NMI: Parity Check/ Memory Parity Error -- System crash

[JohnnyCAPSLOCK]

Ok, I'm back with another possible solution. My boss noticed that after he installed Vista SP1 Release Candidate his parity error had gone away. I am running XP on my laptop so I thought maybe the same correction that seems to be in Vista SP1 RC will be in Windows XP SP3 Release Candidate. You can find it pretty easily by searching on Google. I've installed and this sucker ran all weekend through to today Tuesday morning. No blue screens. Typing on it right now. This is with a Dell D620 and Atheros card. I've not stress tested it yet, but this thing would just blue screen at the login prompt if I let it sit. Can someone else try this if they have Atheros problem and let me know if it is just coincidence that mine have seemingly gone away?

JC

[slider.cs]

Hi all, i brought my xps m1710 in august 06 and as far as i can remember ive pretty much had the problem from the start, it only happens to me playing games. In particular during BF2. It used to only happen in xp and vista used to be fine but i have since gone to just using vista and it continues to happen even though its vista. It doesnt happen everytime i play, but once it starts when playing it continues. I have gone some noghts of playing without it happening at all. Every time i play i use my laptop cooler to help keep the heat down, and this works pretty well. It certainly stops the laptop fans coming on full speed even when playing. I dont remember it ever happening outside of a game. I havent read every single thread in the forum but am i right in thinking its only dell computers. If so has everyone got dell quickset installed, just a thought as it controls a few things. Also im still trying to work out if i have a headset connected each time it happens aswell. This uses usb and the audio jacks. Trying to cover every angle here. The fact that it only ever used to happen in xp and not vista interests me. Although its now happening in vista and i dont have xp installed anymore. Does it boil down to drivers? Lets hope someone can solve this once and for all

[arana]

this happened to my boss twice, it is not a memory issue, we swapped memory with another dell (same model D630) and the error still occurred, also had issues with GARBLED video with or without the port replicator be it on external monitor or laptop display (dock station etc)

[eleron911]

LOL.
I had the issue twice so far, ONLY in XP SP2, not in vista(which I had before) , and ONLY after trying to resume from S3(standby).
I`m hoping it will stay that way.
By the way, did no hardware upgrades or changes, and I never used my wireless card ,never had a chance to. System is from august 2007.

[slider.cs]

Well, after going through several stages in eliminating different things im hoping ive solved the problem. Ive tried removing the wireless card and the modem with no joy. I then decided to reseat my graphics card. That was 3 days ago and i havent had a blue screen since so far. I have an xps m1710 and to reseat the graphics card is quite easy. During this time i have played Battlefield on a few occassions and been fine. This was when it would really happen. The night i decided to reseat the card it happened 3 times, so hopefully problem solved and i will keep you updated.

[slider.cs]

Well about 5 days later got the blues screen so, that didnt solve it

[machoo888]

God damn it, second time I've spent over an hour writing a post only to be logged out from inactivity, ****ing lame.

Well alright, I've been following the forum for moooonths, and have tried to post before, but I'm totally sick of repeating my super post, So I'm going to keep this sweet and simple due to my well deserved irritability.

I've got an E1705, also purchased in August 06. Also getting the NMI parity error. This computer was tank, I was very proud of it, I used this computer non-stop, we're talking 4-16 hours a day for well over a year.

I saw my first NMI parity error bsod almost exactally a year after I had owned the computer, I figured it was software, so I just reinstalled and was fine, but it still poped up here and there irregularly. It became a dibilitating problem, at the time I did graphic design professionally, so losing work became a frequent problem. I'm also a fellow WoW player, nothing worse then getting having something happen in the middle of a instance or raid. So I just put it on the back burner for a while that way I wouldnt waste any more time playing catch up because of Dell's subpar manufacturing.

Every month I took a few days to play with it. Software, drivers, take it apart, multimeters, thermometers, put it back together, w/e. I always knew it got hot, but when I had it torn apart to actually test the heat buildup it wasnt a problem (of course). After months of trying, I gave up, came very close to scrapping it out, or making a ghetto-desktop out of it, but I've already got waay nicer desktops, I dont need another, I need my laptop to work.

Its been a good 6 months since my E1705 first went out of commission, I havent touched it in a month or two. Here is what I have found, and this is something that might help if we could collectively test to do a mass deduction.

First off: Drivers, short term fix, I havent had my E1705 run longer then 4-5 days after a driver swap/update, and I tired to keep it janked up and swap out drivers every 2-3 days, meh, dosent work.

