Nyx_the_Fallen

Hey all, I've been shopping around for options, and I'm wondering if anyone here has experience in my situation. I work in IT, so WFH is pretty much the norm at this point, and probably will be for the foreseeable future. My home PC is a custom-built gaming/data processing machine, which serves me well. I have a dual-monitor setup, with my main monitor being a curved 27" 2k in landscape, and my secondary being a 27" 1080p crappy monitor in portrait. I may invest in an ultrawide at some point... we'll see. My work computer is a laptop with a Thunderbolt port, and my personal laptop also has one. I am also limited by the fact that my monitors only have one DisplayPort port, so if I want to use DisplayPort (for my 144hz refresh rate and FreeSync), I have to switch the monitor cable as well. Problem is, I have to switch (at least) once per day between my desktop and my laptop. I am... very... tired of switching my mouse/keyboard/monitor cables between computers. Here's the workflow I would prefer: Laptops (either one, not both at the same time, obviously) connect to a single USB-C cable which connects to my keyboard, mouse, and monitors. My desktop connects to the same. Whatever solution I find has to support 144hz 2k connections from my desktop. I'm thinking maybe a switching station combined with a Thunderbolt hub could work? But I'm having trouble finding reliable reviews/products that fit the bill. I appreciate the help!

That was actually my first thought when this all started happening. Originally, I had some G.Skil Ripjaws memory that was DEFINITELY on the QVL list. I tried running it with stock frequency, XMP, and manual overclocks. I've since tried two other sets of RAM, both on the QVL list, both from different manufacturers, and at all sorts of different frequencies. No dice. Same story for the SSD--I've tried two different brands, tested the absolute bananas out of them, and still no issues. I can conclusively rule out everything except for the motherboard and the processor. I've replaced the motherboard twice, and Gigabyte has sworn sideways both times that there's nothing wrong with it. I'm inclined to believe them, as having two different motherboards with the exact same super-fringe-case issue seems highly unlikely.

Does "stable for two months with no BSODs" count? I've had the processor in an Asus board for that long . Believe me--if it's a potential fix, I've tried it. Voltages on RAM and CPU, every test known to man, etc. I have literally replaced every component in this computer (aside from the processor). It's not a coincidence that it is perfectly stable in every motherboard but fails within minutes in two different, brand-new Gigabyte ones. Watch the video if you haven't--I really can't think of another explanation.

It was a set of like... ten or so BSOD codes. And I had tried the processor and RAM in other boards and they worked fine. So literally everything pointed away from CPU problems. I also absolutely hammered the RAM overnight with memtest86, and it was flawless.

The title's mostly a joke. But... if you want it, I'll send it. Here begins a tale of wasted time... I have spent the last three(!) months trying to fix my computer. Today, I have finally discovered what the issue was. Let's start from the beginning... Way back in August of last year, I built a new rig. Ryzen 7 3700x, Gigabyte x570 Ultra, Sapphire Nitro+ 5700XT, and other goodies. Basically, it was an upper-mid-tier rig that I didn't plan on upgrading for quite some time. Every once and a while, I'd get a blue screen. I figured it was BIOS issues with my RAM, and it was infrequent, and it was never while playing games, so I wasn't too worried. But it became more frequent. By February, I could last around 10 minutes--except when gaming! It would only crash under minimal load. After taking it to Microcenter, they were sure it was the graphics card. I RMA'ed it. Best experience I have ever had with an RMA: the tech called me the day it got there, told me there was nothing wrong with it, but sent a new one anyway, "just in case". Barely over a week from send to receive. Unfortunately, he was right--the graphics card wasn't the issue. This left the motherboard and the CPU (I had ruled out everything else--but I won't bore you with the details). I decided to go to Microcenter and pick up a cheap B450 board just to gut-check myself. It ran like a champ--my issue was definitely the motherboard. I sent it in for RMA with Gigabyte. Ohhhh LORDIE, that was the worst. It took them over a month from the time they received it to send it back--and they didn't even do any repairs! It was still broken. A month later, after many two-hour holds with Gigabyte's customer service, I finally got a refurbished model in-hand. We are now approximately three months from February, when I took the computer in to Microcenter. The same issue--it was like December: no bluescreens during high load, but under low load, it was only a matter of minutes before I would see PAGE_FAULT_IN_NONPAGED_AREA, SYSTEM_SERVICE_EXCEPTION, or something else. I was going insane. I spent days doing research... and I found it. My processor specifically doesn't work with Gigabyte boards. I thought, at first, it couldn't be true. But then I tried it in a Gigabyte B450. Same issue. Tried it in a MSI x570. Perfect performance. It's Gigabyte. And then... I found this: a video describing my exact issue--same error codes and everything! (To be clear, my CPU is not an engineering sample.) This is the only mention of this issue I can find... anywhere. I have no idea why it occurs. I have tried literally everything I can think of to fix it... and no dice. What should I do?

