Sunrex

Started: http://www.ocforums.com/showthread.php?t=584402

Due to lack of replys on that forum, I decided to ask for help here. To look up what the issue is just visit the link above.

Update on problem:

I have changed the processor to a RMA that I recieved back from AMD, it has a better cooling system from what I'm seeing.

Turned on the PC, loaded up windows vista setup. Blue screen of death. Removed a ram stick, did it again. Blue screen of death.

Upon going into BIOS, and setting everything to optimized settings the blue screen went away. Upon further research, this new CPU wont support FSB 217 like the other one could. This could mean one of two things, the CPU came near the end of the silicon plate (Or whatever it is) thus, quality was reduced... or, the other CPU was BARELY running at that FSB speed, meaning data corruption could have happened at random...

I have yet to actually leave it off all night and boot in the morning, where most of the issues seem to occure. So I dont know if it fixed the issue or not. I also noticed what looked like some artic silver 5 on the actual CPU socket (On the motherboard) that could be causing the issue, but it seems weird it only happens every few hours to days...

If the data corruption is fixed in the morning, im happy. And im going to call AMD and RMA the CPU that caused the corruption.

Any other suggestions in case this doesn't solve the problem?.

Nenkitsune

after reading your threads, it sounds like your system just isn't stable at those speeds. what happens if you run it at stock speeds?

__________________
BE-2400 3.105ghz 8 hours orthos stable
BE-2400 3.484ghz on AIR PROOF!
7750 BE 3.654ghz suicide Validation

IEATFISH

Welcome to the forum. If you could put in your system specs here, we can see the rest of your system.

Do you have thermal paste on the processor in between the heatsink? Also, make sure your RAM voltages are what the manufacturer recommends.

__________________

Sunrex

Honestly?. I don't know. I just set the settings for optimized, everything is stock. Memory is at the timings recommended by newegg and people in the reviews. I still need to see if the data corruption occures, seems to happen when the PC is left off for a long time. (Overnight).

Ok, I added what I could off the top of my head. No internet at home so I'm at the library.

I did have thermal paste on the last processor, but I learned the hard way the default thermal tape/paste/whatever it is that comes with the stock heatsink is best for cooling. AS5 gave me up to 67C on full load, but I recall the stock tape giving me around 61C.

IEATFISH

You should definitely try to keep your processor under 55C, that gives you some buffer before 60C.

How is your airflow in and out of the case? Also, if you are overclocking (I assume you are) are you going up in tiny amount and then backing off when you get an error? You should only be bumping it up 5at most at a time.

__________________

thlnk3r

Sunrex, data corruption issues would be more like missing system 32 files or NTDLR errors after POST. If you're getting random BSOD's while in Windows then that sounds like an unstable OC or perhaps maybe even a failing component. I'd recommend manually setting everything to factory settings including your processor speed, memory and chipset speeds. Start off testing with either Orthos or Prime 95 for a few hours. Have you ran memtest on your current set of memory to verify that isn't the culprit? I typically run tests 5 and 8 each for at least 20 minutes.

When running at 217HTT (reference clock) did you also place your memory on a divider?

Let us know

Good luck

__________________
AM2+/AM3 Overclocking Guide
Socket AM2 Overclocking Guide
Socket 754/939 Overclocking Guide

Terms of Service

Originally Posted by txtmstrjoe:
"Steppings merely indicate possible outcomes due to the law of averages being in your favor."

Sunrex

I have no idea what memory divider your talking about is.

The RMA CPU was no go. Still getting data corruption.

I am getting system32 errors and NTLRD errors after post. Heres a experience I had after keeping the system off for a few hours and attempting to reboot:

File: \windows\system32\drivers\volsnap.sys

Status: 0XC0000221

Info: Windows failedc to load because a critical system driver is missing or corrupt.

Attempted to use last known good configuration, blue screen of death. Reboot. Tried again, works.

Used reliability and performance monitor. All good, two warnings. One was about no anti-virus, and the other was: THe average disk length is 4. THe disk may be at its maximum transfer capacity due to throughput and disk seeks.

Now, I have my hard drives mounted sideways rather then flat.This is because my PC is a micro case. I have the case open for circulation, otherwise my GC reaches 100+C.

It might be that I have the jumpers removed from the hard drive. So I have a full 3Gbps rather then 1.5Gbps. I don't know if this will cause errors in IDE mode.