Second: This IS a manufacturing fault.

Third: This IS a termal problem.

Will Dell ever recall our computers? No chance in hell. They WILL make half assed attepts at fixing a problem their senior techs and design staff KNOW is unsolveable? Guys, you have to understand, its cheaper for them to replace a couple hundred mobos to pacify you rather then spend MILLIONS replacing every laptop that are using these manufacturing designs. Is there any REAL solution for our problems? Hell no, just like Veterns, Agent Orange and the Gov't, they're just going to wait for our complants to die out. And in the mean time continue to brainwash consumers into beleiving that Dell products are "high quality."

The current state of my E1705 is compareable to that of a human in late stages of leperocy or AIDs. But I dont even need to get into my machine, because its no diffrent then any of yours, just waay further ahead as I've used this machine as a test bed to hopefully solve everyone elses problems.

Two things, if you get video artifacting right from bootup, please raise your hand. We're talking matrix-code-esque artifacting, fractured characters, gibberish, w/e. You know, even people that dont start to receive graphic artifacting until windows, raise your hands too, I have a test you all need to try to assert that this is a thermal problem.

Ice your lappy, boys and girls! You heard me right! Stick that ***** on that empty space in your fridge where you just removed those 2 week old chinese leftovers yesterday. The whole machine. Stick it in the freezer in a plastic bag. Just get your machine on ice, right now. Let it sit for a good 1-2 hours, about as long as it takes to properly freeze a tray of ice cubes. The colder the better. When you take it out, be ready to boot it as soon as its removed, dont worry about condensation, as long as you didnt put your machine in there while it was hot, you'll be fine. Turn it on.

Do you still have those artifacts at boot? In windows?

Yea. Thats what I thought.

Same thing here.

Dont get all clever with it and try to ice your laptop before every use, this will only make your probems worse.

Now, back to the loads of previous deductions. YES, this is a PCIe fault. This going to occur in two diffrent ways, both listed dozens of times by dozens of diffrent people. You're either going to **** up your wireless card/slot, video card/slot, or both depending on the ammount of heat damage thats been done. Even worse, a new motherboard/wireless card/video card isnt a gaurenteed fix because this problem has just as much to do with the laptop shell as it does the hardware installed inside it.

So after all this, what? Whats the solution? Did we all just get ****ed by Dell? Does this spell out iminent doom for our laptops no matter what we do?

I've got the solution right here guys.

Moderation.

If you're still under warrenty, then you're one lucky *******. If you're not, then you're one unlucky *******. Either way, you're going to have to deal with Dell, beg them for replacement parts, pay, w/e.

Once you DO get set up with new hardware, assuming they dont replace your hardware with something else that got RMA'ed ("refurbished") for the same problem you already have, you gotta watch your useage like a HAWK.

Keep an eye on your temps. ANYTHING over 140 degrees fahrenheit is bad. If that is sustained, you WILL cause PHYISICAL damage to your hardware. Tempatures inside these laptops WILL reach and sit at temps as high as 190F if you dont keep an eye on it. When your laptops reach core temps that high, dont be concerned about your PCIe bus/devices, at this point you're risking your RAM and CPU too.

Is it fair that we should have to babysit our laptop's temps, use 3rd party programs to FORCE our fans to spool at max speed so while we are using it, so it sounds similar to a 747 during flight prep? No, hell no, but this is a problem we all bought in to, and a problem that is likely to never get fixed.

We're all just a bunch of scrubs that bought into DELL hype. We should all get gold stars. Or at least a golf-clap from DELL.

That being said, anything else anyone has to add would of course be greatly apperiated. I may not be correct. I may be totally wrong. But at this point, we're beating around the common-sence bush. I bust my 8-10 hours a day as an underpaid computer technican at the biggest small computer place in a town of ~500k people. I may not be a Dell "certified" technican (/laugh) but I know way more then your average joe, computer guy, ugh, geek squad employee, ****, I got $100 on the fact that I would troubleshoot circles around a dell technician. This being said, dont take this post for fact, but as an incredibly well deducted answer for a seemingly unfixable problem.

Replys/Kudos/Constructive Critism/Flames are welcome as always, but I dont know if anyone will be able to offer a better answer then this.

Sorry to everyone else with a $1.5k-2.5k paperweight, I feel your pain.

~Machoo

[Zaron3d]

I'm long-winded too, so do what I do.... write your replies in notepad and then click reply and cut-and-paste. That way you have time to proof-read!

So anyway, here goes....I'm sure somewhere down the line someone will search for "Memory parity e1705" in Google as I did, and I'm happy to be able to give some answers.

