|
Problem description: I've been seeing graphical glitches both in games and while web browsing on my PC. I suspect it's signs of GPU failure but I don't have personal experience to know if this is really what it looks like. I see brief flickers of certain graphic elements in games or menus and UI on websites, I also occasionally see black pixelated looking errors on screen, which I'll attach screenshots of that I've been quick enough to capture. Problem has been going on for months and seems like it's getting worse. I mostly just want to make sure that going through the process of RMA'ing my card is the right move. Maybe someone can look at this and know exactly what's going on. Only have until December until my GPUs warranty runs out. Attempted fixes: I tried disconnecting my second monitor to just run my main monitor. I tried cleaning and re-seating my GPU as well as cleaning my PC in general. I've made sure my graphics drivers were up to date. I've checked temps under load and at idle which seemed more than fine. Recent changes: I'm unsure as it's been a while but I don't believe so.   -- Operating system: Windows 10 Pro 64-bit System specs: https://pcpartpicker.com/list/rhqn4s GPU: MSI RTX 3070 GAMING X TRIO Monitor: ASUS TUF Gaming 27" 2K HDR Monitor (VG27AQ1A) I have Googled and read the FAQ: Yes. Google had hits similar-ish to what I was seeing from people on reddit, but no one diagnosed the problem aside from someone suggesting to disable hardware acceleration. That sounded more like taping up a poster over a crack in the wall to me. I don't have any money for a new GPU. Any opinions or advice is very very appreciated.
|
#
¿
Nov 5, 2023 02:37
|
|
|
|
| # ¿ May 10, 2024 04:04 |
|
|
Zogo posted:I'd run DDU: https://www.wagnardsoft.com/forums/viewtopic.php?t=4731 and then reinstall the latest GPU drivers. There were graphical glitches on the side monitor too, but not exactly the same ones. Similar rectangles on the screen but not black, and they didn’t show up in screenshots. I can see if I have a backup cable to try and I’ll look into DDU thank you.
|
|
#
¿
Nov 7, 2023 03:43
|
|
|
down1nit posted:Yeah that's VRAM corruption even. You can try MSI Afterburner and give it a little bit more power if you like, it could help. I have a month left of warranty (to my knowledge. MSI is really dodgy about admitting they even have warranties). Started the process thank you. I suspected it was, but I don't have experience with it, and wanted second opinions before I went though all the hassle and expense.
|
|
#
¿
Nov 11, 2023 08:19
|
|
|
down1nit posted:Did they ever send you a replacement? They did. They didn't repair mine they just sent me a refurbished card and didn't diagnose or tell me what was wrong with it. The new card runs better but the fans or whatever buzz under load so that sucks (I have hyperacusis). Also the led strip is washed out for some reason but I don’t care that much about that. The uh, problem didn't go away though so I guess it wasn't the card. Any ideas what could be causing this if it's not the GPU? Cheap Shot fucked around with this message at 21:13 on Dec 3, 2023 |
|
#
¿
Dec 3, 2023 05:32
|
|
|
I haven't managed to track down the cause of the screen glitching but I'm wondering if this is related? I was playing cyberpunk and my computer crashed. I rebooted it and it crashed again. First time my monitors lost input and I heard a short angry buzz. I had to hold down the power button. The second time it did the same with the screens losing input, and then the computer restarted on its own. After some googling, I checked the event viewer which logged both events. I don't know what any of it means and it wont let me open the related files. Here's all the logged events around the first crash that happened while playing cyberpunk and watching twitch on my second monitor.  code:code:code: code:code: There were a number of identical events logged before each crash but I'm not sure which if any are related. The event logged right before both crashes was MsInstaller: code:Did they send me a bunk GPU back on my RMA? Is this a RAM issue? Neither? Cheap Shot fucked around with this message at 08:02 on Dec 13, 2023 |
|
#
¿
Dec 13, 2023 07:54
|
|
|
