Sign in with
Sign up | Sign in
Your question

E6300...I think I need some experience to help me here.

Last response: in Systems
Share
December 13, 2007 4:15:59 AM

BACKGROUND: I built this computer May '07. I usually maintain an overclock of 415Mhz@ 1.32V w/memory at 833Mhz 4-4-4-12 @2.1V



SYSTEM:

CPU - E6300 Allendale/L2
Mobo - EVGA 680i SLI
Memory - 2x1GB Geil PC2-6400 GX22GB6400UDCA....I believe they're D9's
Hard Drive - 2 Single Platter 160GB SATA Western Digitals
Vid - PNY 8800GTS
PSU - PCP&C 750 Silencer


The Problem:

Recently my system has been incredibly unstable manifesting hard lock ups. I've gone through 3 motheboard's and another set of Corsair memory in my attempt to troubleshoot the hardware. Running the CPU overclocked seems to make the problem worse. If I under-volt to 1.2 my processor seems to run stable for a bit longer, but I'll give you the details chronologically.

1st Mobo (original)-- BFG Tech 680i SLI, BSODS happen every so often but not very common; once a month? From the begining I had issues overclocking this guy. I was able to get her stable at about 415 for the FSB at 1.32V. Everything seemed fine except for the random bsod w/memory dump. At first it was due to a memory leak from nvidia drivers, other problems were usually dealt with, but every so often I had BSOD's that were unexplainable.

Last week after watching a .avi on the internet my computer had a hard lock. The mobo initiated restart right away and gave me an error code that defined said no memory was present. No memory present? This ensued for about 5 or 6 restarts. I removed one stick and it booted. The hard drive was corrupt. I attempted to fix the given issues to no avail and followed up with a clean install of XP. For a time I was unable to run dual channel or even have two sticks of memory. Eventually I was able to fire up dual channel mode in original DIMMS and everything seemed normal.

Friday 07Dec, COMPUSA is going out of business and I had a warranty plan for the mobo that might be flakey...sweet $300 to spend in store. However, all they had available was a Bad Axe 2, 975 available. I got it...same problems -- 3 beep POST code telling me no memory was present. Now the memory issue is 10 times worse. Out of 209823408 times I rebooted only twice was I allowed to commit to two Dimms and utilize dual channel. Otherwise, I was relegated to using just one memory stick. So I purchased a second set of Corsair modules thinking maybe my 680 burned up the other ones, but I never went past 2.3V. I'm sure it was stupid to buy more D9's, but I think that's what they were. Anyways, same problem. Except now it's a bit worse. Even if I POST without error, I have another hurdle; the mobo feedback in the bottom right hand corner locks at "EB". Essentially that had something to do with either the loading of legacy bios something rather or a memory problem.

Now I managed to get my hands on an EVGA 680i SLI mobo and have relegated the issue as to either a Power Supply problem or flakey CPU.

Now I'm thinking, "maybe I have a bootsector virus?". Nope. Headed to repair console, fixmbr & fixboot + format. Turned off comp without a shutdown command. Pulled power from the wall and drained the board. Removed memory, reseated after a minute. Pulsed the power button a few times to make sure it was drained. Now it finally boots correctly for the first time in about a week or so. For some reason ever since this problem started, the proper signal isn't being sent to my monitor to pull it out of power save mode. It ALWAYS turns on after the POST beep. Now it'll get all the way into XP without turning on. I have to hit the source button for it to check the signal and then it works. it NEVER DID THAT BEFORE!. Sorry, it's just irritating.

Anyways, it seems I don't have a bootsector issue and the problem still occurs. I have tried underclocking and every hardware configuration I can think of to alleviate the issue. This is what seems to work -- CPU@stock w/1.2V and I pushed the FSB to 1.5V. I still have popup "windows has recovered from..." error every so often. But it's only hard locked once in the past hour.

I apologize but I don't have the tools to measure VDroop or other essentials. I have relegated the issue to either the PSU or the Processor. Possibly a surge killed something in the PSU or hurt the CPU, I'm not sure. If it helps any I do live in a dorm.

Have any ideas as to how I can safely assume it's the PSU or CPU?

More about : e6300 experience

December 13, 2007 11:09:14 AM

EVGA forum can help you.

http://www.evga.com/community/messageboard/defaultmb.as...

Rarely does a cpu cause hardware failures unless it's damaged/faulty. i.e. dropped on the ground, bent a pin, but fixed it, overvolted, etc.

