nvlddmkm stopped responding

[ripbox]

if it was the case of nv*.sys file being scattered around the place then the problem would be gone after a format and fresh install of vista .. am i correct?
because i have made several fresh installs over the past several month and nether time has it cured the problem
however i did notice a behaviorally change in the way nvlddmkm is dealt with when upgrading from the older drivers to the most up to date drivers

ie 1st driver i had installed from the disc that came with my Asus 8800GTX
nvlddmkm would almost crash on boot or @ start of any graphically demanding game with unlimited amounts of driver restarts - without any BSOD

now on the most up to date driver i have NVIDIA\WinVista64\169.25
more stable but after several restarts you can almost place a bet of a BSOD before you can quit and get back to desktop

 

[stevet]

I battled this BSOD off and on until teh 163.xx series drivers and all of the recommended hotfixes. Then I discovered I could trigger it when I over-tweak my RAM timings. More recently it reappeared, this time found a bad RAM stick.

Point being, if you can't make it go away even with a fresh reinstall then suspect your RAM. Test each stick with memtest86 (at least 5 passes - will take a few hours). If all of your sticks are good then use one of them to test each of the RAM slots on your motherboard. Bad slot would indicate a bad memory controller.

 

[eynestyne]

Had this problem, I contact HP support and they told me to replace the file with one from the nvidia folder. That worked good. The files extension was changed to .old... the new one frm the nvidia folder was expanded and placed in the folder.. Hope this helps

 

[vchad904]

Also had this one for awhile, BFG ended up replacing it on the second call, heck they even upgraded the card to the next revision (7800gs), that was months ago.
Then the other day I saw that BSOD again, and updated to the newest driver. So far no more BSODs. Never did find out what was wrong with the original card....

 

[reality]

Has this been sorted at all? I can't even play any games at all without it messing up.

I don't know much about computers.

 

[vchad904]

Has this been sorted at all? I can't even play any games at all without it messing up.

I don't know much about computers.

I'm hoping my second time around with this issue is fixed with the new driver. As I stated in my first post, the first time I got this BSOD, BFG had to replace my card, but mine wasn't crashing in game. It would crash on profile load or just after.

 

[HillBilly]

I thought I had this problem licked, but as many have noticed one way or another this problem comes back. The problem is the error seems like a general error and it doesn't pin point any actual problem. Any ways, I switched out my video card, still broke, then every ram stick out one at a time and no failures. I than placed two sticks in and ran counter-strike after 2 min. computer would lock up. I decided it must be the memory controller, plus I was spending so much time on this one problem, it would be cheaper and quicker to just replace the motherboard, and see if that works.

I purchased a new evga 780i, I would like to make a point why I switched away from Asus and went to evga, mostly because all of the positive reviews about their support, hopefully Asus and other companies may learn from these companies - 24x7 tech. support phone services (LSI Logic also offers 24x7 service). I have already used evga support 3 times and I am very happy. I have had no board problems (I had a highpoint / LSI Logic controller board conflict problem, but its FIXED. Thank you Highpoint). Plus I overclocked my e6750 to 3.5 ghz without any problem, my temps are only 30 C idle 35 C load so plenty of room to grow.

I will RMA my Asus P5n32E-SLI, back to Asus, maybe I'll build me another computer and put Linux on it and try to learn Linux.

Thanks for all your help and hopefully they come up with better error codes in the future that will help narrow problems like this down.

 

[ripbox]

hmmm would be ok if i had an asus board but no i have an abit... maybe i could loos the asus 8800gtx and get 2 xfx 8800 gts's oh and some new faster ram

 

[stevet]

quote:
16. OK so in our Admin CMD window we type the following commands in sequence
17. takeown /f C:\Windows\System32\DriverStore\FileRepository\*.*
18. The command above will give you ownership of all the files under the FileRep folder
19. cacls C:\Windows\System32\DriverStore\FileRepository\*.* /G Username:F
20. Replace username with your windows account name or login name
unquote

Just a caution on this. Right after you delete the nvlddmkm.sys files be sure to re-enter the cacls c:\windows et.al. command this time typing SYSTEM in the the username. This resets the directory/file security privileges. Failure to do so can reset in an inability to install drivers for usb devices, monitor, etc.

And if the BSOD continues test your RAM with memtest. I had my system stable for several weeks and then the BSOD reappeared. Found I had a bad RAM stick. Pulled it and no BSOD.

 

[iamc3k]

Well, no luck for me.

I ran Jrobert's procedure. I have run memtest86+ v2.00 overnight repeatedly. I have run Prime95 v25.5 overnight repeatedly. I never get ANY errors. Yet, I just tried to play CMSF (a BF.C game) and it crashed with nvlddmkm within 18 seconds. Crysis, COH, WiC are all seemingly stable.

At this point, I think I will become an ATI customer.

I will also start to aggressively pursue Apple and Linux as my OS's of choice.

