Severe problems after motherboard upgrade

[Mikeybb]

I apologize if this is in the wrong place, but I thought it was pertinent since the motherboard was one of the components upgraded. First, I had a stock Dell motherboard with 8 gb (2x4) of no name RAM, an i5-3340, and a GTX 980. I bought an MSI Z370-A Pro, 16 gb (2x8) of HyperX RAM, and an i7-8700k. After installing the components and their drivers, I noticed that when I started up Watch Dogs 2 to test everything, it would run fine (in game) for about 10-15 seconds and then both of my monitors would go black and I'd have to power the computer off using the power button.

After looking around for a bit, I had read in several threads that you need to refresh (I believe that was the term used) Windows 10 after changing components. I did this and retested the game afterwards and the test went similarly, but... the computer powered itself off and restarted, this time with horizontal lines from the time the monitor received signal. Windows will not boot and I have an error code of DPC Watchdog Failure, or something similar. If I enter my BIOS menu, there are short horizontal white/gray lines all over the screen, but they move when I move my cursor around.

I took the case apart and took another look at everything and discovered that my SATA cables for my HDD had been leveraging the rear of the 980 towards the top of the motherboard. I remembered that the card had been difficult to put in and this is likely why. I looked at the card and can't see anything obvious wrong with it or the PCIE slot on the motherboard, however after putting it all back together and moving those SATA cables, it is doing the same thing.

What happened? And how can I fix it? I've pretty much accepted that I'll likely have to pull the HDD that had windows on it and wipe it to get a good reinstallation, but I'm praying that the graphics card isn't toasted. Please ask me whatever might help you come to a conclusion about this problem I'm having. I'm more than willing to troubleshoot, I just really don't know what to do now...

I appreciate your help,

Mike

PSU is an EVGA NEX650G

 

[COLGeek]

What is the make/model of your power supply?

 

[SkyNetRising]

Manufacturer/model name of PSU?

 

[Mikeybb]

What is the make/model of your power supply?

It's an EVGA NEX650G. As I was shopping on PC Part Picker, it was adding up the power draw and it seemed like that PSU should still be alright.

 

[Mikeybb]

Manufacturer/model name of PSU?

It's an EVGA NEX650G. As I was shopping on PC Part Picker, it was adding up the power draw and it seemed like that PSU should still be alright.

 

[COLGeek]

Bumping is not allowed. Please refrain from doing so in the future.

Power should not be an issue. The PSU is sufficient to power your rig.

Just to confirm, you did not perform a full, clean install of Win 10. Is that correct?

 

[Mikeybb]

Bumping is not allowed. Please refrain from doing so in the future.

Power should not be an issue. The PSU is sufficient to power your rig.

Just to confirm, you did not perform a full, clean install of Win 10. Is that correct?

I apologize for bumping and will definitely refrain from doing it in the future. I have not done a clean install of Windows 10 yet. I did a refresh, or something like that, which gave me the options of keeping my personal files or deleting them. I did that both ways and after deleting the personal files, it shut the computer off while in game. Now, no Windows. Thanks for your reply.

 

[COLGeek]

Is anything overclocked (CPU, GPU, memory)? If so, return to stock speeds for troubleshooting. A stable system is a must before ramping up settings.

Have you tested your memory?

Also, this statement is worth investigating...

I took the case apart and took another look at everything and discovered that my SATA cables for my HDD had been leveraging the rear of the 980 towards the top of the motherboard. I remembered that the card had been difficult to put in and this is likely why. I looked at the card and can't see anything obvious wrong with it or the PCIE slot on the motherboard, however after putting it all back together and moving those SATA cables, it is doing the same thing.

Did you install the proper standoffs to the case to mount the motherboard properly?

 

[Mikeybb]

Is anything overclocked (CPU, GPU, memory)? If so, return to stock speeds for troubleshooting. A stable system is a must before ramping up settings.

Have you tested your memory?

Also, this statement is worth investigating...

I took the case apart and took another look at everything and discovered that my SATA cables for my HDD had been leveraging the rear of the 980 towards the top of the motherboard. I remembered that the card had been difficult to put in and this is likely why. I looked at the card and can't see anything obvious wrong with it or the PCIE slot on the motherboard, however after putting it all back together and moving those SATA cables, it is doing the same thing.

Did you install the proper standoffs to the case to mount the motherboard properly?

I had nothing overclocked, I changed the memory speed and voltage in the bios as it was still set to my old DDR3 RAM settings, I set it according to the info on the RAM packaging. I have not checked the memory, I'll look around to find how to do that without access to Windows.

When you say that the quoted statement is worth investigating, what do you mean in particular? I've thought about moving the GPU to the x8 slot and retrying everything, but wanted to hold off as I really don't know what I'm doing right now and I'd rather have a little more direction. The card was visibly not straight, which I should have noticed when I put it in, but I didn't. Could you explain your line of thinking? My thinking is that if the card was leveraged, it could have promoted bad contact with the pins in the PCIE slot on one side of the card which would have been fine for web browsing, etc. However, when there was a stronger demand on the card, it couldn't satisfy that demand due to contact problems.... I checked the gold tabs on the card and see no evidence of any arcing or any defects.

I have standoffs in all the mounting locations.

 

[COLGeek]

Set the memory to auto in the BIOS. Let the motherboard set it. I was concerned that you may not have used standoffs and was flexing the mobo.

