Is my PSU failing?

[Se213]

First of all my specs are:
AMD FX-6300
CPU Cooler: Rosewill RCX-Z1
GeForce GTX 950
CoolerMaster V650
G-Skill ripjaws 8gb
Gigabyte GA-960A-UDP (rev.2.x)
I've been getting complete lockups in games and sometimes shutdowns and my 3.3v readings in hwinfo show 2.04v. Is it time for a new PSU, MOBO or what ? I don't know what to make of these numbers..
https://gyazo.com/703dd5a99b955e92263958dda6bdb7d3

 

[rolli59]

hwinfo has been known to be wrong and it can really be anything that is causing the lockups but are you getting BSOD's with those shutdowns?

 

[Se213]

hwinfo has been known to be wrong and it can really be anything that is causing the lockups but are you getting BSOD's with those shutdowns?

No, what happens is it mostly freezes as in I can't do anything but restart it with the case, no audio loop or anything but sometimes it just turns off as if it was unplugged from the wall. I just checked in my BIOS and it said +3.3v- 3.067, +5v- 4.800, 12v-11.880. Also when I turn up my gpu and cpu fans in games it freezes a lot less and sometimes not at all.

 

[Sedivy]

Well that 3.3V is not within the 5% tolerance, true, but not sure how reliable hwinfo's readings of it are. I'd recommend a voltmeter and doing it manually.
Otherwise, you can try and find another computer with a compatible psu in the similar voltage range and swap it in, see if it still happens.
While that could be psu, it could be other things too. Freezes often come with memory trouble and since this happens during games, it could also be your cpu or gpu under stress.
You could try memtest, prime95 and furmark to test them out individually to rule them out and see what gives you lockups. Keep in mind memtest will require at least 8 passes and even then you're not certain it isn't bad. It could also be your drivers, windows, or the game installation itself so just from that it's hard to tell. What are your temps at?

 

[rolli59]

If this is only happening when gaming then the GPU could be in question so trying it in another system would confirm either way.

 

[Karadjgne]

You should see my pc. It doesn't run anything 12v. Asus suite reads 10.2v, SpeedFan reads 8.2v, even bios reads 10.2v, hwinfo reads 8.2v, Hwmonitor reads 10.2v. And yet the trusty fluke meter reads 12.18v at pinouts on the mobo.

Seriously doubt the pc would boot without the 3.3v being close to spec at least as the cpu uses 3.3v, as does the ram. Even your gpu has a small amount of 3.3v usage.

 

[Karadjgne]

Lockups during gaming are primarily caused by audio drivers. Win10CE has some issues with legacy drivers, especially the Realtek ones from that generation mobo and prior. Has to do with driver inability to accept 64bit processing. I'd start with a trip to Gigabyte website and install any and all last dated mobo drivers, especially audio and Lan. It might also mean a trip over to Realtek website to get further updated drivers as Realtek will most definitely include any of the latest patches.

 

[Se213]

Lockups during gaming are primarily caused by audio drivers. Win10CE has some issues with legacy drivers, especially the Realtek ones from that generation mobo and prior. Has to do with driver inability to accept 64bit processing. I'd start with a trip to Gigabyte website and install any and all last dated mobo drivers, especially audio and Lan. It might also mean a trip over to Realtek website to get further updated drivers as Realtek will most definitely include any of the latest patches.

I'm getting LAN and audio drivers now but I turned both my GPU and CPU fans up just out of curiosity and the freezes happen less often when the fans are working harder, idk what that could mean

 

[Sedivy]

Well that usually means temperature issues but they're not obvious on the screenshot you posted so you should check where your temps are at while gaming.

 

[Karadjgne]

There's 2 sections to any video card. The graphics processor (gpu) and the voltage regulatory circuitry and ram. Of the 2 sections, only the gpu actually has a temp sensor, so while the gpu might read 60°,the VRM's / ram might be hitting 90° and you'll have no warning. Overheating ram will definitely cause issues with freezing, and overheating VRM's will cause gpu slowdowns, basically the same as the voltage circuitry on a motherboard will give a cpu fits.
The cure is airflow. The gpu needs it in copious amounts. Clean heatsinks and fast running fans combined with a good supply from pc intake fans.

 

[jr9]

- I don't recommend trying to use HWInfo with AMD processors, it will give you bad readings. I also don't recommend trusting it for voltage readings either they are often wrong. If you actually had 2V on your 3.3V your PC would not be running.

- Your issue could be a lot of things really. If you have another PSU I would try swapping that in first. If you do not have access to one, I would have a shop perform a diagnostic rather than playing the "buy a part and hope it fixes the issue" game.

- Giving us your max CPU and GPU temps would be very helpful. Furmark is good for GPU and has a CPU stress tool in it as well. CoreTemp is good for CPU temps.

- Bad motherboard can also cause this. AM3+ boards are known to have electrical issues under load.

 

[Se213]

- I don't recommend trying to use HWInfo with AMD processors, it will give you bad readings. I also don't recommend trusting it for voltage readings either they are often wrong. If you actually had 2V on your 3.3V your PC would not be running.

- Your issue could be a lot of things really. If you have another PSU I would try swapping that in first. If you do not have access to one, I would have a shop perform a diagnostic rather than playing the "buy a part and hope it fixes the issue" game.

- Giving us your max CPU and GPU temps would be very helpful. Furmark is good for GPU and has a CPU stress tool in it as well. CoreTemp is good for CPU temps.

- Bad motherboard can also cause this. AM3+ boards are known to have electrical issues under load.

