Upgraded to a FX-8302E and I'm BSODing.

MarcoPolo1667 · 20 Feb 2015 at 10:27

So after being stuck with a FX 4350 since I got this rig I thought it was time for an upgrade. Checked everything out and finally went with the FX-8320E since I didn't need a MOBO upgrade.

So I received it yesterday and installed it, turned it on and suddenly I'm BSOD every couple of min's. Fiddled with some settings and read some things online and after DOWN-CLOCKING from the base clock of 3.2 down to 2.8 it seems to have stopped the BSOD.

But surely that shouldn't have to happen? Surely I should be able to push this cpu up past its base clock?

Here's my spec's:

CPU: AMD FX-8320E
MOBO: Asus M5A78L-M LX
GPU: R9 270X 4GB
PSU: 650W

The BSOD error is this: clock interrupt secondary processor.

joeyjojo · 20 Feb 2015 at 10:29

Bios update and check your cooler is up to the task (what is it?) first IMO.

Edit: BIOS 1603 according to http://www.asus.com/Motherboards/M5A78LM_LX/HelpDesk_CPU/

cheesyboy · 20 Feb 2015 at 10:32

Are you on the latest bios?

(interestingly, the FX-4350 isn't on the support list for the board at all. THe 8320E is, though, so should be fine)

It's not the best board for the high TDP FX chips, so instability could be related to VRMs struggling. 8320E is a 95W job, so should be ok, but might be worth pointing a fan at the VRM area to see if that helps.

MarcoPolo1667 · 20 Feb 2015 at 10:34

joeyjojo said:
Bios update and check your cooler is up to the task (what is it?) first IMO.

Edit: BIOS 1603 according to http://www.asus.com/Motherboards/M5A78LM_LX/HelpDesk_CPU/

Yep my BIOS is updated to 1603. And the cooler is the stock fan which came with it.

Just to add, BSOD even happens on boot. That was until I under-clocked it now it seems to be working fine as of yet on 2.8

- Edit -

And I know, the rig was built by YoYo Tech and I only noticed that the cpu wasn't supported a couple months later. So before buying I checked to make sure this cpu was supported and now I'm getting issues, it makes no sense. :\

VincentHanna · 20 Feb 2015 at 10:59

I would be looking to replace your cooler ASAP.

Scurburg · 20 Feb 2015 at 11:02

What sort of temps are you getting? (should display in bios)
Whats vcore set to? Maybe worth trying to up it a little to see if you can achieve stock, but that only if temps permit.
As others have said, a new cooler is highly recommended for these chips. The stock one is pretty poor.
Also what speed is your ram rated and running at and what voltage?

MarcoPolo1667 · 20 Feb 2015 at 11:08

Scurburg said:
What sort of temps are you getting? (should display in bios)
Whats vcore set to? Maybe worth trying to up it a little to see if you can achieve stock, but that only if temps permit.
As others have said, a new cooler is highly recommended for these chips. The stock one is pretty poor.
Also what speed is your ram rated and running at and what voltage?

Currently with TS3 and 3 tabs of Chrome open my CPU temp is sitting at 20-23C.

And to be honest, I'm kinda noobie when it comes to this stuff, so I have no idea where to find those setting.

cheesyboy · 20 Feb 2015 at 11:12

It's very unlikely to be a problem related to using the stock fan. Those things are crap, but they'll cool the CPU fine at stock settings - as long as they're correctly fitted.

It's most likely to be either a duff CPU, or a mobo issue.

Can you test the CPU elsewhere?

MarcoPolo1667 · 20 Feb 2015 at 11:16

cheesyboy said:
It's very unlikely to be a problem related to using the stock fan. Those things are crap, but they'll cool the CPU fine at stock settings - as long as they're correctly fitted.

It's most likely to be either a duff CPU, or a mobo issue.

Can you test the CPU elsewhere?

Sadly not. My old PC died on me and I don't have a spare any where. What's my best bet here? Buy a new MOBO or getting a refund from Scan on the cpu?

cheesyboy · 20 Feb 2015 at 11:24

Scan aren't great for returns. They'll charge a testing fee if you return as faulty (if they find no fault), and I suspect they'll knock off a fee for opening the package if you return under DSR (or the new DSR-replacing rules).

That said, if it is actually faulty then a new Mobo won't help.

Try cooling the VRMs with an extra fan to see if that helps. If not, I reckon a return of the CPU is the next thing to do.

Scurburg · 20 Feb 2015 at 11:36

MarcoPolo1667 said:
Currently with TS3 and 3 tabs of Chrome open my CPU temp is sitting at 20-23C.

