4670K low low idle volts, instability at no load?

Kanifee · 27 May 2014 at 21:20

Need some help, currently have my 4670K @4.4Ghz and in games it is rock solid, few hours of watch dogs, Dayz epoch/arma2/3 no problem maxing temps at 59c, as much prime95 as I am comfortable giving it which is around 2 hours and temps max out at 70 under water now my fan is turned up, still though I get intermittent restarts when doing menial tasks like opening a web page or YouTube videos that shouldn't be putting anything on the cpu really. This happens maybe 1 or 2 times a day when the system is under minimal to no load what so ever and the cpu is running at 800mhz!

Then I noticed looking through HWiNFO my idle vcore is fluctuating around 0.110v and 0.200v. I have watched it dip as low as 0.050v a couple of times and this has me thinking that this may be the culprit as that seems very very low even for a modern cpu?
I am at the moment running.

Multi x 44
Vcore offset + resulting in 1.320v
LLC 50%
uncore x 33
ring voltage 1.100v
vccin 1.900v
EiST enabled
C-states enabled ( have tried all of them and they just lower it over the average more than anything else)

And that is where I am now, stuck! I have no idea if this idle voltage is normal even as abnormal as it looks to me, I also have no idea how to raise it to make the system more stable at or leaving an idle state.
I would really appreciate any help you can offer, would be nice to know your idle voltages too please.

wazza300 · 28 May 2014 at 13:22

it should be stable imo,its designed to run low voltas esp on haswell

any bsod codes when it crashes? enabled xmp for your memory and running latest motherboard bios?

Kanifee · 28 May 2014 at 15:19

I will make a note when it turns up again, no xmp enabled and bios is on v.1 as it was released about 10 days ago.

Kanifee · 29 May 2014 at 10:50

Just had the random shut off again sitting reading a web page. Vcore was set to manual this time at 1.3v, multiplier on adaptive x8-44. Im starting to wonder if it may be a psu issue, I didn't get any warning or blue screen just instant power down as though the plug was just pulled.

My psu, a corsair HX520 is getting on a bit now, I purchased it in January of 2009 from the members market abd though it has been sitting power down for the last 9 months or so it could have easily had another 3-5 years use before that. As high quality as they are that is still a big ask.

The event logs in windows just show service stops as you would expect and this.

The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.

Log name: system
source: kernel power
event id: 41
level: critical
task category: (63)

- System

- Provider

[ Name] Microsoft-Windows-Kernel-Power
[ Guid] {331C3B3A-2005-44C2-AC5E-77220C37D6B4}

EventID 41

Version 3

Level 1

Task 63

Opcode 0

Keywords 0x8000000000000002

- TimeCreated

[ SystemTime] 2014-05-28T14:11:53.199526100Z

EventRecordID 3566

Correlation

- Execution

[ ProcessID] 4
[ ThreadID] 8

Channel System

Computer kanifee

- Security

[ UserID] S-1-5-18

- EventData

BugcheckCode 292
BugcheckParameter1 0x0
BugcheckParameter2 0xffffe00124236028
BugcheckParameter3 0xbf800000
BugcheckParameter4 0x124
SleepInProgress 0
PowerButtonTimestamp 0
BootAppStatus 0

Weird thing about it is if it is a psu fault the system is fine under full load. Hours of armaII and no issue, it is only at idle or near idle speeds.

I cant think of any other way to test for psu stability either.

+Edit

Looking at the voltages in aida64 and HWiNFO they show how ineffective software monitoring are, wildly different readings on both with aida being the concerning ones, would my system even power on with such a low 12v feed?

I should have really stumped for the boards with voltage read points but returning to pc's form nearly a year hiatus I didn't really intend to OC, that is until I started tweaking and getting bitten by the bug all over again! Its a great board the Z97 G55 SLI but it isn't enthusiast targeted, may have to add it to the MM.

Hagar · 29 May 2014 at 14:15

Is your PSU compatible? Some are not and bottle out at low voltages.

wazza300 · 29 May 2014 at 14:23

could be psu,could also be early bios and not mature enough yet idk

the 0x0124 error is not enough cpu v (bugcheckparameter 4)

idk on those boards but you might be able to add touch more cpu v so it doesn't drop so low? or raise the lower turbo ratio limit?