 

[Mikeybb]

Set the memory to auto in the BIOS. Let the motherboard set it. I was concerned that you may not have used standoffs and was flexing the mobo.

I see, the force on the motherboard would have been parallel to the surface of the motherboard, not perpendicular. I was more worried about the PCIE slot due to the direction that the force was applied. Thanks for the input.

 

[Mikeybb]

UPDATE: I got Windows reinstalled and was trying to get the Nvidia driver installed for the GPU. When I tried to install it, the screen would go black with white and grey horizontal lines across it. I was trying to install the driver because I was stuck at a crazy low resolution on my main monitor (should be 2560x1440) that blew everything up to the point that I couldn't install the software that came with the motherboard (wouldn't fit on screen and whenever I pulled the box up to press install, it would snap down immediately). After this happened once or twice, I decided I'd better try to test the ram prior to anything real bad happening. However, this time it did the lines as it was booting windows and wouldn't actually complete the booting process. I switched the GPU to the other card slot and it made no difference. I also tried removing the GPU and got no signal to the monitor without it. In addition, I pulled the motherboard battery for approximately 1 minute and retried with the same results.

At this point, my limited knowledge says it could be the GPU, the RAM, or the board. But, I'm unsure of how to test these things without access to the OS or another machine. I plan on taking the GPU apart tomorrow to check all the electrical components inside it. Also, why would I get no signal to the monitor with the GPU out? Onboard graphics should kick in, right? I appreciate all the advice so far.

UPDATE to the update: I just tried it once more and got the automatic windows repair process and it concluded that it can't fix it... I don't understand what is happening. In addition, later on this evening, I pulled RAM sticks and tried to turn it on with each one removed. Each of them got to the "Your windows is done, hace a nice day" screen. So... unless there's a better way to test RAM without spending a while installing the OS each time, I believe it's good at this point. The Windows that I originally started on was on my HDD, I clean reinstalled it on my SSD. Both were corrupted/didn't work after a few power cycles.

I'm not sure what this all points to... I don't think it would be RAM, from my understanding of what it does. Also, I can't think that two drives (the SSD just barely used) are bad. I keep coming back to the GPU or the mother board from my stupid installation mistake, but I don't understand how that could cause the Windows problem. Does anyone have a troubleshooting hierarchy? I've been following random internet hits, which may be the wrong approach. I'd rather not have to reinstall Windows every few power cycles, if I were to follow the sticky at the top of the forum page.

I upgraded this system to be able to run a program that, in part, finds chimeric sequences in genetic material from environmental sources for my master's thesis. I ran out of RAM capacity and I have a policy of upgrading to something good instead of wasting $75 on more DDR3 RAM when it will become obsolete sooner than DDR4. I hope this wasn't $600 ish bucks down the drain.

Example of what I'm experiencing when I'm trying to boot up sometimes:

I really do appreciate all of the suggestions, as this is definitely not my area. Thanks

 

[Tinstaafl]

To test your RAM without loading Windows, boot the computer with MemTest86. https://www.memtest86.com/

Since your computer powers on and self tests, and you can access the BIOS, I agree that troubleshooting the GPU should be the next step, based on the following:
1. you experienced graphics issues in the BIOS without having Windows running (horizontal white/gray lines all over the screen).
2. you moved the GPU to another slot and that made no difference.
3. you have integrated graphics to test your system with and rule out the GPU.

If you remove the GPU to test the integrated graphics, you will need to switch your monitor cable to a graphics output on the motherboard, rather than the one on the GPU, or else there is no way to get the signal to your monitor. Your motherboard has 1 Displayport, 1 DVI-D, and 1 VGA for outputting the integrated graphics.

 

[Mikeybb]

To test your RAM without loading Windows, boot the computer with MemTest86. https://www.memtest86.com/

Since your computer powers on and self tests, and you can access the BIOS, I agree that troubleshooting the GPU should be the next step, based on the following:
1. you experienced graphics issues in the BIOS without having Windows running (horizontal white/gray lines all over the screen).
2. you moved the GPU to another slot and that made no difference.
3. you have integrated graphics to test your system with and rule out the GPU.

If you remove the GPU to test the integrated graphics, you will need to switch your monitor cable to a graphics output on the motherboard, rather than the one on the GPU, or else there is no way to get the signal to your monitor. Your motherboard has 1 Displayport, 1 DVI-D, and 1 VGA for outputting the integrated graphics.

Thanks for the advice. I'll do the memtest86 boot after I'm done with the GPU. This morning, I stripped the GPU down to the board and can't see anything that looks fried. I'll likely post up on the Graphics Card portion on the forums to get some advice on what to look for. I tried running it on the integrated graphics through the board's DVI-D port and got a whole lot of nothing. My other monitor has Displayport, so I'll be trying that as well. Just ran out of steam last night and was excited to crack open the GPU this morning.

These are the pictures I took of the GPU, the solder doesn't look real great on the back side. It's quite dull and whenever I've soldered things, I've ended up with a shinier finish on the solder.

 

[The Original Ralph]

JUST a suggestion, but when re-installing the Nvidia drivers, you should uninstall the old driver first, in safe mode using DDU

here's the instructions if you want, from the GeForce forum: scroll down in the thread to about the 3rd post
https://forums.geforce.com/default/topic/1035287/nvlddmkm-sys-crashing/?offset=1

and while you're over there, you might ask what the most current stable Nvidia driver is - they seem to feel a number of the Nvidia drivers are buggy - fortunately, all the previous drivers are still available from Nvidia