Question Computer Still Shutting Down

[DeadSulfur]

So I thought I was done with this. I bought a computer from CyberPowerPC (specifically a Cyberpowerpc gma 3200 bst). It worked fine for a bit, but then it started to shut off while gaming. I thought this could either be a temperature issue or a PSU issue. The temps weren't amazing, but they theoretically shouldn't have been the problem. I read online that CyberPowerPC PSUs are notoriously bad, so I went ahead and swapped it out with a Corsair RM 750x (750W). It worked fine since I installed it just before Easter.

Fast forward to tonight, and my computer shut off twice while trying to play Overwatch. I tested my temps the second time and they never got above 60 C. I'm not sure what's going on at this point, and I don't know of a good way to test things. In terms of event logs, it's essentially just saying the generic "shutdown unexpectedly" critical event.

Here's the specs:

• Windows 10 Home 64 bit (10.0, Build 17134)
• System Manufacturuer: MSI (450M Bazooka)
• System Model: MS-7A38
• BIOS: h1.0
• Processor: AMD Ryzen 7 2700X Eight-Core Processor (16 CPUs), ~3.7GHz
• Memory: 16384MB RAM (GDDR4)
• Display: NVIDIA GeForce RTX 2070 (EVGA), 8031 MB VRAM
• 240 GB SSD (where the OS is installed), 2 TB HDD

Any help would be greatly appreciated. I tried out CyberPower tech support originally, but essentially ended up taking things into my own hands with the PSU swap (which was a nightmare by the way). After that fixing the issue for quite some time, I'm not really sure what it could be.

Oh, and I haven't updated drivers beyond automatic Windows updates. I know old drivers can cause issues, but new ones can also cause problems. I don't see a bluescreen, so I feel like it's likely not old graphics drivers, but I thought I'd try ruling out what I can first.

 

[Phaaze88]

Incompatible drivers do indeed cause issues, but they don't cause instant shut-offs, like you're describing. This points to a hardware issue.
Nothing else in the PC was tinkered with, right? Double check all the psu connections, to be sure nothing is loosely plugged in.

 

[DeadSulfur]

Incompatible drivers do indeed cause issues, but they don't cause instant shut-offs, like you're describing. This points to a hardware issue.
Nothing else in the PC was tinkered with, right? Double check all the psu connections, to be sure nothing is loosely plugged in.

Nothing was intentionally tinkered with anyway. Unplugging the original plugs from the original PSU was quite challenging, so if it is a loose cable problem, I don't believe it was one originally. It has quite a few additional cables for LED shenanigans. They're cool, but I wouldn't have gotten them had I built the computer myself.

One thing I feel like mention is I've had some mouse issues with the computer. I haven't gotten around to fully debugging it, but I've noticed with my wireless mouse that it will seem to freeze sometimes. This will even happen on desktop without much going on. I've noticed it mostly while playing Siege, but I think it's more from me just caring more. It kind of comes and goes as well. I'm fairly certain it's just the mouse freezing and not the entire computer when it does this, but it's quick enough that it can be hard to tell.

Out of curiosity, could this be the cause of bad wiring/electricity in the house? I live in a rather old duplex, and I haven't changed where I've had my computer(s) plugged in since I moved in (the lights didn't shut off or anything, just the computer).

 

[DeadSulfur]

I changed my power settings to use a maximum 70% power for my processor. After that I was able to play games for a few hours without any shutoffs. That leads me to believe it must be power or temperature related. I wouldn't think it was temperature based on the numbers I've seen and the numbers it's been under without shutting off, but I suppose it could be something really quick that's causing it that doesn't even show up in Afterburner.

 

[Phaaze88]

I can't give a concrete answer on the electrical issue. Someone else may be able to chime in on that.
But, you have household appliances that EASILY draw more power than the PC, and they're not instantly shutting off, right?

I've had a similar issue with one of my previous mice, a Razer Naga, I believe. The tracking would just fail, leading to the freezing you're describing. Things you can do about it: update the firmware, clean the pad and belly of the mouse(alcohol and a soft cloth to avoid scratching the laser device), use the cable connection - IF your unit came with one - if none of that helps, consider replacing that mouse.

It still sounds like a psu issue, and your last post further points to it. Kinda nuts, as that's a good unit there, but I even the best parts have their share of duds, it's just less likely.
What are you using to check temps with? Msi Afterburner, Core Temp, and HWINFO are some examples. 85C is the thermal limit for a 2700x, and I can see it shutting down if temperatures were to keep rising beyond that, but going by your first post, you're not getting anywhere near that.

The power supply is still under warranty, so it wouldn't hurt to RMA it.


***Geez, I tried to reply Friday night, and after I clicked send, I got sent to the maintenance page, so the post didn't go through. I didn't see a notification about this thread until recently.

 

[DeadSulfur]

I can't give a concrete answer on the electrical issue. Someone else may be able to chime in on that.
But, you have household appliances that EASILY draw more power than the PC, and they're not instantly shutting off, right?

I've had a similar issue with one of my previous mice, a Razer Naga, I believe. The tracking would just fail, leading to the freezing you're describing. Things you can do about it: update the firmware, clean the pad and belly of the mouse(alcohol and a soft cloth to avoid scratching the laser device), use the cable connection - IF your unit came with one - if none of that helps, consider replacing that mouse.

It still sounds like a psu issue, and your last post further points to it. Kinda nuts, as that's a good unit there, but I even the best parts have their share of duds, it's just less likely.
What are you using to check temps with? Msi Afterburner, Core Temp, and HWINFO are some examples. 85C is the thermal limit for a 2700x, and I can see it shutting down if temperatures were to keep rising beyond that, but going by your first post, you're not getting anywhere near that.

