• 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.

i7 3770k - blue screens and problems, stuff!

m0nstr0s1ty

m0nstr0s1ty

m0nstr0s1ty

Associate
Joined
31 Oct 2011
Posts
15
Hi all,

So, I don't post here that often. Having just read over a few other topics, I should, because some stuff on here is really quite useful. Generally I post here because I need help. I'm an IT pro. of 18 years, but sometimes, we all just need some other people's advice / input. The bottom line, is that's what this topic is about. I need some help from anyone that can offer it.

I'll outline my issue.

I have an i7 3770k, which I purchased in November (from Overclockers), and until 3 weeks ago was working fine. Then, I started experiencing blue screens. Many blue screens. In troubleshooting, at first I started with the simple things; re-installing Windows, updating BIOS, drivers, etc. None of which even resolved my issue, in fact, it began to get worse.

The blue screens (of which I have records but will not post as there are so many) vary so much that I was struggling to pinpoint where to start troubleshooting; the blue screens related to both hardware (such as memory management) and software (driver errors etc.).

I also checked the Event viewer and under the System log, there were warnings regarding the processor threads timing out (ID 19). Google wasn't much help for that one as most people said "ignore it!". Having not experienced that problem before.. I did ignore it.

In the process of troubleshooting, I started with the software/driver related blue screens first. I stripped down my machine and rebuilt it without connecting/installing any removable devices (with exception of keyboard, mouse etc) and using a single SSD and blu-ray optical drive. No other hardware connected. I re-installed Windows probably around 15 times - Windows installed fine, no problems - but when it came to installing drivers, or even once Windows was installed (thus during general running), I would start getting blue screens. I created restore points after every driver issue to see if I could pinpoint one in particular causing an issue, but it seemed to be completely random. No joy.

I then moved on to testing hardware. Fortunately I have so many builds that the spares I have are compatible in various set-ups so I could test most hardware. I tested hard drives (both mechanical and SSD), memory, graphics cards, wireless cards, optical drives, water and air coolers. Nothing made a difference, I'd still get blue screens. So I ruled out any hardware issues, with the exception of the possibility that it could be the processor.

So, I attempted to test the 3770k using Prime95. As soon as I clicked the stress test button, Prime95 crashed. On one occasion I had an instant blue screen. I tested multiple times to no avail. I guessed it had to be the processor. I was confident there was no physical damage; no heat damage, and it hadn't been overclocked. So I phoned Overclockers and RMA'd the CPU.

I had an email from Overclockers this morning - they ran a test on the CPU for 30 hours in Prime95 with no errors. So, it's not the processor.

Here's where it gets really confusing. On sending back the RMA, I swapped out the processor with a Celeron G1610. I haven't had one blue screen since installing it. No problems whatsoever.

I explained that fact to Overclockers Tech; but as far as they are concerned the (3770k) processor isn't faulty (which is fine; no complaints there). They explained that they couldn't support the rest of my system (fair enough, even though all my parts for the last 10 years are purchased from Overclockers; a minor issue). I am assuming that Overclockers didn't want to risk giving false information without diagnosing the system as a whole, good on them. :)

They advised that I take my PC into a local tech. lab to see what they make of the issue. I have no problem with doing that, but before I do.. anyone got any ideas?! I am at my wit's end. All the parts I (should) be using are under 6 months old. It just doesn't make sense. I can't RMA literally every component as it will cost a fortune.

Being an IT pro is clearly no help here, it's not a "standard" problem, so that's why I am begging for some help!

So, here's the full spec. of the system:

i7 3770k (currently using G1610)
ASUS Maximus Extreme V
Coolermaster Seidon 120 (currently using Akasa Voodoo Venom)
Corsair Vengeance 1800mhz (4x 4gb dual channel, matched pairs) but currently using Corsair Vengeance 1600mhz Low Profile (2x 4gb matched pair)
OCZ ZX 1kw PSU
Samsung EVO 500gb
WD Black (1TB x2, 2TB)
akasa internal card reader
LG blu-ray optical drive
Geforce 680 SLi (currently using single Geforce 570HD)

Any help anyone can give would be very much appreciated.

There is one thing I can think of which occurred as I was typing this out: the problem, clearly, lies within the processor infrastructure; but neither processor is faulty, yet one processor works and the other doesn't. The only difference I can think of between the two, is that the G1610 doesn't need to draw power from the 8-pin CPU power socket, yet, I assume the i7 3770k would definitely need to do that. Possibly the next place to look?