Zogo posted:All three are possibilities. I have updated the bios yes. It should be the latest version. I'll see about running memtest thank you. Edit: drat, happened again. Both monitors lost input and I got disconnected from my discord call. I had to hold down the power button. Cheap Shot fucked around with this message at 08:00 on Dec 16, 2023 |
|
#
¿
Dec 16, 2023 03:09
|
|
|
Zogo posted:You could run https://memtest.org/ at some point to check RAM health. So I couldn't manage to get this to work and I got the problem a bunch again. I'll try and lay out what I've done tonight in about the order I did it. got the issue again and checked the event viewer and it said this: code:code:I created the memtest boot usb but I couldn't find it in BIOS and then my monitors went dark again while in BIOS so I had to reset. Then both my monitors shut off again while I was typing this reply a bit ago and now I'm having to retype it. I don't seem to have anything in the event viewer from that but I guess that makes sense since I was in BIOS and not windows? Also noticing a lot of MsiInstaller entries about "Microsoft gaming services" and "GameInput" in the event viewer, all over the place. I looked it up and found people saying GameInput was safe to uninstall and even suggesting it was causing my problem so I removed it in add and remove programs. MsiInstaller: "Product: Microsoft GameInput -- Removal completed successfully." although an hour or so later I see MsiInstaller "Product: Microsoft GameInput -- Installation completed successfully." So that didn't stick. I see multiple entries after that also saying "Product: Microsoft GameInput -- Installation completed successfully." several more times after and before crashes so I don't know what the gently caress is going on with that. I disabled "automatically restart on windows system failure" to see if I could see if there was an error message on screen but I couldn't get any display on the monitors to see, even swapping ports and even from displayport to hdmi and switching my monitor to hdmi input. code:code:code:code:code:code:code:code:I recall I enabled XMP profile (well, the asus equivalent DOCP) at one point when a friend told me about it. That was just within a few days of installing the new video card and also getting and hooking up a new second monitor. I've reverted that and turned off DOCP to see if that's what was causing the blackscreen and crashing issue. I'm not sure what it means if that resolves it though. Bad ram I assume? I couldn't get BIOS to detect the memtest usb so I'm a little stuck there. I was starting to think the problem only happened when watching video, but then it happened in BIOS so I'm not sure now. I'm really begging anyone who knows what these error codes mean, or how to open more detailed log files to enlighten me please. I'm very lost. Could the screen glitches I showed in my screenshots be caused by bad RAM, and could these system errors and bluescreens and blackscreendiagnositcs be caused by enabling the DOCP profile with bad RAM? Or is it possible other hardware like my motherboard or gpu are to blame? I don't really have money to replace things or even take it to someone to get it looked at. Times aren't good for me. 
Cheap Shot fucked around with this message at 09:23 on Dec 18, 2023 |
|
#
¿
Dec 18, 2023 06:02
|
|
|
down1nit posted:What tools do you have there? Any second pc or spare parts? I have screwdrivers. No spare pc, and the only spare part I have is an old 980 which I used while the 3070 was out for RMA. The only receipt I have is the original one for the 3070 that I RMA’d. I may have e-receipts for other parts, I’m unsure. Most of the parts in my computer would be about 4 years old now aside from the gpu which was 3. As far as the crashing/losing monitor input problem goes, it didn’t happen again last night after disabling DOCP but it doesn’t happen consistently so I can’t say for sure that’s what was triggering it yet. It was happening a lot yesterday before I disabled that though. Far more frequently than in days before. Thank you for offering to help me! I really appreciate it. My phone plan doesn’t allow calls outside Canada so I’d have to stick here. Or discord or something. Cheap Shot fucked around with this message at 21:07 on Dec 18, 2023 |
|
#
¿
Dec 18, 2023 21:02
|
|
|
I disabled DOCP, but it still lost monitor input and crashed just now, requiring a press of the power button to turn it off. I thought for sure that must have been why it was crashing but now I don’t know.
