• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Curious issues and Event log

Roverno

Roverno

Roverno

Associate
Joined
24 May 2011
Posts
1,790
Location
West Sussex
Hi all,

First off, not running any overclock currently, and havn't for a while. I had notived some strange cpu issues recently with occasional crashes, and unbalanced core temps and constant boost clock.

In the windows event viewer i am getting the following issue, that repeats 14000 times to fill the event log faster than i can mash F5 to refresh.

A corrected hardware error has occurred.

Reported by component: Processor Core
Error Source: Corrected Machine Check
Error Type: Cache Hierarchy Error
Processor APIC ID: 6

The event log ID is 19.

The APIC ID changes between 0 and 7, i assume referring to cpu core.

I have tried googling but i am not had any great success from that.

Does anybody here have any idea what might be causing this issue and how i can solve it?
 
I would run the intel processor diagnostic tool.

Have you looked at updating your BIOS version and chipset drivers?
 
I have run the Intel diagnostic tool. But that didn't show anything, which I didn't expect tbh. Bios is on 12.8 iirc, which is the latest. Will double check chipset drivers when I get home this evening. Have you had the issue yourself?
 
Installed all the latest intel drivers and chipset drivers. Same issues still :-/

Any more ideas anyone?
 
Have you set your BIOS settings back to default?

*edit - seems like lots of i5-3570k get this error too. Most go away once the user removes the overclock. As you say you are not overclocking can you raise the voltage to the CPU slightly without any overclock and see if the errors disappear?
 
Last edited:
So i overclocked slightly. Set frequency to stay locked at 3.8Ghz and set the voltage to 1.2v, OCCT suggests that this is an accurate voltage reading 1.19v during loading.

Also set the Memory to the XMP profile.

Error still occurs just the same as before.
 
touch more cpu vcore should help,is it a whea uncorrectable error? if so more vcore

might need more vtt voltage aswell (memory controller) with 16gb of ram,
 
Roverno said:
First off, not running any overclock currently, and havn't for a while. I had notived some strange cpu issues recently with occasional crashes, and unbalanced core temps and constant boost clock
Click to expand...

Core temperatures could be causing the crashes? Might be worth pulling it apart and giving it some fresh TIM - or if it is fresh, maybe the application was bad.

Failing that - might be worth trying a different PSU, just in case the PSU is beginning to breakdown (depending on the PSU - how many rails etc, could be just affecting CPU voltage rather than anything else).

Worst case, given that your chip has been overclocked/overvolted, then it could be silicon degredation - although in theory giving it a bit more voltage, you should be able to get it stable.
 
Updated my signature to prevent any confusion. It was well overdue.

Upped to 1.22v, seems a lot for 3.8GHz?

Core temp, according to HW monitor, although hotter on one core, is only by 4-5 degrees. and nowhere near anything high. Only hitting 70C or so.

PSU is corsair AX 860. New less than 6 months ago.

I did buy second hand so maybe it was cared for poorly previously. Dont really feel like i can justify a full upgrade just to solve a minor cpu issue though. Was planning for the budget to go on watercooling soon.
 
what about raising memory controller voltage? no idea what its labelled as on msi boards,it will be stock 1.05v try with 1.15v (might need more of that than cpu voltage)

odd core running hotter isn't anything to worry about,i have a few cpu's that do that
 
Memory controller (VTT) seems to be called CPU IO on msi boards. In bios it was reading 1.256. The recommended maximum seems to be 1.25... The auto seem to be running overkill already.
 
any other voltage at 1.05v? seems very high to me

1.2v is max safe,what speed is your ram running at? try with a lower ram speed say 2.133mhz and manually set vtt to 1.05v then see if it still crashes

just try it for testing,can always up ram speed later on
 
1.07 is the minimum i can set CPU IO (vtt) to. if i try any lower it just changes to box to 'AUTO'. I have booted with that voltage (1.07) and 2133 MHz though. Just going to post this and will run quick stress test
 
try with it at 1.15v with 16gb and 2133mhz speed

it can crash if its too high,same as too little,and 1.25v was way too high
 
Short term stress seems to be ok. But the event manager is still creating errors like there is no tomorrow. I cleared it, and within 10 seconds it has 14000 errors.
 
2133mhz and 1.15 vtt. Stress test on LINPACK mode for a couple of minutes.

Much fewer error 19 codes, a few thousand though. Got one error 47 code, which relates to memory.
 
You must log in or register to reply here.
Back
Top Bottom