I updated drivers and everything and it's still causing issues, it's worse now, it has froze 3 times in a row at the fortnite main menu with all fans on 100% and I checked to make sure they were actually being controlled by the software and they all were. A while back my CPU fried in this PC due to a bad USB 3.0 header and a new CPU fixed that but maybe the motherboard is just bad, I'm not sure how to test it without taking it somewhere, maybe I could buy a new one since they aren't that much, my CPU and GPU temps are very low and like I said even with all fans at 100% it still freezes, CPU tops out at about 50 and GPU was topping out at 65 which is because of the hard curve I had in afterburner. It seemed to ease up the harder I ran the fans but now even that isn't helping.

 

[Sedivy]

Ok if you open event viewer in windows and check for critical events within a few minutes of the time of the crash, what does the event ID say and the description?
You could take it to a repair shop and they can test it with other components to see if they can reproduce the issue, but not sure new board wouldn't cost about as much as diagnosing your old one.

 

[Se213]

Ok if you open event viewer in windows and check for critical events within a few minutes of the time of the crash, what does the event ID say and the description?
You could take it to a repair shop and they can test it with other components to see if they can reproduce the issue, but not sure new board wouldn't cost about as much as diagnosing your old one.

I've been monitoring event viewer and the only error I have is a kernal power error which just comes from restarting the PC by the button on the case. The description says something along the lines of system shut down unexpectedly.

 

[Se213]

UPDATE: I used OCCT to test my CPU and GPU and the CPU: Linpack and CPU: OCCT tests ran fine but the GPU test froze my PC the same way it does in game within 5 seconds. I then ran the PSU test and the entire PC shut down like I unplugged it the second I hit go, any ideas?? I think either PSU or MOBO but idk how to know for sure without another one to interchange which I don't have...

 

[jr9]

Windows Event Viewer will not help you in this case. It is only useful for software and OS level issues. You have a hardware issue.

If you don't have spare parts or access to any, take it to a shop and have them run a diagnostic. It is most likely either have a failing power supply or motherboard. I feel it's about a 60/40 chance in favor of failing PSU. If your PC ever shuts off like that then it generall can only be one of those two things. Other failing hardware generally yields blue screens or resets.

 

[Sedivy]

You don't mention kernel event id but probably 41 (63) which is hardware (or driver) failure.
I would say psu or gpu. But since gpu test only freezes you and gpu+cpu (which is their psu test) crashes you, likely it's your psu. Which is odd really as that's top rated psu but maybe you had bad wiring in your place or power surges or even heat issues that helped fry it prematurely. To make sure, best put in another similarly ranged psu in terms of wattage and check if you have the same problem.

 

[Se213]

You don't mention kernel event id but probably 41 (63) which is hardware (or driver) failure.
I would say psu or gpu. But since gpu test only freezes you and gpu+cpu (which is their psu test) crashes you, likely it's your psu. Which is odd really as that's top rated psu but maybe you had bad wiring in your place or power surges or even heat issues that helped fry it prematurely. To make sure, best put in another similarly ranged psu in terms of wattage and check if you have the same problem.

It is a 650W and the one I was going to order was a 550W, is that enough within range? It appears to be enough for my system but I'm trying to save money since I'm a poor college student

 

[jr9]

- If you're trying to save money then don't gamble on what parts will fix your problem. It could just as easily be your motherboard as it's an old AM3 board.

- It's impossible for a driver problem to cause hard shutdowns.

 

[Sedivy]

I wouldn't order anything at the moment. use a friend's computer that has a similar wattage psu or someone in family. Don't waste money until you're certain it's psu as it could be other things as well.
As for moneysaving, scrimping on psu is a quick way to decrease the lifetime of your other more expensive components so I wouldn't.
Personally I like this list http://www.tomshardware.com/forum/id-2547993/psu-tier-list.html and if you're money conscious go with something from tier two. Stick with 650W. Also, surge protector might not be a bad idea.

 

[Karadjgne]

Driver issues can cause hard shutdowns. Depends on the driver. There's more than a few drivers intimately tied to the cpu, voltage controls, pcie etc any of which can give ntkernal fits, and you get instant shutdown, bsod, reboot etc.

 

[jr9]

BSODs, resets, freezes, blackscreens yes for sure, but I've never seen driver issues in the field that caused complete instant shutdowns. It's generally BSOD if it's drivers (90% of the time it's display drivers) or blackscreens resets into reboots if it's a failing GPU. Instant shutdowns as if the plug was pulled in my book generally indicate a electrical problem.

 

[Se213]

You don't mention kernel event id but probably 41 (63) which is hardware (or driver) failure.
I would say psu or gpu. But since gpu test only freezes you and gpu+cpu (which is their psu test) crashes you, likely it's your psu. Which is odd really as that's top rated psu but maybe you had bad wiring in your place or power surges or even heat issues that helped fry it prematurely. To make sure, best put in another similarly ranged psu in terms of wattage and check if you have the same problem.

I went to BestBuy and got a power supply and the GPU test that used to freeze within 5 seconds ran for 15 mins fine and the PSU test that crashed it within 1 second also ran for 15 so I think my power supply was the issue, thanks to everyone that helped me it's a shame I have to pick one answer.

 

[Sedivy]

Good to hear. Strongly recommend you get a surge protector at some point to plug your pc and monitor into. Often even good psus fry because of power surges and bad wiring so to spare your shiny new psu going the way of the old one, a simple power bar with built in surge protector can be a lifesaver.