First of all, lemme say that just recently a friend let me have his e1705 laptop to use to help him do some web development work. He did warn that "It crashes to blue screen randomly" saying it was "some kind of overheating issue". Sure enough, after the laptop sat running for anywhere from 30 minutes to 2 hours, it would go right to the dreaded bluescreen of death that I know you all have already come to loathe.

Well that would never do. This was a really nice laptop - 2ghz dual core processor, 2 gigs of RAM - and I wasn't just going to accept that this was something it "just did". So what to do?

First, I considered it might be a drivers issue, which doesn't make a LOT of sense, since it was a memory issue, but what the hell. I went to Dell and updated all their software. Rebooted, held my breathe and waited... CRASH

Then I scanned for viruses, malware, spyware, and general disk errors. Reboot, wait... CRASH

I understood that laptops don't generally just start overheating so this was most likely a cooling issue. Unfortunately I will say it's a freaking 2 hour ordeal to get to the fans, but it is possible. Please note that doing this yourself WILL VOID YOUR DELL WARRANTY, but if you're brave enough and you have a small screwdriver, you can find everything you need here : http://repair4laptop.org/disassembly_dell.html. Note that you will have to remove the LCD screen to open the case (yes really) but it's not as bad as it sounds.

So I open the case and remove both cooling fans (on cools the processor and one cools the video card) and I was horrified to find a mutant caterpillar living in each one. Okay, not really, but the size of the dust bunny in each was amazing. This thing sucks up dust better than my vacuum cleaner! It was apparent that this was the problem. With this giant dust ball blocking the sensitive equipment the fan was supposed to be cooling, this was obviously the problem! The parts were overheating from not being cooled and BAM! I cleaned and dusted the internal parts thoroughly, flossed dust from the blades of the fans, reassembled the laptop, patted myself on the back, powered it up and all was well... for about 45 minutes. CRASH!

After much weeping and wailing and gnashing of teeth, I decided to test the memory directly. I had already run burn-in tests and the memory passed with flying colors. This is because memory errors are somewhat random on the chips. Burn-in tests can assault the chips, but if it doesn't hit the problem area in just the right way, the memory will work. I mean, you can boot into Windows and use it for a while, right? No reason why the memory wouldn't pass a test. But just to be sure, I took out one of the chips and rebooted. The laptop actually seemed more responsive on the one chip. Maybe this is it, I thought. I used it for about 15 minutes and CRASH!

This is going nowhere, I thought. Such a shame too because it's a really nice laptop. So I swapped to the other chip, rebooted... and it's been running for 2 days straight. Hasn't even hiccuped. I can now say with at least 95% assuredness that the problem is solved.

So let me back up here. The MAIN problem is the fans aren't / didn't cool. This problem MUST BE ADDRESSED, otherwise if you swap bad ram out it's just going to get fried again. But chances are this is just bad, cheap RAM (who the heck is Hynix anyway?) and was destined to fail anyway. Nevertheless, there are plenty of good components, and although 1gig of replacement RAM is only $26, you should price the processor and video card! If your laptop is over a year old like this one, expect that yours has the same buildup.

I hope this serves you well. Your laptop isn't broke, Dell tech support is a joke for not knowing how to fight this problem, and if this fix saves your day, feel free to drop me a tech support donation at my Paypal account - rush3rd@gmail.com... but an email of thanks would be good too!

Your techie pal in the know,
Rush Montgomery III

[switchmaster]

I have a Dell Latitude D620. I got the memory parity error for a couple of weeks off and on until the laptop finally refused to boot up (it got into a rebooting/error loop). Today I put the hard drive from the fried laptop into a different (refurbished) D620 and it works fine. So this is definitely a hardware issue. By the way, you can get the temperature monitoring program at http://www.diefer.de/i8kfan/index.html. My CPU temperature is now at 44 degrees and everything is working fine.

[martin79]

I had the NMI error for over 6 months, with everything in the laptop replaced. Dell finally replaced the laptop. The only diffrence i can see in the replacement system is that they are using Kingston RAM, not the no name crap i had before. I've now been running for 2 months without seeing a problem.

Also just a note to everyone that thinks heat is causing the problem, no it's not, heat is a symtom of a problem, not the problem it's self. Ramping the fans up is only masking the problem, not fixing it.

If your under warrently, keep on at Dell, they will replace it. If not and your in the UK, check out the consumer rights act, it's likely you can get them to take action, so long as the laptop is under 3 years old.