As was foretold, we've added advertisements to the forums! If you have questions, or if you encounter any bugs, please visit this thread: https://forums.penny-arcade.com/discussion/240191/forum-advertisement-faq-and-reports-thread/
Options
Physically moving my computer into another room has ruined it
UncleSporky
Registered User regular
Registered User regular
Computer built in 2014.
Intel Core i5-4670 3.4GHz 6MB Cache Quad-Core Desktop Processor BX80646I54670
Gigabyte LGA 1150 Intel B85 HDMI SATA 6Gbps USB 3.0 Micro ATX DDR3 1600 Intel Motherboards GA-B85M-D3H
Kingston KHX16C9B1RK2/8X HyperX Red 8GB (4GB 512M x 64-Bit x 2 pcs.) DDR3-1600 CL9 240-Pin DIMM Kit
EVGA 500B 500W 80PLUS Bronze Certified ATX12V/EPS12V 500W Power Supply 100-B1-0500-KR
Started with a 750ti but in 2016 upgraded to:
EVGA GeForce GTX 1070 SC GAMING ACX 3.0 Graphic Card 08G-P4-6173-KR
256 gig solid state primary drive with a 512 gig platter secondary drive. Still running Windows 7 (this changes).
I've been playing Satisfactory. It's fun. I realize my computer chair is kind of hurting my back and it's so much more comfortable in the recliner in the living room, so I decide to move my computer in there and play on the TV, not something I've ever done before.
Previously stable computer that I could leave running multiple days suddenly starts bluescreening very quickly after turning it on, usually while running the game but not always. Kernel data inpage error with the error code 0x0000007a that might indicate hardware issues. Check event viewer and there are atapi errors corresponding with every bluescreen, "the driver detected a controller error on \Device\IDE\IDEPort1."
Scan the RAM, it's fine, run chkdsk, it's fine. Open computer, reseat RAM and SATA cables and mobo cables. It's not that dusty inside, nothing really caked on, clean it out a bit. I remember in the past on various computers I've gotten strange errors when trying to use the front USB ports not directly part of the motherboard, so I switch the kb/mouse/wifi to be connected in back instead.
It feels like doing all that worked! I play Satisfactory for 4 or 5 hours and then start winding down just reading stuff online. Bluescreens again. I laugh and go to bed. The next day I try again, play video games for a few hours, get ready to take a break, bluescreens again. The computer is now politely waiting for me to finish using it before dying. Event viewer no longer is getting the atapi errors, the bluescreen says the error happened in ntfs.sys.
Ok, so the whole computer is old. I have a new hard drive I had bought in preparation to update eventually. I unplug the two internal drives for later pulling of files since they seem to still work, and plug in the brand new drive by itself, just a 512 gig SSD. I install Windows 10. I install everything I would need, NVIDIA drivers and control panel, browsers, Steam, PDF viewer, VLC, WinRAR, y'know. Get it all set up normal.
Oh my god it's amazing, so clean and nice. I play Satisfactory and it doesn't crash. I perform various updates and restart and things are going great. I leave Satisfactory running overnight to keep generating resources in the game. I wake up and it's fine and the computer is fine and I'm happy and think everything is solved.
Assertion failed: [File:D:\ws\SB-200518130817-657\UE4\Engine\Source\Runtime\Windows\D3D11RHI\Private\D3D11Util.cpp] [Line: 198] Unreal Engine is exiting due to D3D device being lost. (Error: 0x887A0006 - 'HUNG')
Satisfactory dies. I can no longer launch it, this happens every single time. Some posts online seem to think this could be related to hardware issues.
As I'm researching this, my wifi adapter also starts dying. I hear the telltale "doodloo" sound of Windows thinking I unplugged it, it shows no connection, then after a minute the connection comes back. It does this every 5 to 20 minutes.
Repeated event viewer errors:
WLAN Extensibility Module has failed to start.
Module Path: C:\Windows\system32\Rtlihvs.dll
Error Code: 126
Looking up info on this it could be driver related. So far I was using the default Windows drivers and they were working fine, so I'll try the official Realtek download instead.
I will say this only started after installing NordVPN again, so it's possible something about the way they twerk the network drivers caused this.
We have caught up to the present. At least Windows 10 hasn't bluescreened and crashed yet. But I'm really getting fed up with the whole situation. I guess I'll reseat and blow out the video card and see if that changes anything.