|
|
#
¿
Dec 19, 2023 03:29
|
|
|
Zogo posted:Did these issues happen while using the 980? No, the blackscreening is a recent thing. To be honest though I didn’t use my computer very much while the 980 was installed so I’m not sure. I couldn’t even really run ffxiv without it sounding like that gpu was trying to achieve liftoff. I was actually kind of surprised how poor stuff was running on it. I feel like I used to play games like that on it no problem. The things I did before the blackscreening issue started: Removed the 980 and Installed the 3070 I received back from MSI after RMAing, but still experienced the graphical glitch in my OP. I used DDU to do a clean install of the 980 drivers and then used it again to install the 3070 drivers. I Enabled DOCP. I bought a second of my main monitor to replace an old side monitor. I installed both monitors on arms and replaced the displayport cables with new longer ones to reach. I think that’s everything I did recently. The blackscreening started sometime after all that. I couldn’t get that memtest to work so I tried the passmark version which I was able to get to work (on my second try. It blackscreened while I was making the usb the first go.) This was the result.    I didn’t realize how long the test would take so I regret starting it at 1am. I’m also surprised it didn’t blackscreen throughout the whole test. Right before I ran it I unplugged both my 1440p monitors and instead plugged in this ancient 2006 dell using a dvi to hdmi adapter. I haven’t yet gone to desktop and run it for a while to see if it still blackscreens/crashes in this configuration. I’m really tired and I’ll have to look into the steps you’ve suggested tomorrow after some rest. Both of your suggestions sound like they might be a bit over my head at a glance but I’m too sick with stress so maybe sleep will give me a clearer look. Thank you for continuing to help!
|
|
#
¿
Dec 19, 2023 10:12
|
|
|
Zogo posted:
I'm not really sure how to use it and it seems pretty limited on the free trial but it says my drives are in good health.  So far it hasn't blackscreened while just using it with this single older monitor. I'll keep using it to see if it happens. Could bad displayport cables cause this? These are the ones I got https://www.amazon.ca/dp/B07JG2LGB9?ref=ppx_yo2ov_dt_b_product_details&th=1. Maybe there's a problem with them that I hadn't considered? Or could it be that the new monitor I bought is bad somehow? My monitors are both VG27AQ1A. That sounds unlikely to me but I don't know for sure. Cheap Shot fucked around with this message at 22:56 on Dec 19, 2023 |
|
#
¿
Dec 19, 2023 22:49
|
|
|
Zogo posted:Those are probably fine. Yes, you’re probably right because it blackscreened and crashed just now on this monitor too…
|
|
#
¿
Dec 19, 2023 23:33
|
|
|
Zogo posted:Yeah, there can be a tendency to think an issue is resolved if a few hours/days go by with no problems. But you really need a solid 2-3 weeks with complete stability to be confident that the issue is resolved. I’m wondering if MSI just sent me back a bunk refurb when I RMA’d my card? I sent it in because of the screen glitching, but the screen glitching was still present with the new one. I haven't seen the screen glitching in a while too and I’m wondering if some of the fixes I tried for the crashing actually resolved my original issue and this crashing is unrelated? The card they sent me has bad coil whine and washed out LEDs for some reason so I’m wondering if I just swapped a fine card for a hosed one? It can’t even recall the last time I had a crash on this computer prior to this RMA. The screen glitching was annoying and concerning but it was fairly infrequent and wasn’t shutting down my computer. Also I don’t know if this tells us something but my computer stays on longer before crashing after letting it stay off for a while. If I try using it right away after it crashes it starts happening more and more frequently . Cheap Shot fucked around with this message at 00:29 on Dec 20, 2023 |
|
#
¿
Dec 20, 2023 00:27
|
|
|
Zogo posted:All three of those things are possibilities. Thank you, I’ll try that. The program I was using maybe wasn’t reporting temps accurately. The temps seemed fine, good even, but I’ll see if I wasn’t getting the full picture. It happened even just idle on desktop however so I’m not sure.
