I
IHaveNotASingleClue
I'm desperate and I am looking for any kind of assistance with this because I am at a complete and total loss. I posted this problem about two weeks ago and could find no solid answer.
This, if anyone wants past information: Soft-Lock/Hanging? Processor Stops Processing But Computer Keeps Running? System "Crash" But Desktop Still On?
TL;DR: Computer "Hangs" randomly: everything will suddenly stop animating/playing/whatever, but mouse can still move, some windows can still close, some button animations still play, but nothing responds, commands do not work, computer will not close of it's own volition even if the Start Menu can be pulled up and clicked through. No artifacts besides hanging windows and screens. Windows look "Frozen" but videos will be showing a constant loading circle. Occasionally still able to use internet while this is happening. Reset button on Tower restarts PC and it is perfectly fine afterwards.
New computer worked fine for a week before it started doing this. Started after running some intensive programs that worked the entire system for the first time. Checks indicate OS is perfectly fine, no corrupted sectors or files.
Specs: Windows 10, Fully Updated
Ryzen 7 3700X
MSI B550 Tomahawk
G.Skill Ripjaws 32 Gig DDR4 RAM
GTX 970, then later a 1070 Ti
SeaSonic PX-850 Power Supply
(I do not believe there are any inconsistencies with this combination of hardware and most of them seem to have good reviews everywhere)
New Information: Alright, so this happened a few times a while ago before I just stopped using my computer after posting on the Microsoft Tech support. Literally the only response I've gotten about this at all was about my video card potentially not working with the latest Nvidia Drivers, so they said to roll back and try an older one. I go back to a driver I was using perfectly fine on my last computer and it seemed to work for a day before doing the hang again.
I decided maybe my Graphics card was finally starting to die. I hesitantly ordered a decently-priced one off Ebay (I was trying to wait for a 3070) to test this theory out, because it made sense it would be going: it's over 6 years old, used heavily for gaming (not abused, but I used it for many hours every single day for pretty much everything), and the computer wasn't "Crashing" in the usual sense, it seemed to only be the interface on screen (No other components appeared to be shutting down during this time), on top of the fact it is a bigger system demanding more of the same card now, so I figured it cant hurt.
Well I got a used 1070 Ti off Ebay, slapped it in there and ran a ton of stress tests on it (Heaven Benchmark, The UserBenchmark little program for fun, my million browser tabs, intensive games, music, etc.) and literally nothing went wrong. No slowdowns, no instability warnings, no critical anythings. Processor got kinda warm but it seems to always do that and it never goes above 75 C. New GPU never went above 65 C.
I tentatively hoped it was fixed, and I went about my life. It worked flawlessly for about four days straight, two of which I was using it heavily the entire day with zero issues and no indication anything at all was amiss.
Last night I used it for a few hours, played some more WoW on it and wound down for the night with an hour or so doing research on a new desk chair and Pinterest stuff (I was also transferring 130 gigs of backup files to a different drive but the CPU wasn't even blinking at this). And then right before I was about to close everything down it did it again. This time it happened more suddenly instead of the more gradual dying of responsiveness the other times had, and very few things responded to clicks: windows did not respond or shutdown when I told it to and menu's did not open up, but the video I had up still tried "loading" endlessly and the graphics for buttons indicated they were being hovered over and pressed. Just nothing was happening.
I quickly (gently) opened up the case and looked in, thinking maybe the Tomahawk I have would tell me something with those little On-Board diagnostic LED's on it. None of them were on, and from the looks of it everything else was fine: fans were spinning, RBG lights I didn't bother setting up were strobing. I do not remember the Processor making it's strange little beeping beeping noises it makes when it's thinking, but maybe I wasn't paying close enough attention at that moment.
Again, only Event 41 and 6008 this time (And the 6008 I surmise only because the Reset button goes through a different method of forced shutdown than the power button holding method so it registers as an "unknown Internal shutdown" or whatever). Literally nothing in the logs leading up to the moment it hangs. No crash logs anywhere. No Dumps. No anything's indicating anything at all.
So now I am a couple hundred dollars less for a graphics card I probably didn't need and I'm back to square nothing.
Suspicions: I dont know. I have literally no idea. I have no leads. Nothing about this makes sense to me and the fact I have found nothing even slightly relevant to it in hours and hours of searching does not help. All the tests, checks and fixes I have attempted have done nothing.
It could be the power supply? Maybe it somehow isn't delivering consistent power and it causes the entire system to just hang in some weird half-dead state? But it has done this both with the computer running full-throttle as well as barely anything at all. Wouldn't it be more consistent if it were power delivery failure? (Besides it's a Seasonic Platinum 850, it better not be bad, on top of the fact it's way overpowered for my current system).