I came to that conclusion because you went through 3 680i boards with similiar problems.

Have you installed ntune? Get rid of it.

Do you do file sharing or play online games? Does it bsod while you're doing that?

Have you memtested for 8 hours? Constant bsod's can mean one thing: memory. Either config or fault.
December 13, 2007 11:45:38 AM

It happens almost immediately after booting. I've tested two sets of memory sticks. One by Geil, the other by Corsair. Both give similar results. It doesn't always present itself through BSOD though. It could happen at any given time and it's a hard lock majority of the time.

I've used Orthos, forcing a hard lock but with no errors prior to the hard lock or BSOD. I've also done memory testing on both sets with no errors reported.

I'd also like to point out that both 680i's were A1 and that the third board I had used in testing was an Intel D975XBX2 otherwise known as their Bad Axe 2. I've had no issues with severe temperatures either. I can't say I've brought out a multimeter to check out my V's but I did notice that the bios is reporting .04Vdroop and my 5V is 11.93 at it's worst. The 3V's are 3.28 stable.

This is the first time in a long while I haven't been able to figure out what the hell is wrong. I'm about to go buy a $38 celery to see if it's a CPU problem. Could I have had an over voltage that effed up my cpu to some extent? As soon as I dropped the CPU into this newer EVGA 680i the board reported that the CPU was a 16x multiplier at 4.something GHz. Of course it never attempted to post at those settings...but it was still odd none the less.
Related resources
December 13, 2007 12:31:06 PM

My guess is bad ram or a bad PSU. Try running your Bios at default values for a day or so. If it dont lock up at default then you probably have a bad overclock. If it does its Ram or PSU. Maybe your ram doesnt like volts over 1.8. You need to run defaults to narrow it down.
December 13, 2007 12:32:25 PM

What was your overclock? Maybe you ran it to high and to hot and the CPU is indeed bad.
December 13, 2007 12:46:28 PM

Do you know anyone who will let you borrow their CPU? Did you upgrade the BIOS? I don't think it is the CPU, I have the same chip and the Asus P5N32-E SLI and I've had the same problems in the past (lockups, BSOD, memory dumps) and I tried all the BIOS's for this board and finally found one, the 1102 BIOS, which fixed all my issues. Now I have the system running perfectly stable at 1600 where before this BIOS it wasn't stable a stock speeds.
December 13, 2007 9:37:09 PM

BTW, speedfan can show vdroop under charts tab, voltage & vcore1 & 2. Have it open while stress-testing. Note the differences in voltages.

Copy & paste your posts in evga forum. They can help you.
December 14, 2007 12:43:29 AM

I apologize if I come off as abrupt but I think you all missed something very important. I'm not able to even idle under XP without a hard lock up. My definition of a hard lock up is when the screen freezes, maintaining the last valid frame that was output from the video card. The computer remains on, but is not responsive; only leaving a shadow of it's last visual output on the screen.

I've stated my overclock at the very begining. 415GHz@1.32V. I am lucky if the system will even let me open a browser window let alone an email client. If I attempt to install any drivers I risk instigating a hard lock and corrupting the XP install. I've had to format 4 times alone before I was able to get through a clean install of XP without any hard locks that corrupted the installation. When I state that the system is unstable, it's completely unreliable and I'm lucky if it stays for more than 5 minutes. This post alone has taken 4 re-boots. The BSOD's are not common. But I would conclude they manifest for the same reason that is behind the hard lock ups. I called PCP&C and one of their senior techs was scratching his head but was leaning towards a PSU issue the same way that I am.

I came to the forums, not to rehash what I've already done but to find out if you all had any clues that would definitively point out a bad processor or PSU.

Again, I apologize if I sound rude but I was hoping I could avoid redundancy. I've mem tested, cpu tested, looked at thermal temps, and whatever else I could come up with. The problem is definately hardware, but it's not memory or the motherboard as I've already gone through the steps you all have given. At this point I'm no longer overclocking either. I've under-volted and under clocked in hopes of gaining enough stability for RMA's, threads such as these, or other clues to the puzzle.

