Freezing after stress testing
 

Hey guys I am dumbfounded right now. I can pass say prime95, but the damn PC freezes when I stop the test. I don't have any screenies because I can't do shit. I'm at 4700mhz and have passed stress testing at 1.35-1.40 trying higher vcore to fix this. I've tried ram from 1333mhz to 2133. Same shit. VTT has been changed. Realtemp, cpuz and anything else open will just "stop responding" and I can't open anything else nor restart. I have to hold the power button.


2600k 100 x 47
P67A-UD7 BIOS F3h JZ
2 x 2GB Corsair Dom 1866 C9
Asus 580 Direct CUII
C300 64GB x 2 @ R0
Samsung F3 1TB x 2 @ R0
Samsung F4 2TB x 1
Corsair AX850w
Win7 64-bit w/ SP1

same thing with LinX ?

temps decent ?


First thing I would try would be some different stix if above is 2 yes's

Maybe your os is shot to shit? Or your stick is baddeeeee.

Could be the OS...

Check ICH temps after stress testing it shouldnt rely on the ssds, but ICHs get worked hard with high speed SSDs especially in RAID0.. wouldnt think the stress testing would hit them but...

Also LLC maybe set to high?

Good point on ich, mine gets warm evwn with the room being frozen but i dont have a fan on it.

Yes with LinX but not near as bad. My hottest core temps are 73c. I very well could be ram, but it was chugging along at 2133 7-9-7 last time I had turned this rig on.

Could be. I'll try uninstalling SP1 laterz.

LLC is set at lvl 1 so volts should be right around where they are set in bios. I haven't upped the ich volts but I did think about it. Then I didn't because I couldn't think of what would stress it. I'll try that for sure. How big of a bump usually? I do have all the ports filled up and two R0 arrays.

I've tried up to 1.14 PCH volts. I'm going to try updating the firmware to the 007 for my SSD's before moving any further.

I highly doubt it's the SSD's.
OS most likely.

OS is my next step. I mean prime will keep running and show it passing tests if I don't touch it. After an OS change who knows. Guess ram slots? I'm just trying to go from basic to hard in order without skipping steps because with my luck, the step I skip is the one that fixes it. It just blows my mind since it will continue to run/pass prime95 as long as I don't touch it.

some type of voltage spike or something when it stops?

Hell if I know, but the boot up is faster with the new firmware. What I'm experiencing is what the new firmware is supposed to fix. Well not to a T, but something like it. Freezing and slow response and such.

As far as a voltage spike... idk man. I mean at first I was thinking the same thing because I only noticed it when I stopped the test. Then I noticed if I let it run long enough that it would pull the same stunt while running. Running prime blend now to see what it does. I'm thinking my next steps are:

1. Fresh BIOS flash
2. New OS
3. Swapping RAM slots
4. Switch RAM
5. I don't know....

Hardlock?
Hardlocks are usually heat or power related.

Eventually yes but it starts first with me trying to make say realtemp the active window. Then it will "stop responding". After that I will hard lock immediately or just be able to move the mouse but can't open any programs or windows. If it doesn't hard lock I will see the little spinning thing most of the time like it's trying to do something.

Got it. I understand now. That happens to me sometimes if I lean on a system too hard for too long benching. It's a heat issue usually, in my circumstances anyway.

re-timming in a few.

Ok I tried a run at stock clocks for giggles and its doing the same thing. :( I'm guessing there is some software issue?? Guess I will skip down to OS and go ahead with a tim job. This is agrivating as hell!

You on a c300?

Ya two in R0

I am still on firmware 2 and not one problem. Dunno if you can flash back to it though. I had all types of problems with my drive on my 1366 Asus platform but haven't had those problems on 1155, 1156 or AM3 platform. I also had problems getting my 2600k stable. I could get it to pass any stress test I threw at it but then it would crash periodically while cruising the web, burning a dvd and what not. Trying to remember what I did. I messed with some voltage. Think it was the pch or vccio. I do know that I ended up putting it on auto and the problem went away. Let me plug in a USB keyboard and look at the BIOS to see what I jacked with.

EDIT: It was the PCH voltage. Dunno if this will help.

you may already answered this question and i may of just missed it but can you move the mouse or it just 100% locks up dose nothing?

i'd say try jaw down all the settings and just run everything 100% stock and then a new os with just a stander sata hdd if you have one or ide just to see what's up

if that dosn't lock up at all then you start one by one put the settings back
it's gonna take FOREVER but that's the best way doing it to me at least

Well my SSD's came with 006 and Crucial says you can't back flash to the older firmware. I just wrapped up the re-install of the OS and proggies. I have a few ideas of things to test out with this fresh install with the first being testing without SP1 installed. Plus I'm going to test P95 on a HDD and SSD to see if that makes a difference. I tried all kinds of PCH volts with no help. Hell I was having this problem at stock clocks!!

Edit: Well I'm running some custom Prime95 blend right now and I just noticed my RAM is slightly warm to the touch. This is at 1.6v and default 1333mhz. I gotta go to bed but I'll have to pay close attention to those temps tomorrow. For all I know that could have something to do with it too.

New tim job on CPU + fresh OS without SP1 = everything fine at 4.7ghz. Installed the Corsair ram cooler and the memory is cool to the touch now. Next step is to install SP1 to see if things get fubarred again. Temps on the CPU are the same btw. Contact was great when I pulled the cooler off last night.

Nailin it down bro ............... :thumbsup:

It just shows you how long since I've overclocked in a case, hehe. I'm used to a bench table and fans everywhere.

I was just doing a search and ran back across this thread. The actual fix was a reg hack. Here is the thread on crucial forums. Drives work fine now. :thumbsup:

http://forum.crucial.com/t5/Solid-St...-me/td-p/38766

Nice find Chuch :)

Thanks for updating us :)

Did you have high performance mode set under power settings?

I have seen mentioned that one should enable hotplugging in BIOS to get applications to work better, but I thought that was only linux based things like Gparted or other bootable apps. And I have not needed to do so.


EDIT: Holy Crap! MS has known aboutthis problem for a long time

http://msdn.microsoft.com/en-us/wind.../gg463521.aspx

Great stuff guys,

This has been driving me crazy

Rich I enabled high performance mode both os installs. This is a problem with RST drivers and crucial drives. LPM hax fixed my issue like everyone else on crucial forums. You guys need a pic guide on how to to this? I know I had to create the reg entries before running the .reg file.

No pics needed here, but feel free. Glad you got it sorted out!

Intel RST10 drivers are apparently causing issues with lots of newer drives. OCZ and corsair both have recommendations on using MS drivers with certain drives.