Hey all, I bought the Gigabyte x570 AORUS ULTRA this past August, and I figured I would share my experience with all of you, just in case you're looking at motherboards that are guaranteed to be compatible with the next generation of AMD CPUs. First, the good: The board looks and feels high-quality. It's thicc, black, and most pieces you'd end up doing anything with are metal. The VRMs and their heatsinks are BEEFY, and I never even approached having power delivery issues on the CPU. RAM support is pretty good. The QVL list has grown a ton since launch, and I've had luck overclocking non-QVL RAM. I have heard whispers online about some QVL ram not working on XMP, but I haven't run into this issue. Onboard WiFi and Bluetooth is flawless... almost. I was going to run an Ethernet cord to my new office for gaming, but honestly, the WiFi is so good that I didn't bother. However, I kind of hate their antenna. There's really no good way to hide it. No issues with audio. PCIe slots are awesome--they're metal-reinforced, and had no problems supporting my 3-fan giant of a GPU. This is a weird thing to like, but the built in IO shield is pretty awesome. As is the IO itself. All of that makes this board look pretty compelling. On to the bad... Mine broke. I'm not sure what happened, but it just started randomly bluescreening constantly. I can't speak to their overall durability, though, because I am a sample size of one. This is the worst RMA experience I have ever had, hands-down. Originally, I thought my issues were my graphics card. I sent it in to Sapphire. They called me the day they received it, said they couldn't find any issues, and shipped me a refurb the next day. Total RMA time was under a week. Gigabyte? Today is the mensiversary of starting the process. The card arrived at their repair center on the 28th of last month, early in the day. They didn't acknowledge receipt until late the next day. And since then, in the intervening 21 days? No contact, whatsoever. It's like talking to a brick wall. In summary, if you get a working model and it stays working, this is a great product. But... watch out. If anything goes wrong, you're done for. Perhaps Gigabyte will redeem themselves somehow, but at this point, I will never purchase another of their products unless I can get MicroCenter's protection plan for it.

Yeah, I'm pretty frustrated... From what I've read, IRQL errors could mean... Just about anything. And ntoskrnl.exe is just as unhelpful. Like, lol, thank you, it's part of the kernel process. Appreciated.

Aha. Got it to crash again. Most of these are IRQL_NOT_LESS_OR_EQUAL or DRIVER_IRQL_NOT_LESS_OR_EQUAL, and whatever issue it is is hiding behind ntoskrnl.exe.

Yeah, I linked .zip files that include all of my minidumps in the OP. Those have both the error code, string, and driver. It seems to be most often related to ntoskrnl.exe, or a driver (.sys file) associated with ntoskrnl.exe.

Correct. As far as I can tell, RAM is not the issue. However, I also do not have another set to pop in and test.

No misunderstanding--I had my RAM at its built in XMP profile of at 3600. I disabled that and set it all to factory default. I ran Memtest86 on both settings. Clean bill of health.

No overclocks, except RAM. I've tried running the PC at stock settings for RAM and at its built in XMP profile. Same results--nada.

Yes to all four!

EDITS: Added info on system. I'm having very strange issues with BSODs. Sometimes, I'll boot and it will take three tries to even log in. Other times, I'll boot, have no issues for a couple of hours, and then BSOD. Interestingly, I've also seen it get into a "long" boot loop: When idle, it will BSOD, restart, be fine, and then BSOD again every 40 minutes or so. Files that might be useful: Minidump Report 1 Minidump Report 2 Perfmon Report I've done the following: BIOS is up-to-date. XMP is on. Run sfc/ scannow and several DISM repairs. My initial sfc/ scannow caught some corruption and fixed it. Ran Memtest86--clean bill of health. Completely reinstalled Windows (this seemed to help for a while...) Updated all drivers to the most recent version. Ran Driver Verifier (no issues, actually had BSODs while running it that weren't caused by it) Reseated all hardware except the CPU (I don't have thermal paste on hand) Thoughts: Could it be the PSU? If so, how would I check? How can I try to figure out what these issues are? My specs are: Windows 10 Pro Aorus Ultra x570 AMD Ryzen 7 3700X 2x8 GB Crucial BALLISTIX 3600 GHz Sapphire Nitro+ 5700 XT WD Black 1TB NVMe SSD Cooler Master Gold 750 Watt Fully Modular

Here's another set of minidump reports. I'm so frustrated.

See update above...

I would think broken hardware as well... But I can't prove it. And I don't know what would be broken? BIOS is up-to-date... Chipset drivers are also. Is there a way to print all driver/firmware info so that I can include it in the post?

Hey guys, UPDATE: I strongly expect this has something to do with AMD's flaming garbage heap (Adrenaline) I completely uninstalled it and let Windows search for a driver for the graphics card... Haven't had any issues yet. However... I would like to have FreeSync along with the other features I paid for... Any suggestions? Link 1 and Link 2 to .zip files of dumps and perfmon report. I am at my wit's end. I've built computers before, but I have never had rampant BSOD issues like this. We're talking every 15-30 minutes... or not for three hours. Here's what I've tried so far: Reseating all hardware Updating all drivers Running Driver Verifier (I got so frustrated with this that I completely reinstalled Windows) This actually helped for a day or so (only two BSODs in ~6 hours), but as of today, near complete instability is back. I updated all drivers after the reinstallation Running Driver Verifier (again) -- no drivers whatsoever are causing issues now. I've actually gotten BSODs while running Driver Verifier... that weren't triggered by Driver Verifier. Running Memtest86 (before and after the Windows reinstall) -- 4 passes, all 13 tests, no errors. Help! I have... just no idea. Here are my system specs: AMD 3700X TRIXX Sapphire RX 5700 XT (8GB) Aorus Ultra X570 mobo 16GB (2x8) 3600 GHz Crucial Ballistix RAM Be Quiet! big boi cooler Cooler Master MWE Gold 750-watt power supply WD Black 1TB SSD Let me know if there are any other details I should include... I think I've covered all of the requirements, but I'm happy to run any other tests necessary.