[SOLVED] New PC dead after 4 weeks - heartbroken...

[TheRealMacDaddy]

Hi all,

Firstly, I've tried for so long to fix this I'm literally willing to pay whoever can help me fix this. As I'm sure you can appreciate, in the middle of lockdown, to have an issue like this is pretty troubling.

I’m in desperate need of some advice. I’ve tried for a week to fix this issue, I’ve posted on all kinds of forums, websites, I’ve called professionals, everything.

I also made a thread on reddit about a week ago with some further in-depth information here: https://www.reddit.com/r/buildapc/comments/gt0ioj/new_build_pc_died_after_4_weeks_please_help/?utm_source=share&utm_medium=ios_app&utm_name=iossmf

Long story short is:

Built a new PC about 4 weeks ago. Computer worked fine for about 4 weeks, but would crash every few days, restart itself, and work as normal.

Last Thursday, the computer crashed, restarted itself, crashed, restarted itself, and since then, it hasn’t worked properly and has been boot looping.

After a week of troubleshooting all kinds of methods, including removing RAM, removing SSD, HDD, switching RAM slots, reseating GPU, everything, nothing worked.

I had some artifacting on some occasions when trying to boot the PC when my SSD was removed, and thought my GPU may be the issue. I contacted CCLonline (who I bought the GPU from), who refused a replacement/refund (despite being within 30 day warranty) and quoted a repair only with up to 45 working day (8-9 weeks) total wait, for a GPU I had bought 30 days ago...

I thought the issue may have been that I installed the incorrect drivers, and a few days ago, I managed to boot the computer into windows and had a manic rush to install DDU, and remove my drivers.

After removing the drivers, the boot looping instantly stopped and the computer worked fine. I reinstalled the (correct) drivers, and the ran Cinebench with no issues and used the computer for a couple of hours with no errors, nothing in event planner, no BSoD. I even played highly intensive games all evening yesterday and frequently checked my temperatures whilst gaming all night with no occasions where the temperatures were too high.

Today, I turned the computer on again, and yet again it crashed on the windows desktop and boot looped. I used it playing games for hours yesterday with literally no issues at all. I turned it on today, got to the windows screen and crashed. How can it possibly work flawlessly all day yesterday, just to completely die again today?

I’m at a total loss at this point...the only BSoD I’m getting at present is SYSTEM_THREAD_EXCEPTION_NOT_HANDLED

I just tried completely wiping the computer, uninstalling windows and all files, but the PC crashed yet again whilst in the Cortana section where you select if you want to track diagnostics/connect to WiFi etc.

Can anyone advise what I can do here? It’s my first build and I saved for months to afford this computer, only for it to completely die after 4 weeks. CCLonline want up to 45 working days to replace the GPU which means I’ll be unable to use the computer (I don’t have integrated graphics on my ryzen 5 3600).

I’m completely lost 🙁

I’m more than happy to pay anyone who can help diagnose it, or can direct me to someone who can offer professional help. I just want to be able to use my PC! 🙁

 

[Cere]

So my first thought is overclocking. What are your specs? The motherboard may be overclocking even though you aren't physicaling doing so yourself.

My next thought could be a cooling issue. What are temperatures like in your system? Are your fans and cooling system running like they should? is the thermal paste on the CPU covering the CPU good but not to thick?

There are a number of things that can be causing this. But your specs would help to see if its an issue with something in the BIOS.

 

[Deleted member 2720853]

As the user above said please include all of your specs including the make and model of your PSU.

 

[TheRealMacDaddy]

So my first thought is overclocking. What are your specs? The motherboard may be overclocking even though you aren't physicaling doing so yourself.

My next thought could be a cooling issue. What are temperatures like in your system? Are your fans and cooling system running like they should? is the thermal paste on the CPU covering the CPU good but not to thick?

There are a number of things that can be causing this. But your specs would help to see if its an issue with something in the BIOS.

Hi - thanks so much for replying, I really appreciate it.