But seriously, is my motherboard dying? Why is everything now terrible forever because I moved my computer into another room?
Intel Core i5-4670 3.4GHz 6MB Cache Quad-Core Desktop Processor BX80646I54670
Gigabyte LGA 1150 Intel B85 HDMI SATA 6Gbps USB 3.0 Micro ATX DDR3 1600 Intel Motherboards GA-B85M-D3H
Kingston KHX16C9B1RK2/8X HyperX Red 8GB (4GB 512M x 64-Bit x 2 pcs.) DDR3-1600 CL9 240-Pin DIMM Kit
EVGA 500B 500W 80PLUS Bronze Certified ATX12V/EPS12V 500W Power Supply 100-B1-0500-KR
Started with a 750ti but in 2016 upgraded to:
EVGA GeForce GTX 1070 SC GAMING ACX 3.0 Graphic Card 08G-P4-6173-KR
256 gig solid state primary drive with a 512 gig platter secondary drive. Still running Windows 7 (this changes).
I've been playing Satisfactory. It's fun. I realize my computer chair is kind of hurting my back and it's so much more comfortable in the recliner in the living room, so I decide to move my computer in there and play on the TV, not something I've ever done before.
Previously stable computer that I could leave running multiple days suddenly starts bluescreening very quickly after turning it on, usually while running the game but not always. Kernel data inpage error with the error code 0x0000007a that might indicate hardware issues. Check event viewer and there are atapi errors corresponding with every bluescreen, "the driver detected a controller error on \Device\IDE\IDEPort1."
Scan the RAM, it's fine, run chkdsk, it's fine. Open computer, reseat RAM and SATA cables and mobo cables. It's not that dusty inside, nothing really caked on, clean it out a bit. I remember in the past on various computers I've gotten strange errors when trying to use the front USB ports not directly part of the motherboard, so I switch the kb/mouse/wifi to be connected in back instead.
It feels like doing all that worked! I play Satisfactory for 4 or 5 hours and then start winding down just reading stuff online. Bluescreens again. I laugh and go to bed. The next day I try again, play video games for a few hours, get ready to take a break, bluescreens again. The computer is now politely waiting for me to finish using it before dying. Event viewer no longer is getting the atapi errors, the bluescreen says the error happened in ntfs.sys.
Ok, so the whole computer is old. I have a new hard drive I had bought in preparation to update eventually. I unplug the two internal drives for later pulling of files since they seem to still work, and plug in the brand new drive by itself, just a 512 gig SSD. I install Windows 10. I install everything I would need, NVIDIA drivers and control panel, browsers, Steam, PDF viewer, VLC, WinRAR, y'know. Get it all set up normal.
Oh my god it's amazing, so clean and nice. I play Satisfactory and it doesn't crash. I perform various updates and restart and things are going great. I leave Satisfactory running overnight to keep generating resources in the game. I wake up and it's fine and the computer is fine and I'm happy and think everything is solved.
Assertion failed: [File:D:\ws\SB-200518130817-657\UE4\Engine\Source\Runtime\Windows\D3D11RHI\Private\D3D11Util.cpp] [Line: 198] Unreal Engine is exiting due to D3D device being lost. (Error: 0x887A0006 - 'HUNG')
Satisfactory dies. I can no longer launch it, this happens every single time. Some posts online seem to think this could be related to hardware issues.
As I'm researching this, my wifi adapter also starts dying. I hear the telltale "doodloo" sound of Windows thinking I unplugged it, it shows no connection, then after a minute the connection comes back. It does this every 5 to 20 minutes.
Repeated event viewer errors:
WLAN Extensibility Module has failed to start.
Module Path: C:\Windows\system32\Rtlihvs.dll
Error Code: 126
Looking up info on this it could be driver related. So far I was using the default Windows drivers and they were working fine, so I'll try the official Realtek download instead.
I will say this only started after installing NordVPN again, so it's possible something about the way they twerk the network drivers caused this.
We have caught up to the present. At least Windows 10 hasn't bluescreened and crashed yet. But I'm really getting fed up with the whole situation. I guess I'll reseat and blow out the video card and see if that changes anything.
But seriously, is my motherboard dying? Why is everything now terrible forever because I moved my computer into another room?
Switch Friend Code: SW - 5443 - 2358 - 9118 || 3DS Friend Code: 0989 - 1731 - 9504 || NNID: unclesporky
0
Posts
Just for shits and giggles, have you thrown the computer back to the first room again? Just on an extremely off chance that there is a weird power problem in the living room circuit? The chances are near zero but I'd do it just for curiosity.