Please help. :(
 
Last edited:
windows 7 or 8? did you manage to catch the bsod code when it crashed?

it'll look something like this with various numbers 0x00000124

look in event viewer in the warnings/errors and click on some to see what it says

with 16gb of memory you need to enable x.m.p memory profile if it supports it in the bios then try more vccio/vccsa voltage from stock 1.05 to around 1.10v to 1.15v

it could also be psu idk yet but id try memory first,could also run memtest86 software and see if any errors crop up

overclocked or stock it should run and be stable with either just the 4 pin or 8pin psu/cpu leads

also test with the akasa mem card reader disconnected just to rule that out
 
I had a read through, but have you tried the system with a different PSU?

Also is the cpu socket void of any bent pins.
 
Hi - thanks for the quick replies.

Incidentally, I have had quite a few of the error codes in the link of your sig., wazza. I'm sure you know what it can be like trying to diagnose a blue screen - but the information dropped on that list is most useful. I'll definitely use it to test. Also, I am using Windows 7.

In regard to using 16gb memory, the BIOS was set to run the memory at 1866mhz (which is the module speed, not overclocked) and the timings which I have set manually. I didn't adjust any XMP settings; but the BIOS does report that the memory profile is picked up as being an XMP profile, if that makes any difference. I'll have a look at that. I'll also check the voltage.

I didn't mention that I have used individual sticks of RAM with the G1610 and I don't get any errors while using memtest. Apologies, I typed a lot and was bound to miss something.!

Fortunately I do have a spare PSU, it's not in great condition but I think it's still in working order.

There is no damage to the socket on the mobo - that I was very cautious and am quite sure of!

I get the 3770k back from Overclockers tomorrow, so I will test the above to see what happens. The possiblity that I may have missed something, given what you've suggested, is more likely; so when I've tested I'll report back.
 
RAM voltage manually set too? I have 4x4GB of Vengeance (though mine is 1600MHz XMP) and found that 24/7 stability was only reached once I manually set the RAM volts to 1.50. This is on a B3 P67A MSI board though.

I have a 3770k too btw on that setup.
 
Last edited:
Have you tried pulling a couple of your sticks of ram?

My 3770k and z77 sabertooth has a similar strange issue thats developed over 18 months when first booted it fails all mem tests causes random blue screens and sometimes fails to post at all. If however I leave it sat in the bios for half hour to warm up then bam its fine and will be stable indefinitely as long as its not powered down, Ive put it down to full memory controller and the ram not being QVL certed for the saber but it is incredibly annoying, Ive tried upping volts on mem, vcore, vcssa and the only thing the helps is either pulling 2 sticks or running them at 1333mhz
 
I had this, pulled the RAM get a white pencil rubber and rub along the contacts voila! not a bluey in sight for 2 months since, you get black deposits on the contacts which are a PITA
 
you my man are a genius if your ever in bournemouth i owe you a beer rushed home and done exactly that and its touch wood booted and stable first time
 
Last edited:
Hi all,

I said I'd report back once I had done more testing; the good news is that I think I have fixed the issue.. but I have no idea how.

I put the CPU I received back from Overclockers (reminder: RMA tested as not faulty) in the PC and instant blue screens at the Windows splash screen. So clearly something was up.

So, I decided that I would start from scratch. Literally. I pulled the PC apart, removed the components, dropped everything back to factory settings as much as I could; and began to rebuild.

I got everything rebuilt.. with all the components I wanted to use; I set up the BIOS settings, and it seems to be working fine so far; no event viewer errors, no blue screens, no freezing, no crashing.. so now I am going to try and run Prime 95 to see how it handles.

It's unfortunate that I didn't pinpoint the exact issue before I rebuilt, but I just got so sick of it.. I thought I'd treat it as a new build, and fortunately it worked!

Thanks for your help. :)
 
stuie said:
Have you tried pulling a couple of your sticks of ram?

My 3770k and z77 sabertooth has a similar strange issue thats developed over 18 months when first booted it fails all mem tests causes random blue screens and sometimes fails to post at all. If however I leave it sat in the bios for half hour to warm up then bam its fine and will be stable indefinitely as long as its not powered down, Ive put it down to full memory controller and the ram not being QVL certed for the saber but it is incredibly annoying, Ive tried upping volts on mem, vcore, vcssa and the only thing the helps is either pulling 2 sticks or running them at 1333mhz
Click to expand...

Also, that is an issue that I had before my other issues started. When I get my RMA'd watercooling system back, I'll definitely be testing that out with my 4 sticks of Vengeance!

gug54321 - will definitely use that information to test the above problem if I have it again. That is legendary.
 
You must log in or register to reply here.
Back
Top Bottom