Nvidia and Microsoft each are silent on this issue. The widespread nature of its occurrence seems to be hardware independent. The software left is MS's and Nvidia's. I don't care who is at fault. The fact that neither can fix it or even admit it's a problem is why they will lose my business.

Regards,
Ken

 

[Justintoxicated]

Well, no luck for me.

I ran Jrobert's procedure. I have run memtest86+ v2.00 overnight repeatedly. I have run Prime95 v25.5 overnight repeatedly. I never get ANY errors. Yet, I just tried to play CMSF (a BF.C game) and it crashed with nvlddmkm within 18 seconds. Crysis, COH, WiC are all seemingly stable.

At this point, I think I will become an ATI customer.

I will also start to aggressively pursue Apple and Linux as my OS's of choice.

Nvidia and Microsoft each are silent on this issue. The widespread nature of its occurrence seems to be hardware independent. The software left is MS's and Nvidia's. I don't care who is at fault. The fact that neither can fix it or even admit it's a problem is why they will lose my business.

Regards,
Ken

Any chance you are running a creative card and NForce MB?
Any chance you do not have the latest bios for MB?
Pass Mtest and Prime 95 (on both / all 4 cores?)? passing on 1 core does not mean much obviously..
GPU Temps? CPU temps? Voltages fluxuating on either (not sure how to tell this due to software/driver signing)
Have you tried verifier.exe? (the existing solution MS should have used and made more public rather than blocking all unsigned drivers)?

Go to run, type verifier.exe -> Create Standard Settings (Next) -> Automaticaly select all drivers on this computer -> Finish.

Reboot windows and see if you get a bluescreen, write down error msg, boot safe mode, re-run verifier.exe, disable settings -> reboot normal & google error info, find possible solutions...

does sound like a driver/software issue though, cause I would imagien crysis would pretty much be stressing your GPU....

 

[iamc3k]

Justintoxicated,

Thanks for the reply.

As to your questions, I have an Abit IP35 Pro motherboard (intel chipset, not nvidia), I am using the MB integrated audio, Realtek 7.1.

The Bios version is 11. The latest is 14. There is no 12 or 13. 14 seems to have some instabilities and only addresses E8400 series half-step multipliers. I have a q6600 processor. A 16 is in beta.
Mtest/Prime95 on all 4 cores.
GPU/CPU temps are all quite low. Mid 50's to mid 60's. (My cooling: an Antec p182 case, 2 Noctua 12cm intake fans at 800rpm, 2 Nexus 12cm fans exhausting at 800 rpm, GPU has Thermaltake HR-03 plus w/9cm nexus fan, cpu has Thermaltake Ultima 90 w/9cm nexus fan. MB heatpipe has been straightened, lapped, and AS ceramiqued. Capacitor around LGA775/cpu seat have been modified with AS ceramique and individual heat fins. Etc. In short, I have taken a lot of measures to ensure all components are extremely cool. All stress tests and temperature readings verify that I have achieved very cool operating temperatures.)
Voltages: I have mildly increased the MCH voltage from 1.25 to 1.33. CPU volts have likewise been slightly increased. RAM voltage calls for 1.8 to 1.9; I am using 1.85.

There is evidence of a .05 vdroop to the cpu. That has been countered by an increase in BIOS cpu volts.

I have been running with an overclock. The cpu stock is 2.4. I've been running at 3.2. I am going back to default clocks and trying to see if the 'nvlddmkm' is still an issue at default values.

What does verifier.exe do?

I will run it and let you know what happens.

Thanks,
Ken

 

[iamc3k]

Justintoxicated,

I just ran verifier.exe; twice. Both time I received a BSOD on reboot; both indicated an issue with "klif.sys". The BSOD cycled off before I could read all the information.

A search of "klif.sys" shows it to be a "Kaspersky Labs Internet Firewall" file. I run Kaspersky Anti-Virus, so no surprise there. I do not consider "klif.sys" to be a problem or a factor in the nvlddmkm errors.

Regards,
Ken

 

[Justintoxicated]

Justintoxicated,

I just ran verifier.exe; twice. Both time I received a BSOD on reboot; both indicated an issue with "klif.sys". The BSOD cycled off before I could read all the information.

A search of "klif.sys" shows it to be a "Kaspersky Labs Internet Firewall" file. I run Kaspersky Anti-Virus, so no surprise there. I do not consider "klif.sys" to be a problem or a factor in the nvlddmkm errors.

Regards,
Ken

IMO I would work on these issues first. Whatever your firewall is doing it is making your system unstable enough to cause BSOD. Try removing and re-installing or checking for updates and see if you can get the "first" BSOD to go away. The Nvidia one has probably not shown up yet due to the frist BSOD, one step at a time to get your vista stable! Once you get rid of all of these your system will be much more stable.