I haven't put it back in the other room again yet, like you said the chances seem low that that would affect anything. And also at this point I've changed enough that it might not be a good test anymore, I'd probably want to plug back in the old drives too and see if in the complete old state of things everything would be fine. But at this point I'd rather be able to get this clean install stable and work from there.
And, on that note...I installed Realtek's drivers and it stopped the wifi issue immediately and it hasn't happened since. I then verified the Steam install files of Satisfactory, and also ran another reasonably demanding 3D game (Planet Coaster), which worked just fine. And then I ran Satisfactory again and it was also just fine, and continues to work. So it's possible it was just several coincidences making it look worse than it was...
This is hotter than a lot of people like to keep it under load but it's not death, is it?
Anyway I don't want to declare it good because I've done that like three times throughout this whole process and it's never been true. But we'll see.
Yeah I thought I had read that. Though with the age of everything it's probably better to keep it cooler if possible, but heat doesn't seem to be the issue.
Anyway thanks for reading my part-cry-for-help-part-frustrated-rant. Fingers still crossed. I just hate when I can't pinpoint a specific component that's causing a problem.
Plugging it in with a surge protector and on the old source of power also does nothing. Reseated the video card too.
I hope it's the motherboard dying and not the 1070.
Constant event viewer warnings:
Display driver nvlddmkm stopped responding and has successfully recovered.
It sounds like either your PSU or, less likely, something knocked free against (or possibly from?) the motherboard while you were moving it.
I've reseated everything short of completely dismantling the computer and reassembling it. Didn't see anything that is touching the motherboard that could be shorting it out or anything.
I haven't tried that, I haven't heard of an un-updated BIOS causing screen artifacting or graphics crashes, especially when it worked fine for a few days. It feels like something has slowly been rotting or dying since starting over on the new hard drive.
I do need to install the specific chipset drivers still.
If it turns out not to be the power supply, or if it turns out the faulty power supply fried my motherboard or 1070, then hey, at least I have a nice new power supply to build a new computer with.
What I suspect happened is the wiring may not be so great to that one outlet and it surged into the PSU and now it's not delivering power properly. Though the fact that you're having artifacting and IDE issues makes me suspect it might be RAM too... but also can't rule out motherboard issues at the same time.
I would be very surprised if your GPU got taken out, even if it was a bad power supply. They are pretty robust these days, and one of the things necessary for extreme overclocking is actually doing shunt bypass to allow higher power delivery. While it's certainly possible, I'd say it's highly unlikely, and that the issues are actually unstable or too low power draw, versus it being overloaded and fried.
2D games seem fine, I could run Ultimate Chicken Horse. Video seems fine, I just spent some hours watching Hulu. Everything about the computer is perfect except it can't render in 3D off the GPU.
When the games crash, they leave an entry in event viewer as before that says "Display driver nvlddmkm stopped responding and has successfully recovered."
Normally I might think this has to do with drivers, but again, the way it felt like things were slowly dying as I posted above...the games worked for a while, then crashed, and then stopped working altogether. Even persisted that way after a full driver wipe and reinstall.
So now I will try messing with drivers again, install the motherboard chipset ones and redo the video card drivers again, and if none of that works I will give my card to a friend and see if he has the same trouble.
EDIT: Installed one chipset driver (Intel INF installation). Tried to install another one categorized under chipset, Intel Management Engine Interface, and was told that I already had a newer version of it. Also installed Realtek audio driver, and UNinstalled NVIDIA's drivers again. Still plugged into HDMI through the 1070, but with no drivers installed and stuck on 1024x768...Planet Coaster runs fine in full 3D. Well, it doesn't run "fine," it's a slideshow, but it works. Interesting...
EDIT2: After installing the NVIDIA driver, crashes exactly like before, immediately. Same event viewer log. "Display driver nvlddmkm stopped responding and has successfully recovered." Also tried tweaking a number of settings, going back to 1024x768, no screen scaling, nothing else open at the time, still crash.
1) Do you actually have your display plugged into the graphics card and not the onboard video by accident?
2) Consider running DDU to completely wipe away any remnants of old drivers.
3) You said you got the new GPU put it, based off previous discussions, is it safe to assume that it was a PSU and not GPU, and that was just a typo?