The power supply is still under warranty, so it wouldn't hurt to RMA it.


***Geez, I tried to reply Friday night, and after I clicked send, I got sent to the maintenance page, so the post didn't go through. I didn't see a notification about this thread until recently.

I get what you're saying about appliances that draw more power, but I'm wondering specifically for this outlet. If the wiring is shoddy I wonder if it's just not always able to draw enough power. (one example of the places shoddy wiring is the majority of the outlets aren't grounded, though the one I'm using is one of the few)

I've been using MSI Afterburner with the on screen display. For many games it usually doesn't go above 65 even after long play sessions. For Siege though it will sometimes get up to 80, though I've reduced that by capping the framerate to 120 fps. It didn't happen with Siege in this most recent case, and I've played plenty of it after "fixing" it the first time.

 

[Phaaze88]

It's only a stopgap, though.
Perhaps try moving and testing your setup temporarily in other rooms? That would help narrow things down.

 

[DeadSulfur]

It's only a stopgap, though.
Perhaps try moving and testing your setup temporarily in other rooms? That would help narrow things down.

That's definitely been the plan.

I installed HWMonitor to see if I could learn anything from the voltages. I set the max processing power back to 100 percent and booted up Overwatch while trying to keep tabs on the stats. I played for a bit without any shutoffs and then booted up Siege and played for a bit without shutffs as well. I didn't notice any strange numbers for voltages at least. I'm not certain what are the ideal numbers for a lot of these power related (watts, amps, volts) related stats though, but the +3.3V, +12V, and +5V values all had very minimal variance.

Temps seemed to be what I'm use to (Overwatch maxed around 65, and Siege around 73). The max temp recorded was 85, but this was during Siege and very, very briefly. For instance, something loaded. I don't know how the threshold for shut off works, but I'm fairly certain super brief temperature spikes aren't uncommon. I would assume they need to hit 100 for more than a few moments at least, but I could be wrong there. Regardless, the last time it shutoff was with Overwatch, and I didn't notice any spikes at the time (the largest spike for the game today being 75, once again very, very briefly). Either way I could use a better air cooler, at least a better fan temperature plan (right now when it gets to 80 it never goes above 65 percent fan use).

I'm not sure at this point. I would assume if it were a PSU problem it would be a bit more consistent (it shutoff back to back the other day for instance), but maybe not. I did upgrade the graphics drivers, since after talking to CyberPowerPC tech support they suggested trying a different graphics card to see if it were driver related (why they wouldn't just suggest updating the drivers is beyond me). Since I hadn't updated graphics drivers at all between replacing the PSU and now, I feel like that shouldn't be it, but who knows.

 

[Sloppymess]

That's definitely been the plan.

I installed HWMonitor to see if I could learn anything from the voltages. I set the max processing power back to 100 percent and booted up Overwatch while trying to keep tabs on the stats. I played for a bit without any shutoffs and then booted up Siege and played for a bit without shutffs as well. I didn't notice any strange numbers for voltages at least. I'm not certain what are the ideal numbers for a lot of these power related (watts, amps, volts) related stats though, but the +3.3V, +12V, and +5V values all had very minimal variance.

Temps seemed to be what I'm use to (Overwatch maxed around 65, and Siege around 73). The max temp recorded was 85, but this was during Siege and very, very briefly. For instance, something loaded. I don't know how the threshold for shut off works, but I'm fairly certain super brief temperature spikes aren't uncommon. I would assume they need to hit 100 for more than a few moments at least, but I could be wrong there. Regardless, the last time it shutoff was with Overwatch, and I didn't notice any spikes at the time (the largest spike for the game today being 75, once again very, very briefly). Either way I could use a better air cooler, at least a better fan temperature plan (right now when it gets to 80 it never goes above 65 percent fan use).

I'm not sure at this point. I would assume if it were a PSU problem it would be a bit more consistent (it shutoff back to back the other day for instance), but maybe not. I did upgrade the graphics drivers, since after talking to CyberPowerPC tech support they suggested trying a different graphics card to see if it were driver related (why they wouldn't just suggest updating the drivers is beyond me). Since I hadn't updated graphics drivers at all between replacing the PSU and now, I feel like that shouldn't be it, but who knows.

hey, just wondering if you found the problem out because same thing is happening to me and i cant figure out what the source of the problem is, maybe bad mobo or psu?

 

[DeadSulfur]

hey, just wondering if you found the problem out because same thing is happening to me and i cant figure out what the source of the problem is, maybe bad mobo or psu?

Sorry for the delayed reply (I hadn't checked this in some time), but I still haven't figured the problem out. I had someone take a look at it, and he found that the GPU and air cooler had some packaging plastic still on them (they evidently didn't remove all of it when building the computer somehow). He removed that, and it seemed to fix the issue, since he had found by looking at bios logs that the computer was shutting off due to heat.

I don't know where these logs are found. I can't find them, but that's what he said. It didn't add up (monitoring my temps, they were still higher than I'd like but definitely below shut off), but I hoped it was the problem and stayed optimistic. A little less than a month later it shut off again. I gave it back to him, but from the looks of things he didn't really check it much. It's back in my possession, and I'm probably going to RMA it at this point. From what he found, there's a decent chance it's a faulty motherboard.

I honestly wish the problem were more consistent. At this point I just want a solution, but the damn thing is hard to debug. When it only crops up as an issue every month or so in spurts, essentially anything could end up being a false positive.