|
|
#
¿
Dec 20, 2023 08:23
|
|
|
I haven't been able to test yet because of the crashing. I found this reddit post thats eerily similar to my experience. Edit: More and more I’m thinking it’s the GPU. I guess I should have been more suspicious when the card they sent me had bad coil whine and a washed out LED strip. And I never had any crashes prior to the RMA. The service slip does say they swapped my card, so likely they just swapped mine out for some poor abused card on its deathbed that they got a signal out of and called good enough. Cheap Shot fucked around with this message at 10:53 on Dec 20, 2023 |
|
#
¿
Dec 20, 2023 10:19
|
|
|
A bit of an update I guess. I received a card back from the second RMA today and the thing is full of dust and covered in oily stains on the shroud and the visible parts of the board. My hands are actually itchy from handling it. They offered to have it “extensively tested” before sending it to me this time (why the gently caress aren’t they doing that anyway?) but I doubt this thing was ever even refurbished let alone tested. I think it’s pretty clear that MSI was my problem the whole time. For the record, my pc hasn’t black-screened or crashed once after swapping my older backup card in while I waited. I guess I have to scrounge up the funds for a new one. loving crazy. Cheap Shot fucked around with this message at 19:43 on Jan 10, 2024 |
|
#
¿
Jan 10, 2024 19:23
|
|
|
 Here’s some of the caked on dirt in the fan after blowing compressed air through everything. I’m trying to imagine the pc this came out of. There’s no shot they actually refurbished this thing right?
|
|
#
¿
Jan 11, 2024 04:45
|
|
|
down1nit posted:Refurbished does actually mean "used but tested." Companies are contracted to test and clean these *usually* but it's such a poo poo job brushing and blowing fragile rear end fans all day that it's probably underpaid kids without facemasks It’s already crapping out. I just gave up and am bankrupting myself trying to build a new computer in tyool 2023-24. Just gonna swear off MSI forever now, and also Corsair while I’m at it since everything I’ve bought from them has not held up.
|
|
#
¿
Feb 3, 2024 11:41
|
|
|
Welp. Spent all my money on a brand new computer and it just crashed on me. Screens went black and it rebooted. Just like the last pc was doing when it wasn’t freezing on no signal. What the hell could be the problem?? The two sata ssd’s I wiped and carried over? They’re the only things from the last computer but I ran checks on both of them. The monitors? The powerbar my pc is plugged into? Or is this just a cruel coincidence? I wish I knew how to diagnose this poo poo. I’m so run down and exhausted. Edit: This is the information I was able to pull from the minidump  Does this help at all? I'll copy paste the seemingly important lines for ease of lookup. 0x00000119 00000000`00000002 ffffffff`c000000d ffff8a82`289b72e0 ffffbb8d`c74a5e90 watchdog.sys watchdog.sys+3ad0 Watchdog Driver 10.0.19041.3636 (WinBuild.160101.0800) ntoskrnl.exe+3fd6f0 There were also a lot of lines in the dump reader and three at the top were red so here they are:  Edit 2: I'll just go ahead and pull some info from the event viewer too. These two in application: https://pastebin.com/BcPAVF7a https://pastebin.com/3wFPCxFi This one in System, not sure if it's related: https://pastebin.com/tW6vx1Vs Cheap Shot fucked around with this message at 12:34 on Feb 10, 2024 |
|
#
¿
Feb 10, 2024 10:38
|
|
|
Zogo posted:Is the OS fully updated and also is the motherboard on the latest BIOS? OS and BIOS should be up to date since I updated them. For updating bios I used the gigabyte command center, which, I dunno. I ended up uninstalling that software because it seemed like it might be causing issues. Since I built this new one I’ve had one instance of it failing to wake up from sleep and needing a restart, and one instance of the display drivers needing to be restarted with the key press combo to get picture back. And then now this crash and reboot from playing a game while watching a stream. Exhausting. Does the crashdump or eventlog info have any useful information? Cheap Shot fucked around with this message at 23:49 on Feb 10, 2024 |
|
