I have a Corsair AX 1200i and it is working perfectly.

cheers
 
Last edited:
Can be hard to diagnose certain issues, might be temperature related, could be that your PSU is starting to fail going by some of the symptoms you describe.
 
Thank you,

I knew I'd left out some things...

My temperature control is very unorthodox yet very powerful. I have very low temperatures in my case it is a huge case. and I have 10 HDDs in the case so I care about the airflow and the heat.
My CPU's thermal paste has been renewed.

I have a Corsair AX 1200i and it is working perfectly.
I have one symptom; frozen picture, no mouse/keyboard and I can't even shot the PC down with the power switch for 'Hard Shutdown...'

I have a bunch of Q codes and they don't seem very accurate to me. And since I turned the PC on this last episode it has been almost four hours without that frozen picture symptom.

Also, correct me if I'm wrong: If there was a PSU related issue I wouldn't have the frozen picture issue, and those specific Q codes, I think...
 
Also, correct me if I'm wrong: If there was a PSU related issue I wouldn't have the frozen picture issue, and those specific Q codes, I think...

I was thinking that the PSU at a certain point fails to deliver enough power to the GPU and then it freezes. Maybe try a different power cable in a different spot on the PSU, and have you tried to use the video card in one the other PCIe slots to rule a fault out there?
 
I already swapped Zotac with Asus..... And in the process, I've changed the port because it required a different output
 
Since you only see the hard lockup rarely, this will be exceptionally hard to diagnose because there is no "easy trigger" for the problem that you can replicate for testing.

I have one symptom; frozen picture, no mouse/keyboard and I can't even shot the PC down with the power switch for 'Hard Shutdown...'
That's a solid lockup which definitely indicates a hardware problem.

There was some screwy messes with PCI-E 2.0 vs PCI-E 3.0 during the X79 chipset era as that is when things were changing. Your CPU, for example, only supports PCI-E 2.0, but the video card supports PCI-E 3.0. Of course the video card is backward compatible to PCI-E 2.0. The P9X79 WS is "PCI-E 3. 0 Ready" because Intel didn't have PCI-E 3.0 CPUs at the time the board was introduced. Anyway, I would check the BIOS to see if the GPU slot is set to PCI-E 2.0 or 3.0. I'm guessing it should already be set to PCI-E 2.0, but double-check. If it's 3.0, try changing it to 2.0. The P9X79WS is a good board and I'd not expect it to give you trouble even with it set to PCI-E 3.0 but given that the 3930K isn't PCI-E 3.0 capable it's worth a look. I don't know what else to check. It's just board, CPU, memory and video card, right? No USB cards, LAN cards, sound cards, etc.? It doesn't seem like a PSU issue to me.

The reason why I'm bringing up the PCI-E setting is because I'm pretty sure that in my previous main rig which was using the Asus Rampage IV Formula X79 and the I7-3820 (same Sandy Bridge E as your 3930K) the BIOS was configured for PCI 3.0 even though it wasn't supposed to work. Like you, I kept the BIOS up-to-date and I swear at some point Asus decided to just enable PCI 3.0 on the board even though X79 wasn't "officially" supposed to support it. If I am remembering correctly, programs like GPUz would even show PCI 3.0 x16 link speed during the GPU load test. I had a GTX 690 originally and then changed to SLI 1080s and then SLI 1080Tis. It's possible that the 690 was PCI-E 2.0 and PCI-E- 3.0 didn't happen until I installed the 1080s but I can't remember. Either way, I never had any lockups with any of them so I left things set to AUTO. The system has a Xeon E5-1680v2 @4.2 GHz and a single 1080Ti now and runs at PCI-E 3.0, but that CPU officially supports it because it's Ivy Bridge.
Refer to this post by Nvidia: https://nvidia.custhelp.com/app/answers/detail/a_id/3135/~/geforce-gen3-support-on-x79-platform

It may have nothing to do with your issue, but a hard lockup like you are experiencing points me to some low level signalling lock and the PCI BUS is certainly in that area.
 
i don't know what the B2 error code is on intel but on AMD it's typically memory or GPU related but not in the sense that either are going bad but it's usually a communication error. that being said my guess based those other systems(which were all AMD) that have shown B2 errors is that the board is likely dying. as far as the screens getting stuck it's usually related to a driver crash in the background but because it's failing to recover the drivers the system gets stuck.
 
Since you only see the hard lockup rarely, this will be exceptionally hard to diagnose because there is no "easy trigger" for the problem that you can replicate for testing.
That's a solid lockup which definitely indicates a hardware problem.
Refer to this post by Nvidia: https://nvidia.custhelp.com/app/answers/detail/a_id/3135/~/geforce-gen3-support-on-x79-platform
It may have nothing to do with your issue, but a hard lockup like you are experiencing points me to some low-level signaling lock and the PCI BUS is certainly in that area.

That's very interesting, thank you. Excellent info with links and everything...

So we are saying that even though all these years despite the PCI-E discrepancy, what worked deviously, would stop working., and why not...

There are some cards but I took everything out just one drğive and the base.

I will look into the PCI-E state of the mobo. and get back.
The machine has been working since the last incident. It's just maddening...
Anyways thank you both for inputs.
 
Bump up DRAM, CPU and chipset voltages by a wee bit.

My corsair memory always loved extra voltage.

Not saying this will fix things, but a little voltage at this age won't hurt and *may* help stability.
 
So we are saying that even though all these years despite the PCI-E discrepancy, what worked deviously, would stop working.
Well, it could also be that the motherboard is slowly dying. Signals are getting faults. Aging capacitors. Cracked solder joints. Who knows.
 
I had a very strong feeling about the motherboard and went to the other way. (GPU). But now I am told that CPU could also not be excluded from the usual suspects' list as I initially was told otherwise by a technician. But I feel like the motherboard got really tired and those Q-codes are too random and scattered all over the board. The problem is that there aren't any decent X79 boards. I am okay with X99, but there aren't ant of those too... Like I mentioned I don't want any of those Chinese rebranded X70 boards...

So I don't know. I might try to send it to the service as a last resource... And research a bit more of I don't know what! :/
 
Back
Top