I truly do appreciate all the help you all have given. I was just hoping for a bit more than what I've already accomplished.
December 14, 2007 12:51:38 AM

Well if your sure it isnt memory its probably the PSU or you ran your CPU to hard and it is dieing.
December 14, 2007 1:00:58 AM

That's what I'm thinking...the CPU is dieing. I had an Allendale and they're not known for their overclocks. I pushed as hard as I could without killing it with voltage. But I know that 680's are notorious for burning up memory with higher voltage than set in bios. I've heard the same problems exist for the Processor, just not as bad. But I've always noticed a Vdroop of about .04

I'm at a loss. I haven't been stumped like this in a few years. After all the overclocking I've done, first time this ever happened. I'm gonna say the PSU took a hit and is effing up my stability.
December 14, 2007 1:17:44 AM

Whats the multiplier on your chip?
December 14, 2007 12:54:32 PM

E6300 = x6, x7 multipliers

On a good note. I have spoken with a Tech at PCP&C. He's come to the conclusion I have, it's a power stability issue. However, we're both wondering how that happens while using a 750 Silencer. We're both opposed to the idea it's the Silencer causing the issue but we both agree that it's a power problem. Either way, they said to RMA so they could test it for me and whether it passes or not, they'll replace it.
December 14, 2007 1:52:38 PM

This is gonna sound stupid, but did you try to swap power cables with eg, the monitor? I know a power cable is really simple, but they have been known to cause problems like yours, and they are REALLY cheap (~$1) to replace, doesn't hurt to try I guess. btw, I have my E6300 stable at 3.2 (458x7) at not very much voltage, so I don't think you've damaged your cpu but oc'ing (unless ur mobo is oc unfriendly). Good Luck.
December 14, 2007 1:56:07 PM

I'm no expert but I would say you can safely eliminate the MOBO and MEM from your list, the fact that you have seen the exact problem with different boards and mem is good enough for me. And I'm thinking your power supply is ok too. So how about trying something easy. Like switching your video card. Clear your drivers and find somebodies card you can drop in. I have seen too many of these video Drives cause too many problems not to at least check that out as it looks like you have tried everything else.....you did say something about slow wake up...something is trying to resolve.
December 15, 2007 12:52:40 AM

when things get hot they expand, and if they get close to something else there can be a short. for example, a misplaced riser connecting your mobo to your mobo tray or case.

if you have swapped out everything imaginable (it sounds like you have) and not found the problem, swap out the case. Do you have any pictures?

GL
December 16, 2007 6:17:17 PM

1) The power cord trick didn't seem to do it.

2) I've already reformatted about 20 times in the past week and a half. I went ahead and found a PCI Video card with an S3 Virge chip and plugged that in too. I still got a BSOD after booting from the CD and getting to the process of copying the necessary files to the HD for installation.

3) That's an idea I had thought of but not quite sure how to go about the process of identifying how, what, and where. The only thing installed that you could call similar to a riser is my 8800GTS. I use onbard sound.

I'll see what I can do for pictures; photobucket perhaps. However, I do plan on getting in touch with Intel and looking into warranty options, as it is a retail chip.

However, strangely, Knoppix and Ubuntu work fine. Therefore, I went ahead and installed Ubuntu. The only problem I've had so far is that installing the driver for my 8800GTS killed the GUI. I can't brag and say I'm proficient at anything dealing with linux so I went ahead and just re-installed with a clean format.

I'll try to get the pics up asap now that I seem to have a "working" system.
December 16, 2007 6:28:15 PM

If Ubuntu install you probably have a corrupt windows disk. Stranger things have happened. I skimmed through the posts again, and I didn't see anywhere if you tried a different PSU. If not stop everything else and find one to try.
December 16, 2007 6:35:45 PM

I've had a hard getting hold of an EPS12V Power Supply. I attempted with another PSU that only had a P4 connection and it seemed to work fine. Anyways, with the Stealth S3-chip PCI card I used, the power requirements should be quite low, considering the processor is a 65W TDP, both hard drives are single platter WD's plus a Plextor optical. Other than that I have two 12cm fans that run at either 800 or 1200rpms and a stock heatsink/fan combo for the cpu.
December 16, 2007 6:36:19 PM

I really can't imagine that setup putting much of a strain on my PCP&C 750 Silencer.
December 16, 2007 6:38:11 PM

I'll have to requalify my statement about the Ubuntu installation. It was not perfect. Three installations quit at the same spot --82%; it said analyzing mirror image or something rather. I left it on all night and went to bed. By some miracle the installation worked this morning and I was able to re-install once more without it haulting at 82% for 5 hours, as it did last night before I went to bed and left it.
December 16, 2007 7:26:59 PM

Well if it is your PSU, you'll probably melt your disk before you decide to try a different one lol.
December 16, 2007 9:16:39 PM

:) 

You know, I already noticed a small hair line crack in the center of my XP installation disc. Of all things...it could be a corrupt installation that XP isn't detecting.
December 16, 2007 9:17:17 PM

I'm thinking about overclocking and see how my Ubuntu installation works.
December 17, 2007 4:57:17 AM

Hindsight; Ubuntu overclocked fine, Knoppix and Ubuntu had no problems whatsoever. They are both CD based OS's. After installing Ubuntu onto a single hard drive, still no issues. I've re-installed XP without issues.