Follow these steps to stop your computer from restarting on Blue Screen so you can google more info. They are the same steps for vista as for xp.
Disabling Blue Screen Of Death Auto-Reboot - Windows XP Tips and Tweaks on Tune XP


Read this for more infomation on troubleshooting BSODs.
Troubleshooting Windows STOP Messages

It took me a few days to get everything working on my XP install (Driver wise) and another few days to realize my lockups not involving BSOD were from problems with creative and Nforce Mother Board. I decided to flash to the newest bios for the Motherboard and so far so good but time will tell

 

[iamc3k]

Er, this is a vista forum and I'm using Vista Ultimate x64. Do the XP tweaks work on Vista?

And how does a firewall from a known program, which I have purposely installed, cause a BSOD?

I'd like to know more about verifier.exe and what exactly it is verifying and doing at bootup which causes a BSOD. Do you know?

Thanks,
Ken

 

[Justintoxicated]

Er, this is a vista forum and I'm using Vista Ultimate x64. Do the XP tweaks work on Vista?

And how does a firewall from a known program, which I have purposely installed, cause a BSOD?

I'd like to know more about verifier.exe and what exactly it is verifying and doing at bootup which causes a BSOD. Do you know?

Thanks,
Ken

This is not an Xp tweek, its a microsoft utility.

Changing BSOD's to stay vs auto reboot is the same for XP as vista I just did not feel like writing down all the steps.

I don't know 100% how the verifier works but I have had SEVERAL poorly written programs cause BSOD. after running it, changing bios/drivers/software updates and I have been stable ever since.

Just because your program is well known does not mean it does not have issues. I have bluescreens from Nvidia, Creative, and Logitech (mouse) all cause BSOD's running verifier.

If you want to go to ATI that is fine. In my experience ATI hardware is more reliable than nvidia but the drivers are often lacking compared to Nvidias. You could end up with more/different issues. The grass is not always greener on the other side. I don't play fanboy for either side just IMHO...Having hardware failure rate alone I would buy an ATI card hands down, but there is most certainly a trade off in drivers and mostly in speed right now.

Start by solving the problems caused by verifier (Driver Problems), you can always re-install your firewall later if you must have it. or it may be better to use something that is actualy compatable with vista (Passes verifier.exe).

BSOD is either caused by hardware failure or software. Software problems are typicaly cheaper to solve. Prime + memtest would probably expose hardware issues. Start by getting rid of what is causing your errors. If you choose not to I can't blame vista on this one, and I'm not the one to praise vista myself.
I'm going to change my laptop back to XP when time allows and my new hard drive arrives. But these are for reasons that do not have to do with this thread. I can't say that the verifier will fix your problem, but if the nvlddmkm BSOD is thrown by the verifier then it is almost definatley a driver issue and you should be able to produce it on command and you can play with different driver versions / installs until you get one that works.

Once you can see all the info for your BSOD perhaps you can find out more on what is causing it (IRQ conflicts etc if from a hardware side)

One BSOD can lead to another so you need to start with what you know. you don't even need to uninstall your firewall.... Just run msconfig -> startup and disable the process/es from starting up with windows...

 

[ripbox]

well i managed to fix this problem in a very expensive way

i went out and baught a Thermaltake W0133RB 1200W PSU and a BFG Geforce 9800 GX2 and since he not 1 BSOD relating to nvidia drivers
the only BSOD i have had since is where explorer has stopped responding on boot
ok it not an option for all but my point is that i think the 9xxx series does not contain this driver bug

 

[deltatux]

It could just be the overclocked CPU/GPU causing the nvlddmkm to stop working. However, I have had these issues before with my old GeForce 6600 AGP card on Windows Vista but that was last year, so I'm very sure that was fixed. However, it may still be a problem with their x64. It's either that or your GPU is causing trouble with the driver.

As one poster said earlier, ATi cards usually last better than NVIDIA ones and I do agree with it being worse than NVIDIA when it comes to driver features. However, that is for your decision.

xboxrulz

 

[iamc3k]

I have not had an nvlddmkm error in a while. In fact, I stopped having them when I stopped my overclock.

I've go a q6600 G0 which I clocked to 3.2. It ran Prime95 overnight (7+ hours) without an error. Several times. I also checked my ram (8Gb) using Memtest. Again, 8 hours or more, repeatedly, with no errors. I also benchmarked with 3DMark06: no errors.

I never had an error in any stress test I applied to my machine. All my temps were low.

In desperation, to stop the nvlddmkm issue, I went back to the default CPU clock. Nothing else was overclocked on my machine; Ram timings, PCI bus, nothing else.

I am left wondering if the nvlddmkim error kicked up due to an instability. Possibly the northbridge?

I will OC again and see if it (the error) kicks up again. I will be adjusting my MCH voltages and adding more cooling to the northbridge.

Abit IP35 Pro, Antec p182 case, 4 x 120mm fans, 2 x 92mm fans, HR-03Plus heatsink, Ultima 90 heatsink. What's one more fan....

Thanks,
Ken

 

[deltatux]

The northbridge may well be the culprit in this issue. Make sure to overclock the northbridge when overclocking the system (i think you can do that, not sure, i don't OC my parts except GPU).

deltatux