I've had system32 errors and NTLRD errors since I built this PC, IT IS NOT THE HARD DRIVES (UNless its the jumpers..) I bought a new hard drive thinking the old one was going, no good.

I can only fault the motherboard at this point, I cant think of any other reason I get data corruption when its shut off for long periods of time.

Another issue I've seen is when I tried mounting the partions in a linux live CD. It said it couldnt mount due to blah, blah. Upon asking on the linux forum/IRC, they said it was because I shut down the PC improperly. Sadly, I shut it down right. So I am not sure whats wrong here.

This is seriously pissing me off, I keep losing over 100GB of data every time I have to reinstall. It irritates me more because one second it says it cant read a system32file on boot, and then another reboot and it does it fine.

Prime95 and Memtest showed clean, but this was a while back. It is possible the MB is not supplying the voltage correctly. Perhaps its working at 2.1V and then the ram Voltage drops suddenly for a second?. I don't know.

I know theres a good chance getting a new MB would be great, but I really dont know.. I have yet to really have any problems with this MB other then the data corruption, again that only happens when the PC is off for long periods of time or its on for a few days...

Sorry for the long posts, just trying to fully explain the problem.

thlnk3r

Sunrex, you sound a bit bummed so I'll try and not ask to many questions

I know you may not want to do this but it's part of the process of elimination. Please perform a full surface scan on your drives. This will look for bad sectors. Hopefully this will rule out if any of the drives are bad.

For Western Digital drives, I recommend downloading the Data Lifeguard Diagnostic Tools. Burn this utility to a CD and boot directly to it. You can download this by going here: http://support.wdc.com/product/downl...sid=30&lang=en. Download size should be 2.10MB.
For your Seagate drive, please download the Seagate Seatools. Do the same exact thing (burn as an image to a CD). Those tools can be downloaded here: http://www.seagate.com/www/en-us/sup...loads/seatools.

What you posted above in regards to the NTDLR and missing files ect sounds like a classic unstable OC that eventually corrupted your OS (mbr, critical system files). I have a feeling the first time you tried 217HTT it didn't go so well. More than likely your memory frequency was also overclocked which might of cause of the original problem. Perhaps you did mention this already but did you try reinstalling the operating system? Does the same issue still occur? I would also start testing your machine at stock clocks to see if it crash/bsod's. I prefer Orthoswhen testing.

Let us know

Good luck

EDIT: Just a side note. I have performed surface scans on many hard drives in the past. Sometimes a drive may pass but in fact it may still be bad. It's hard to explain but the only way I can put it is...the performance level of these so called "passed" drives was just horrible. Benchmarks were horrid, load times were abnormal ect. A simple RMA solved my issue.

__________________
AM2+/AM3 Overclocking Guide
Socket AM2 Overclocking Guide
Socket 754/939 Overclocking Guide

Terms of Service

Originally Posted by txtmstrjoe:
"Steppings merely indicate possible outcomes due to the law of averages being in your favor."

Sunrex

Drives are fine, I did the seagate test before (Seagate is new) and it passed. The WD drive works fine on another computer with no corruption. The only difference is that the drive is mounted sideways rather then flat.

The only OC I am currently doing is the RAM, but even then it's at the recommended timings unless its getting corrupted because the FSB makes the ram 766Mhz or something rather then 800Mhz like it should be. I would have to do a OC of the CPU to be able to hit a stable 800Mhz. Your guess is as good as mine.

I'm confident its the board, actually. If it can't supply the voltages correctly then its more then possible data corruption can happen.

Plus, if I can restart the PC and data that was corrupted before the restart no longer is... well, could be a ram problem I guess but memtest ( A hour or two) showed no errors.

The new CPU with the default thermal tape is doing good. I haven't tested it at 100% load, 55% load gives me a max of 42C.

I have noticed the CPU voltage is rather unstable. I have the VCore at 1.350 but it goes up to 1.380/1.430 underload. I would have passed this off as the BIOS adjusting the voltage but I turned off the auto on voltage.

So, no ideas?. Should I just shrug it off and buy a new board?. It's not the PSU in the system, I've replaced that.

my77stang

could it be a hard drive problem? google spinrite (you may have to use torrents) and DL that utility. it will make a bootable cd and will scan / repair any corruption on your drive. its probably the best hard drive utility on the market.

__________________
Proud member of The Gr33n Machine and No Ma'am