And to be honest, I'm kinda noobie when it comes to this stuff, so I have no idea where to find those setting.

Well we all have to start somewhere.

Since I am unfamiliar with your boards bios it makes it a little difficult, but we can try if you are willing to delve. Just a quick question, Did you run a fresh install of windows and which windows version are you using?

Although your temps appear to be fine at low use, its hard to say what they will be like under heavy/full load (I believe there to be a cap at 72c). I'm not sure what you have running would really stress the CPU to find out either. We will probably need a few progs to find out some details of whats going on with regards to temps, load and voltage. If you can find these:
cpuID
hwmonitor
Prime95 (dont run yet, this is to stress the CPU for max temp)

I usually use majorgeeks.com to download them.

Once you have these, please try and run CPUID first. There should be a voltage reading, so if you can let us know what that is to start with.

MarcoPolo1667 · 20 Feb 2015 at 11:41

Scurburg said:
Well we all have to start somewhere.

Since I am unfamiliar with your boards bios it makes it a little difficult, but we can try if you are willing to delve. Just a quick question, Did you run a fresh install of windows and which windows version are you using?

Although your temps appear to be fine at low use, its hard to say what they will be like under heavy/full load (I believe there to be a cap at 72c). I'm not sure what you have running would really stress the CPU to find out either. We will probably need a few progs to find out some details of whats going on with regards to temps, load and voltage. If you can find these:
cpuID
hwmonitor
Prime95 (dont run yet, this is to stress the CPU for max temp)

I usually use majorgeeks.com to download them.

Once you have these, please try and run CPUID first. There should be a voltage reading, so if you can let us know what that is to start with.

OK, I'll download those programs now. And to answer your question... I didn't do a fresh install of my OS which is Windows 7 Home Premium 64 bit. (Mainly because YoYo Tech didn't send me a disc for it with my build)

Here's the stats from cpuID:

cheesyboy · 20 Feb 2015 at 11:54

Big temperature recording on TMPIN2

No idea if that's reliable, but if it's getting recorded at that level, then it could be causing issues.

E:
that's most probably not accurate;
http://linustechtips.com/main/topic/68967-possibly-a-weird-temperature-in-hwmonior/

Scurburg · 20 Feb 2015 at 11:58

That is HWmonitor, but thats ok, we can see some details now. TMP0 is likely to be your CPU/Socket temp, the package temps are for the CORE temps, generally CORE temp will be lower than Socket, so all looks ok there.

I do have a concern though, you +12V is well below tolerance. I believe 11.75 is lowest and yours is well below that, and I would expect the system to be very unstable with those readings.. IT could just be that HWmon is reading it incorrectly though. Which Power Supply are you using?

"Big temperature recording on TMPIN2"
Since it hasnt moved from 128, I think its a redundant sensor, although I could be wrong.

Hagar · 20 Feb 2015 at 11:59

Your voltages are all reading very low 3.3V 5V 12V, PSU issue?

EDIT I get a TMPIN2 high readout on my crosshair (ASUS issue).

MarcoPolo1667 · 20 Feb 2015 at 12:07

All I know it's a 650W PSU. I really don't want to turn my computer off just incase I can't turn the flipping thing back on without BSOD.

And since I haven't said it yet, I want to thank everyone for helping me out here and dealing with my lack of knowledge on the issue. Cheers.

mmj_uk · 20 Feb 2015 at 12:11

MarcoPolo1667 said:
The BSOD error is this: clock interrupt secondary processor.

That's often what you get from failed overclocks so it could be the CPU is just not stable on default voltage, either because the CPU is not meeting spec or because your motherboard is low end.

Does upping the CPU voltage a little help to reduce crashing?

Scurburg · 20 Feb 2015 at 12:17

The PSU could be the cause as cheap ones are not always as good as they are rated. Does it have a brand name on the back or any details about total amp?

MarcoPolo1667 · 20 Feb 2015 at 12:17

mmj_uk said:
That's often what you get from failed overclocks so it could be the CPU is just not stable on default voltage, either because the CPU is not meeting spec or because your motherboard is low end.

Does upping the CPU voltage a little help to reduce crashing?

Judging from the responses so far it looks like my voltage is to blame here. I have no idea how to change it and what to change it to.

Scurburg · 20 Feb 2015 at 12:31

I dont think its your voltage, IMO its your PSU not supplying enough power, in which case upping the voltage will actually make it worse.

If you really want to try though, you will need to reboot and press del to enter bios. It should be under something like advanced settings, we are looking for vcore,
It will likely be set to auto. You either need to set it to +0.1 for offset mode (To bring your vcore up to about 1.2) or set it manually to 1.2 (dont put this for offset what ever you do)