One common theme. I used one hard drive the entire time. Now I don't know if this is coincidence, but I also plugged a fan into a fan header and it proceeded to give feed back to the board and could not hardly turn the fan. I don't know if it's a separate issue or not, but it lit up my head like a light bulb. I DO HAVE A POWER PROBLEM!!!

So here's what I'm thinking...All of these issues have been happening when attempting to run the drives in a Raid configuration. Perhaps this is putting a strain on the PSU. There was only one instance where it failed in a single drive configuration during an XP installation. But so far, the installation is running great. Either I have a HD problem, RAID issue(probably bios), or a PSU problem.

After the fan header problem I vote PSU. Input?
December 17, 2007 5:08:38 AM

I'd also like to point out that while using the Stealth 3D PCI Video card, powered by S3, it resolved the issue of my monitor NOT coming out of suspend once the motherboard posted.

However, once I reintroduced the 8800GTS to the system, the problem resumed. Majority of the time the monitor fails to come out of standby, while once in a while, usually after the computer has been off for several minutes, the monitor is woken from standby as it properly should upon post.

Again, I think this seems to point towards a PSU problem. Ideas?
December 27, 2007 5:34:30 PM

It doesn't seem to be a power issue at all now. Matter of fact, somehow I was able to make my system completely stable. I'm back at a 2.8GHz overclock with memory @800 4-4-4-12. I've done everything I can to recreate the blue screens or hard locks and nothing.

I re-installed using the newest nvidia Raid drivers I could find. Installed motherboard drivers EXCEPT for the Nvidia's network forceware driver something rather that enables "Teaming" n' such.

Anyways...no more SATA corruption or anything else. There is only one lagging issue that didn't disappear. When I boot my computer, sometimes my GFX will send the proper signal to 'wake' my LCD and sometimes it won't. Actually, it will just sit there turned off until I press the source button so that it looks for a signal. Maybe it's an issue with my monitor but it's an NEC 20WMGX2.

Oh well...I'm gonna sit on this and wait until I get my 780i upgrade from EVGA. I don't think the other forum members are too converned so I haven't posted in a bit but I'll certain try to update that now.

Hindsight tells me that it appears that the 680i chipsets have some sort of manufacturing issue. Nvidia keeps brushing it off as driver related but they would have released patches for these problems by now. Their latest drive is from sometime last winter. Apparently from everything I read it's hit or miss with people and problems related to these boards.

So my analysis says that my first 680i board was flakey, the Intel Bad Axe 2 just didn't like the memory I was using, and this newest board might be flakey as well, but as long as I don't instigate that aspect of the chipset that's effed up, I seem to be ok. That means I can't use nvidia's network forceware something rather nTune controls.

Btw, nTune can kiss my ass. I hope ESA is a better product. Matter of fact, anything that's related to tuning hardware through windows = horse shiat.

In the end, PCP&C said that if I send mine in they'll replace it with one tuned for 680i mobo's. So I'll probably push through with just for the sake of it.
a b B Homebuilt system
a b } Memory
December 27, 2007 6:07:11 PM

I800C0LLECT said:


Btw, nTune can kiss my ass. I hope ESA is a better product. Matter of fact, anything that's related to tuning hardware through windows = horse shiat.


Agreed.
December 27, 2007 6:26:39 PM

Before I upload images of my system for those who had asked...

Tomorrow I should receive my Thermalright Ultima 90 w/ Panaflo 12CM :) 

Otherwise my temps are about 33C Idle. I really don't need the newer heatsink at the temp delta is within 20C...I just like having it.
December 27, 2007 6:50:42 PM

I can't say it's anything fancy, but it works. Cable management is a mess inside this case too. Waiting for an excuse to swap out for a new case.
December 27, 2007 7:04:19 PM

For those interested, my CPU fan is intake and the side fan is output.

I have a third 12CM fan in front of the drive cages for intake and my PSU is an EPS12V so it's output fan is directly on the back of my machine.

CPU Idle = 33C
NB = 35C
GPU = 59C

That duct that lian li provides actually dropped my CPU temps by about 3C.
!