#
¿
Feb 10, 2024 23:29
|
|
|
down1nit posted:The error logs suggest your video card again. HOW loving FUNNY IS THAT? Funny enough to leave me in tears.  Ok lets see. No headset right now. Wired mouse and keyboard. Speakers are some old green plug analogue ones I’ve been using for over a decade with a subwoofer which I mostly dont use because of my ear damage. I’ve got an xbox elite controller sometimes plugged in. A samson C01U usb condenser mic also plugged in. A usb splitter that I don’t actually really need now and could probably unplug until I have a use for it. I think that’s it. Monitors are two of these: https://www.asus.com/ca-en/displays-desktops/monitors/tuf-gaming/tuf-gaming-vg27aq1a/ Cables are both 3m displayport. These: https://www.amazon.ca/dp/B07JG2LGB9?psc=1&ref=ppx_yo2ov_dt_b_product_details I know the last computers crashing was also happening with different cables and different monitors, but this new pc’s crash might not have even been related to my previous pc’s problems. The only files I’ve transferred from my old pc are a few game files to save redownloading them, their appdata folders with saves, screenshots, and a few music and blender project files. Everything else is just staying on a portable external SSD for safe keeping. Edit: I know you said it suggested gpu related issues, but could the errors be cpu chipset driver related? I don’t think those got installed properly maybe. I used gigabyte command center to try and install drivers and updates for my board which seemed like a lovely program and I uninstalled it. I decided to update the chipset drivers from AMD directly tonight. Or could it be related to this? This seemed odd to me: 
Cheap Shot fucked around with this message at 07:30 on Feb 11, 2024 |
|
#
¿
Feb 11, 2024 07:23
|
|
|
down1nit posted:Get yourself in troubleshooting mode lol So it just hasn’t crashed again yet. I’m not sure if it’s something I did resolved it, or it’s just not hit the same circumstances again yet? Hard to troubleshoot. The other computer was hard crashing pretty constantly, although I think that really was the bad videocard MSI sent me through RMA. That card was just fully cooked. down1nit posted:Speakers are fine. Ear health is a treasure. It is more too late for me than you could imagine sadly.
|
|
#
¿
Feb 12, 2024 07:50
|
|
|
|
| # ¿ May 10, 2024 04:04 |
|
|
I guess a followup for maybe even a little closure of some kind. I got the same black rectangles screen glitch on the brand new computer recently. Google never really gave me conclusive answers to why it was happening, with most of the results pointing to gpu failure. There were however some people saying it was from Chrome using hardware acceleration. I had switched to firefox which didn’t resolve the issue, and after RMAing my video card, then having to RMA it again after they sent me a turd, I gave up and blew all my savings on a new computer. Since I got the same issue on the new hardware, I switched back to trying to find a software cause. A friend asked me what programs I was using in common with the last computer. I still haven’t put too much on it yet so I narrowed it down to either discord or sharex. Not entirely ruling out sharex, but it turns out discord has hardware acceleration on by default and it supposedly is chromium based. I switched it off and it hasn’t happened again since. I’m still on watch but I think that was likely it. I don’t quite understand though. It feels like it should be more widespread of a problem if it’s just because of chrome and discord, two very widely used programs. I’m skeptical but the common thread seems to have mostly been while video was playing in chrome and streams/picture in picture video was playing on discord now that I can look back and reflect. I always had at least a browser or discord open on my second monitor. Anyways, shame it went the way it did and I had to waste a load of money and deal with probably the worst RMA service I’ve ever encountered. The card I RMA’d was likely fine, replaced with a completely cooked one, and then a filthy one with rattling fans. I never get away without maximum suffering but at least this is probably resolved now though? 
|
|
#
¿
May 1, 2024 12:15
|
|