Edit: One other quick addendum: Your GPU is in the very top PCIE slot on your board?
1. No, it's plugged into the graphics card.
2. I have tried DDU several times already.
3. Typo.
4. Yeah, it's in the top slot. I think it's the only slot that can accommodate the card? It's in the same slot where it always was and functioned fine for 4 years.
Roll back to a driver from last year and if that doesn't work seems like maybe your graphics card might be bad.
The graphics card has the bonus power plugs plugged in, correct?
Yeah I'm gonna try an older driver, but the best test will be having a friend with a completely different computer try to use it, if/when I can get it to him.
What do you mean by bonus plugs? The only power I have ever had plugged into it is right here, and all of them are plugged in:
I had to do this:
Get REVO Uninstaller, download the NVidia drivers but don't install.
Remove the NVidia drivers via REVO and kill all remaining registry entries when that screen pops up. Even after uninstalling via DDU it still leaves things in Windows.
Click windows button, hold shift and click RESTART. This will bring up the settings menu, click TROUBLESHOOT, ADVANCED, and then RESTART. Select SAFE MODE without networking. Optional: Shut down, then power on, and continue to shut down with power button 3 times to get the Win10 boot to fail 3 times, which will bring up the boot options.
Unplug your internet cable just to be sure.
While in Safe Mode, reinstall the NVidia drivers. Reboot into normal Windows.
The only change is if you're using Win 7, and get it into safe mode via the normal Win 7 method.
Also make sure if you got a modular power supply that they are firmly plugged into the power supply end as well. Other than that, the card may finally have gone I guess. It still sounds very much like a power delivery issue with the drivers and it working fine until there's a heavier load put on it.
I just tried all this.
Used REVO to uninstall NVidia, which felt a little less thorough than DDU. It ran the normal uninstaller then scanned for leftovers. Once it was gone I also ran DDU for good measure.
Then in safe mode I installed the oldest NVidia drivers on their site, which were from November 2019.
Didn't help, same exact situation as before.
I have discovered that every time I right click the desktop and choose Display Settings, I can trigger flickering and artifacting. It flashes and gives me black bars every time. Also sometimes does things like this:
Do you have your old video card to try?
I don't think it's overheating. When I launch a game it immediately crashes, like the act of...I dunno, issuing a 3D draw call, is what's killing it. There isn't time for it to heat up.
Would it be possible that the card's RAM is bad, and trying to initialize some part of the RAM is doing it? One piece of advice for that event viewer error that I saw online was that someone else temporarily fixed it by intentionally downclocking and hobbling their RAM.
Another thing I could try...the old Win 7 installation was blue screening, but after minutes/hours of normal use. I could plug them back in and see if now suddenly the games don't launch on that old installation either.
I had given my 750ti to a friend in a time of need, but he's since upgraded and didn't need it anymore so he gave it back. I might need to use it as a fallback for a while but it would also be a good test to see if something about the rest of the computer/installation is rejecting 3D graphics.
But it could be some GPU processor and ram combination issue, it's so strange.
But yeah GPU is the likely culprit now that we've ruled out PSU. If you have a spare older one for sure try that (or a friend's).
It does. I haven't used it yet but like I said above, I was able to run every game without any NVidia drivers installed, off the 1070. They were slideshows but they actually ran. I would guess the same would be true with the onboard.
It def seems like the gpu is dying but weird it just started after moving the system.
I guess once you get your hands on a friends system you can check.
If you got a spare drive you could install Ubuntu or some other districts and check if you can run anything 3D. Kinda rule out the software issue.
My issue hadn't been instant crash on attempting something 3d, but a cycle of behaving stable for a while, then a period of intense instability to the point of the GPU not even being recognized as present, then back to stable for a while, repeat.
Warframe/Steam: NFyt
She's dead, Jim!
As bad as it is to lose a nice video card after only 4 years, it's also a huge relief to know exactly what the problem is. I have a 750ti I can use if I have to, until new cards come out. But for now I'm playing Disgaea 4 on Switch and have plenty of other things I can do and play, it's not the end of the world.
Thanks for everyone posting in this thread on my diagnosing journey.
Y'know at the outset I do think something was going wrong with the hard drives too...those initial errors were totally different, IDE controller things, ntfs.sys. I don't think a crashing video card would've caused those kinds of errors, but I'm also not an expert on these things. It seems like a pretty big coincidence for the hard drives and video card to all decide to break at the same time after moving a computer into another room.