Graphics card? I'm kind of doubting it considering two different cards are doing this now, unless both of them are failing in the exact same way, in which case someone on Ebay is going to be getting a mouthful. Open-Box my butt, it had a years worth of use-dust on it . . .
Motherboard/CPU? I would assume if it was something with these two the crashes would be a lot harder and more consistent, if the computer started up at all. Maybe this is why the Motherboard Diagnostic LED's are not lighting up? The Processor having a problem would make sense in the case that nothing is actually "Processing", or only giving it a half-attempt at doing anything, but as far as I know Processors are kind of an "All Or Nothing" component: either they work or they dont, and anything else is an outside factor.
RAM? It would explain the rather random nature of the crashes, as they had very inconsistent timings with no similar or seemingly recreatable conditions, but if anything at all tried accessing a bad sector of Ram stuff would start going downhill instantly, regardless of overall system stability. But as with the CPU/Motherboard I would assume it would be a lot harder of a crash and thus a fair bit more apparent. My checks on the Ram have turned up nothing, although that doesn't necessarily mean anything since Ram can (rarely) go for years with a problem and never make itself apparent even with extensive testing, if my research has told me anything.
Hard Drives? All of my drives are healthy (at least, according to the software I'm using to monitor them): I have two brand-new M2's and my three other HDD's all passed exhaustive tests I put them under after reformatting them for the new computer. Two of the hanging incidents did happen either during or right after a program was reading/writing a lot of data to drives (the first one during the UserBenchmark program when it did benchmarking sequences, and the second happened right after the 130 GB file transfers between my backup drives), but I dont know of anything similar happening during the other incidents.
Software? The only thing different between my two computers is now I am using Firefox instead of Chrome, and I have a bit more monitoring software (CPUID, Afterburner, Hard Drive software from WD, Seagate and Samsung), but rarely were any of them running when the hangs happened, usually only CPUID and only then after the second incident alerted me to it being an ongoing problem. I do not use Afterburner actively yet, mostly just to monitor FPS and temps and the like. They were also running fine the week before this started happening after I built the computer.
Audio Equipment? The newest addition to my setup is a decent speaker/subwoofer combo that I use via Bluetooth. I've heard Audio drivers crashing can, for some reason, cause catastrophic failures in systems when they go, but I have seen absolutely no indication of failure in this area, and my laptop works fine with the audio setup. Maybe the Bluetooth is causing some weird, obscure problem.
More Specific Musings: Some new ideas I've come up with might help narrow this down
When I get off work I'll probably start the arduous process of lugging my 50lb PC tower out and putting my older Corsair 750 Gold from my last computer into it and seeing if I can rule out the power supply (Since that one has been going for years just fine without issue and I know it works). To me this seems like the most likely cause, as power supply issues can cause the widest range of failures and obscure behaviors, but since this thing seems to apparently like waiting several days after I attempt a fix before throwing a wrench in my mood this could take some time to diagnose.
When I was looking at my computer last night, I noticed that after the hang started and I took the panel off, the Processor didn't make it's occasional little "beep"/"click" noises it usually does every couple seconds. I mean, it's strange to me that a processor make any audible noise at all but this is my first time dealing with an AMD processor of this caliber, so I just chalked it up to AMD weirdness. Or maybe it's something else making that noise and I just haven't figured out what it is. Figured I'd mention it.
I'm also suspicious of my Boot Drive being on a NVMe M2 that is in the M2_2 slot on the motherboard; my other M2 is only a Sata, but the M2_2 slot doesn't accept Sata, only NVMe for . . . god only knows what reason, so I cannot put my Games M2 drive there. MSI seems to be the only one that has this configuration, all other board makers have it the other way around so the NVMe-only slot is the M2_1 and Sata is M2_2. I dont know if this has the potential to cause issues or not, since the M2_2 slot being occupied seems to cause all sorts of strange issues on other boards, both with the drive and other PCIe devices that get cancelled when it's occupied for some reason, doubly so since it's a boot drive? Maybe something internally is starting up and blocking the OS from functioning with the rest of the system and it causes the hangs? I doubt it, but I figured I would mention it. I've heard stranger things. For the record I do not have any other PCIe devices besides my Graphics card. I'm ordering a M2 converter drive to attempt to remedy this by switching my Sata M2 into the converter and then my boot NVMe into the M2_1 slot so it can be unencumbered by a shared lane and be directly connected to the CPU.
Any and all assistance of any kind I would be forever grateful for, because at this point I am basically chasing a ghost: if it is a specific program, it is leaving no traces at all anywhere about the issues, and if it is hardware, aside from the spare power supply I have no reasonable way to test any individual parts by switching them out, since my last computer was much older and had vastly different components, and I really, really, really do not want to have to start sending hardware back en masse just hoping that it fixes something that might be wrong with the whole thing.
Continue reading...