Specs:

CPU: Ryzen 5 3600

GPU: Gigabyte 5700 XT 8GB OC

RAM: 2x Corsair 8GB 3200Mhz

PSU: Corsair 650W TX650M

MOBO: B450 Tomahawk Max

OS: Windows 10

BIOS: Not sure exact name, but newest B450 Tomahawk Max BIOS

Temperatures are fine as far as I could see. I originally thought temperatures were causing my periodic crashes every few days, but I bought a new fan (cool master black edition) and still encountered crashes every few days. The case has good airflow with 3 case fans too. There's a small chance the thermal paste could be too thick, but I was still getting these crashes whilst using the stock cooler with stock thermal paste.

The system is at default clock. I've tried undervolting and also under clocking the RAM, neither have seemed to make a difference.

Just as an added comment:

A couple of days ago I managed to boot into Windows randomly, and I tried to get into google to install DDU. The PC crashed whilst trying to install it, rebooted, and allowed me to get straight into Windows again. I then uninstalled the drivers and had no issues for the entire next day. I reinstalled the newest drivers, from the correct website, and played games for most of yesterday. I turned the PC on/off several times, ran Cinebench, Apex Legends, browsed google etc. No problems all day.

Turned it on earlier today, played for a bit, no problem. Turned it off, came home hours later, and boom - doesn't boot.

 

[Deleted member 2720853]

5700 XT huh? Try the following:

First off, go into Command Prompt and write the following command to disable the bootlooping in case something like this occurs once again:

bcdedit /set {current} recoveryenabled No

After that, try going into your BIOS and selecting the PCIe Slot Configuration to Gen 3 (3.0). For a MSI board it should be in PCI Subsystem Settings or something like that.

Try not having any monitoring software run in the background, that includes Afterburner, HWMonitor, GPU-Z etc.

Try setting the Windows TDR value (value which tells Windows when to restart the card in case it becomes unresponsive) to give the GPU a bit more headroom for work:

1. Prses the Win Key + R, write regedit and press OK.
2. Go to HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\GraphicsDrivers
3. Set TdrDelay to 8 instead of whatever value it currently is on.

If none of these work, let me know and we can try something else.

You also mentioned you have Corsair memory? Is that Corsair Vengeance LPX? Several people have reported problems with their builds when pairing such memory with Ryzens.

 

[madmatt30]

And again, all specs needed, including both your hard drives.
You clearly have 2 at least because from thatbl reddit thread you disconnected the ssd and still got a corrupted Windows boot screen - unrelated but if thats what happpened your windows install and boot sectors are split across multiple drives which is never a great idea.

Your system isn't completely dead at all, clearly you can get into bios ok??

Just a hunch but I personally feel this is ram related.

 

[TheRealMacDaddy]

5700 XT huh? Try the following:

First off, go into Command Prompt and write the following command to disable the bootlooping in case something like this occurs once again:

bcdedit /set {current} recoveryenabled No

After that, try going into your BIOS and selecting the PCIe Slot Configuration to Gen 3 (3.0). For a MSI board it should be in PCI Subsystem Settings or something like that.

Try not having any monitoring software run in the background, that includes Afterburner, HWMonitor, GPU-Z etc.

Try setting the Windows TDR value (value which tells Windows when to restart the card in case it becomes unresponsive) to give the GPU a bit more headroom for work:

1. Prses the Win Key + R, write regedit and press OK.
2. Go to HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\GraphicsDrivers
3. Set TdrDelay to 8 instead of whatever value it currently is on.

If none of these work, let me know and we can try something else.

You also mentioned you have Corsair memory? Is that Corsair Vengeance LPX? Several people have reported problems with their builds when pairing such memory with Ryzens.

I'll give this a try in a little bit. Can I change the TDR value without getting into Windows?

My RAM is this: Corsair CMK16GX4M2B3200C16 Vengeance LPX 16 GB (2 x 8 GB) DDR4 3200 MHz C16 XMP 2.0 High Performance Desktop Memory Kit, Black

 

[bignastyid]

Please don't create multiple threads for the same issue.
This thread is closed.
Thread that is remaining open
https://forums.tomshardware.com/threads/new-pc-died-after-4-weeks.3613481/