What's new
Carbonite

South Africa's Top Online Tech Classifieds!
Register a free account today to become a member! (No Under 18's)
Home of C.U.D.

Constant BSOD, I've run out of ideas, need tech shop in Cape Town

Shiizzo

Member
Rating - 100%
4   0   0
Joined
Nov 27, 2021
Messages
63
Reaction score
28
Points
1,235
Age
43
I'm getting constant BSOD's every day. I've tried many things, but I can't find the problem. I have done a lot of troubleshooting.
Did memtest86, no issues with RAM.
I removed one module and still crashed. Put in the other module and got more crashes.
Windows is on SSD with +-80% health.
I've used DDU to uninstall graphics driver, still crashes.
PC crashes in Safe Mode with Networking.
I check minidumps with Windbg, whocrashed and Bluescreenview... Nothing jumps out at me (I may be out of my depth).
I'm running driver verifier at the moment, but even that, no idea how to use it.
Done the SFC scan now thing, still crashes.

I don't want to throw parts at the problem, but I'm ok with paying an expert to resolve the problem. Any recommendations for a shop in Cape Town, where you believe they will be skilled enough to figure this one out?

PC:
Asrock B550
AMD Ryzen 3600
32GB G.skill ripjaws
Geforce 3080
Windows 11
 
1. Use HWInfo64 to get some indication of temps
2. If you have a spare drive, do a clean install on that (no other software) and see if it still happens
3. Disable overclock if you have one, even on the RAM (disable the profiles, run at 2400)
 
I removed one module and still crashed.

Have you tried running with just 1 module, instead of removing just 1?
Temps fine on cpu and gpu?
Which psu? How old is it? Are you running individual cables to gpu from psu (eg no single cable splitting into 2x 8 pins)
Did you upgrade form win 10 to 11, or clean install?
 
1. Use HWInfo64 to get some indication of temps
2. If you have a spare drive, do a clean install on that (no other software) and see if it still happens
3. Disable overclock if you have one, even on the RAM (disable the profiles, run at 2400)
Thanks for the reply.

I use hardware monitor currently. Max 64 degrees on CPU, mostly in the 40's. GPU doesn't show up currently because driver isn't installed.

I don't have a spare driver, but it's a great idea so I will get one and try the fresh clean install

I will clock my memory down in the meantime, and see how that goes.
 
Pretty sure Woot offers this, but Carb will inevitably try help, reseat all components 1 by 1,
Clean ram & gpu slots & cpu,
Can also reset BIOS to default, unplug cmos. Reflash bios' if necessary.

Set windows power plan to balanced or low power, see if that changes something.

What do you mean don't have a spare driver @Shiizzo ? get it from nvidia's website?
What BSOD errors are you getting?
 
Thanks for the reply.

I use hardware monitor currently. Max 64 degrees on CPU, mostly in the 40's. GPU doesn't show up currently because driver isn't installed.

I don't have a spare driver, but it's a great idea so I will get one and try the fresh clean install

I will clock my memory down in the meantime, and see how that goes.
Windows should install a basic / recent NVidia driver when connected to the internet and activated.
 
Have you tried running with just 1 module, instead of removing just 1?
Temps fine on cpu and gpu?
Which psu? How old is it? Are you running individual cables to gpu from psu (eg no single cable splitting into 2x 8 pins)
Did you upgrade form win 10 to 11, or clean install?
I am running 16gb RAM currently, so just 1 of the two RAM modules. I've tried both individually, in slot 2.
Temps appear fine, I don't have logs, only real time monitoring.
PSU is "EVGA 220-B5-0750-V3 750 B5 750W 80 Plus Bronze Fully Modular Black Desktop Power Supply", bought November 2021. Indeed I am running a single cable to GPU... I should change this?
I had worse problems on Win10. I updated to Windows 11, which didn't resolve the problem. That was last year, and the problem was faulty RAM, which was replaced/RMA'd. That faulty ram showed errors in memtest86. Current RAM shows no problems.
 
Pretty sure Woot offers this, but Carb will inevitably try help, reseat all components 1 by 1,
Clean ram & gpu slots & cpu,
Can also reset BIOS to default, unplug cmos. Reflash bios' if necessary.

Set windows power plan to balanced or low power, see if that changes something.

What do you mean don't have a spare driver @Shiizzo ? get it from nvidia's website?
What BSOD errors are you getting?
Sorry, spelling mistake. I don't have a spare HDD to do a clean install of Windows.
 
If your win 10 already had issues, a simple upgrade to 11 could bring those along (or make it worse)
I know it's a mission, but a fresh install might help if the hardware all checks out😅

Where in Cape Town are you? Someone might be able to lend you an old platter drive or smoll ssd.
I'm in Tableview. Will scratch around, might still have one somewhere if you are nearby.
 
Where are you in ct? If you close by I can come take a look. Are you crashing inside bios? What error are you getting during the bluescreen? My best bet as it is by far the most common issue in old mobos is dirty dimm slots but then you would also be struggling to get it to boot in the first place. Alternatively what is also common is a faulty SSD/HDD. Everything else is pretty unusual. The fact that youve already had to replace memory leads me to think there may be a problem with your dimm slots.
 
Try getting another kit from a mate or carbie and testing it out.

I had a year of headaches with a mates pc of constant BSOD's
3950x build with x570 board and 3080

Did memtest86.. which for 128gb, you can imagine, took forever.
All dimms passed with flying colours.
Tried each dimm individually and got varying results in stability, but the fact that it all passed memtest86, the ram was ruled out.

We swapped board, cpu, psu, to no avail.
We swapped out the GPU with a GTX 760, and it still kept crashing.
The issue is the crashes were completely random. Sometimes nothing for a week, then 3-8 a day.
We tried pbo on, pbo off, eco mode, resizable bar on and off, xmp on and off, different ram timings, overvolting the dimms to 1.35v at 2166mhz.

