It looks like you're new here. If you want to get involved, click one of these buttons!
Hey all,
Wondering if I could get some advice about a problem that has been annoying me for the last 6 months or so, which causes my PC to sometimes boot into recovery mode. Basically sometimes when I turn my PC on it'll boot into recovery mode, Windows will perform it's various checks to ensure that there isn't any problems with the PC and everytime it comes back stating that no problems where found/fixed. I used to allow recovery to complete every time, but now I'm sick of it happening and I usually force a restart the moment recovery has loaded. It's totally random when it occurs and I've been unable to lock it down to any particular scenario. I always shut down my PC properly and I have tried reinstalling windows 7 (home premium/professional [legit copies]) in the past and it still occurs. I'm wondering if it is an issue with my MB or SSD, but I've been unable to find a record of the problem happening for anyone else with these components. I reiterate that I now don't let recovery complete, I just force a restart and the PC starts perfectly normally after. I can confirm that this never happens from restart, only when the PC is cold booted.
My PC specs are as follows:
Case: Corsair 550D
PSU: Seasonic X760W Gold 80+
MB: MSI P67A-GD53
Memory: 8GB Corsair Vengeance DDR3 1600Mhz
HD: 120GB Corsair Force 3 SATA III
Seagate 1TB Hard Drive SATAIII 7200rpm 64MB Cache
CPU: Intel Core i7 2600k (professionally OC'd to 4.5Ghz)
Cooler: Corsair H100
Graphics: ASUS 570GTX Direct CUII
OS: Windows 7 Professional
If anyone can shed any light/offer a solution, it would be greatly appreciated.
Comments
I'd run a few 'stress test' programs and see if the issue can be replicated.
Prime95 is an open source program that'll 'torture' your PC's CPU / RAM stability. Run it for 24 hours in 'Torture Test' mode and see how well it does.
http://en.wikipedia.org/wiki/Prime95
I'd also run memtest86 to check your RAM.
Gdemami -
Informing people about your thoughts and impressions is not a review, it's a blog.
That's normal If your definition of cold boot is to turn the PC off the "incorrect way"
I can confirm that I always shut the PC down the 'correct way' I have run several stress test programs with no problem. The PC is absolutely fine when started up, no problems reported, the only issue is when the PC is cold booted after a correct shut down.
This sounds like your CMOS battery is not charging properly.
Prime95 is unique in that it'll 'stress' components way harder than any other stress test programs out there.
Including that god-awful 3D-something benchmarking one.
Gdemami -
Informing people about your thoughts and impressions is not a review, it's a blog.
Battery fault
Hard disk fault
RAM fault
Less likely
Case wiring fault
Overheating
When last did you clean out your pc?
Huh?
My pc works fine
I would agree with the overclocking - the RAM and CPU can't take it. What does it mean to have it 'professionally overclocked'? If it wasn't overclocked from the manufacturer - then there is no such thing as 'professionally overclocked'.
Hey guys,
Thanks for all the advice It's definately not an overheating problem, I have plenty of heat monitoring and everything is running nice and cool, I have a very good cooling setup and ambient temperature is always low (living in the UK). As I stated before I'm 90% confident that the problem is with the MB or SSD (I have recently replaced the storage drive and it is still happening). I don't skimpt on components and I had the PC OC'd by www.scan.co.uk who ensure that the OC is stable over at least a 24 hour burn in period so I don't question the OC itself (4.5GHz OC is pretty tame when using WC on a 2600K anyway).
Guess I'm back to the drawing board short of replacing the MB, which I'm loathe to do at this stage. Maybe Quizzical can shed some light on the problem later?
Just because yoour overclock is stable doesn't mean that some programs work well with overclocked CPU's. I have seen this in the past.
Run your PC at normal clock speed and see if it does the same thing. It it doesn't, then it is your overclocking that is the issue.
Not always and this is the key. THat is why you should check it that way. Just my $0.02.
I have done some major overclocking in the past - a 600 MHz unlock AMD Duron (OK dating myself here) overclocked it to 2 GHz but would eventually overheat. It idi have issues with boot up when cold not rebooting.
Basicly you buy quality RAM, once you've had it a few months its failure rate will be low compared to cheaper ram.
But...
During the first couple of months a certain % of electronic components always fail, REGARDLESS of build quality. The high end gaming ram you bought has the same % failure rate when new as the rubish chain stores still in their pcs. It's only once its bedded in that the quality shows.
So do do a ram check, especially if its new ram.
I think your motherboard battery is more likely culprit though.
Overclocks are finicky things: just because it was verified by a 24 hour burn in when the computer was first built, doesn't mean it could do so at the same clock speeds and voltages today.
Heat degrades silicon. The normal expected lifespan of a silicon chip is several years (probably 15+ years). However, when you overclock, it runs hotter. Even if your monitored temperatures are nice and cool, it's still running hotter than it did when it was at stock. That, and your monitored temperatures aren't necessarily measuring the direct and exact spot where it's getting hotter... the internal CPU thermocouples are near, but aren't directly on top of the processing units.
So overclocking will lower the lifespan of your chips - that isn't any surprise. Sure, it's lowing it from maybe 15+ years to 10+ years - not a huge deal. But here's where the fine line comes in.
An overclock your already pushing your hardware near it's design limits. As the chip degrades over time, those limits change as well. That means an overclock that was possible when the unit was brand new, may not be possible 6 months later, even if nothing else at all has changed.
You need to take all your clocks back to stock and see if the problem then goes away. I would bet 10:1 that it's your overclock causing the problem, even though it passed a burn in test when it was first made. Overclocks, especially aggressive ones, aren't something you can just "set and forget".
I saw almost the exact same problem with a computer once: stock it ran perfectly. OCed it would pass benchmarks and stress tests fine; it just never quite shut down correctly and would always start up saying that it needed to chkdsk (although it had no problem rebooting - no clue what the difference was). I told the guy to either drop the overclock, or just ignore the startup warnings. He chose to ignore the startup warnings. PC ran fine for as long as he had it (4ish years iirc).