All that was left to replace was the RAM.
He bought a kit of Corsair LPX 3600mhz and it hasn't crashed once since.
I currently have his G.Skill 3200mhz c16 128gb kit in my pc to test and have not had a crash in almost 2 weeks.

For the life of me, I really don't know wtf, but it's now working.
The only thing I can think of is the Ram not liking the GPU

Is your 3080 an EVGA FTW3?

I suggest you try get another ram kit and do some testing.
 
Last edited:
SSD doesnt sound good man.

Post a screenshot of BlueScreen info. I often have to diagnose these remotely.

Verifier.exe basically stress tests your drivers as opposed to something like 3DMark / FurMark which stresses the hardware. How long as Verifier.exe been running and what guide did you follow to config verifier?
 
Try getting another kit from a mate or carbie and testing it out.

I had a year of headaches with a mates pc of constant BSOD's
3950x build with x570 board and 3080

Did memtest86.. which for 128gb, you can imagine, took forever.
All dimms passed with flying colours.
Tried each dimm individually and got varying results in stability, but the fact that it all passed memtest86, the ram was ruled out.

We swapped board, cpu, psu, to no avail.
We swapped out the GPU with a GTX 760, and it still kept crashing.
The issue is the crashes were completely random. Sometimes nothing for a week, then 3-8 a day.

All that was left was the RAM.
He bought a kit of Corsair LPX 3600mhz and it hasn't crashed once since.
I currently have his G.Skill 3200mhz c16 128gb kit in my pc to test and have not had a crash in almost 2 weeks.

For the life of me, I really don't know wtf, but it's now working.
The only thing I can think of is the Ram not liking the GPU

Is your 3080 an EVGA FTW3?

I suggest you try get another ram kit and do some testing.
Dont kill me but I think I know the issue :)

Is the problematic RAM certified by his x570 brand to work :)

3950x "doesnt like" 4x RAM slots occupied. Is the new kit 2x or 4x modules? I bet its 2. Did you do a CPU microcode update? Usually happen during BIOS update AFAIR.

More RAM slots occupied = more stress on mem controller, which lives somewhere next to the CPU itself. I've seen a pattern with Intel and AMD's higher end CPU's and where 4x RAM slots are used.

As far as I can tell, the CPU overall isn't stressed enough under Memtest, to make it BSOD with the 4x slots occupied. I could be wrong, just a theory.
 
Thanks for the replies!

I'll try to respond as best I can.

CPU I got Jan 2021.
I'm in Kuilsriver. Appreciate the offers of lending me parts.
I had this ram before: G.Skill F4-3200C16D-16GVKB Ripjaws V Series 16GB (2x8GB) DDR4-3200MHz CL16 1.35V
I upgraded late last year to this: G.Skill F4-3600C16D-32GVKC Ripjaws V 32GB (2x16GB) DDR4-3600MHz CL16 1.35V
That was the faulty set of ram, that was RMA'd. I've run it at 3200mhz since I got the replacements, and was stable for a while but then the BSOD's got more and more.

I never heard of QVL lists until this past weekend, when I checked ASRock's website, and my RAM isn't on there... If it is the RAM, and I also think it may be, I will buy another set of the 3200 that worked well for me before... I gifted my old RAM to my niece, so no getting it back now.

3080 is Gigabyte GeForce RTX 3080 GAMING OC LHR Edition 10G GV-N3080GAMING OC-10GD 10GB GDDR6X 320-bit PCI-E 4.0 Desktop Graphics Card (Lite Hash Rate)
BIOS is currently updated to run ram at 2400, will monitor how well this goes.
Tonight I will get two separate power cables to the 3080.

I'm getting a few different BSOD errors... typically:
UNEXPECTED_KERNEL_MODE_TRAP
PAGE_FAULT_IN_NONPAGED_AREA
IRQL_NOT_LESS_OR_EQUAL
and a Hyper-V error I don't currently have on hand.
 
SSD doesnt sound good man.

Post a screenshot of BlueScreen info. I often have to diagnose these remotely.

Verifier.exe basically stress tests your drivers as opposed to something like 3DMark / FurMark which stresses the hardware. How long as Verifier.exe been running and what guide did you follow to config verifier?
bluescreenview: Clipboard01.jpg

Followed verifier instructions from Microsoft support page. Running since earlier today.
 
bluescreenview: Clipboard01.jpg

Followed verifier instructions from Microsoft support page. Running since earlier today.
PC crashes in Safe Mode with Networking - either SSD , SSD cable if applicable or Windows.

Had a look but didnt see it mentioned, have you reinstalled Windows?
 
bluescreenview: Clipboard01.jpg

Followed verifier instructions from Microsoft support page. Running since earlier today.
I need to see some info in the bottom of BSV, can you select the 5x BSOD's and export them, and upload them to your Gdrive please?

(File - Save selected)
 
Reinstalled Windows after getting replacement RAM in November last year. No reinstall since then.

I will take my PC apart tonight, and plug everything in again (except CPU... don't like messing with that). HDD, SSD's, RAM, GPU
 
Reinstalled Windows after getting replacement RAM in November last year. No reinstall since then.

I will take my PC apart tonight, and plug everything in again (except CPU... don't like messing with that). HDD, SSD's, RAM, GPU
I would not worry about taking it apart, especially if you dont like messing with the CPU.

The software will tell you everything you need to know :) Upload the export from BlueScreenView and I'll give you a good idea of whats going on.
 
Have you done a fresh install of windows?
 

Users who are viewing this thread

Latest posts

Back
Top Bottom