Sign in with
Sign up | Sign in
Your question

Losing My Mind

Last response: in CPUs
Share
March 14, 2010 8:32:41 PM

Hello,

Right away, thanks for all your help, much appreciated.

I bought a high end PC from CyberPower in 09 and had nothing but problems with it. CPU fan failure, overheating CPU, graphics card replaced twice (monitors going on and off) and after 3 times shipping it to California and dealing with horrible CyberPower support...I said no more monkey business and RMAed all parts myself (RAM, GPU, CPU FAN, MOBO and PSU). Also, due to constant RAID failures, I purchased 3 brand new Caviar Black HDs and an Adaptec 5805 SATA RAID Controller for a RAID 5 setup in order to avoid any further issues.

After receiving all new/repaired parts:

- GTX 260 from EVGA
- 6GB (3 X 2GB) CORSAIR MEMORY
- P6T DELUXE from ASUS
- V8 CPU Fan from COOLERMASTER
- 850W PSU from COOLERMASTER

I have put the PC back together only to be blown away by problems...and I am lost and unsure what to do. This is my main business PC (I am a pro Audio Producer) and I simply ran out of funds to repair or upgrade.

After a fresh Windows 7 x64 Ultimate install (reinstalled 3 times). I am getting the following:

- My Adaptec RAID controller is running at 86C where normal operating temperature is shown to be 40-55C.
- My Core i7 940 CPU is idling at 47-50C - no overlocking of any kind, 6 case fans and Coolermaster V8 fan with Arctic 5 Thermal Compound.
- I am getting a BSOD every 5 minutes with the error - NVLDDMKM.dll error - no matter what NVIDIA drivers I use, no matter what drivers I rollback and update on the system.

The messed up part is that my hardware is all the same and I never had that BSOD before I RMAed all my parts. Now I have this and all old problems. Please help me as I am desperate to have a working, stable PC, but can't get a break no matter what I do :(  :(  :(  :( 

More about : losing mind

a c 127 à CPUs
March 14, 2010 8:54:02 PM

Well for the RAID failure problems, its strange that would happen on a ASUS mobo especially since you are using a Intel chipset mobo that has some of the best RAID. Hell my old P4P800 setup with a 845P chipset has had a RAID 0 for 7 years straight now and no problems.

The CPU might idle high if the heatsink is not properly attatched. Are you using the stock fan of an after market one? Did you clean the CPU and heatsink off when you replaced it? Did you use the thermal grease that came with the heatsink or did you use a higher quality one such as Artic Silver 5?

Answering those might help us figure out the problems with the heat on the CPU.

As for the BSoD, nVidia is pretty dependable for those. But it shouldn't be that bad. It sounds like the RMAed GPU you got is either overheating or bad. Download this:

http://www.techpowerup.com/gpuz/

It should give you the temps when idle so we can see what is going on there. I can tell you its not because of WIndows 7, its more stable than XP in most every way. It might be the drivers but still it wounds like more of a GPU problem.

Also make sure you have the latest drivers for 64bit:

http://www.nvidia.com/object/win7_winvista_64bit_196.21...

If you did install 32bit drivers it might be the cause of the problem. When you uninstall the drivers, search for a program called Driver Cleaner. Download that and then reboot into safemode after uninstalling the drivers and run that program. When you uninstall drivers, they tend to leave parts there and sometimes even old parts of the old driver which can cause problems. This program will clean those out (including registry entries) and when you install the new drivers it will be completley fresh.

Try some of that and we will go from there.
m
0
l
March 14, 2010 9:04:10 PM

Thanks for the prompt response!

Yeah, I had constant RAID Array failed with latest Intel Storage Matrix and Drivers. 8.9, 8.8 was better but I still had to rebuild and repair once or twice a month.

Before replacing the CPU fan, I used the Thermal Compound Remover and Surface Purifier on both the CPU and heatsink (Coolermaster V8). I specifically bought the Arctic Silver 5 compound and applied it this time. My case is a Cosmos S from Coolermaster and has plenty of airflow and fans! How come I could be idling at 47C according to motherboard BIOS hardware monitor?


In regards to GPU, I will try GPUZ, please keep in mind the following:

1. I had an identical build and Windows 7 x64 before RMAing all parts and NEVER had that NVidia BSOD with default Windows 7 NVIDIA drivers or Latest x64 bit drivers from NVIDIA.
2. BSOD occurs on both, native WDM Win 7 drivers and latest 196 drivers from Nvidia

What is a good idle temperature for Nvidia GTX260 by EVGA?

THANKS!!!!
m
0
l
Related resources
a c 218 à CPUs
March 14, 2010 9:45:05 PM

A. AS5 takes up to a year to cure under normal everyday usage.

B. Have you tried taking off the side panel and using a desktop fan to blow air inside the case ?

http://www.vornado.com/ProductDetail.aspx?ISC_Category=...

C. What is internal case (MoBo) temp w/ side panel on ?

D. driver issues

1. Uninstall nVidia Drivers
2. Use CCleaner registry tool to clean out anything that says nVidia on it.
3. Reboot

E. Did you just pop those drives into the machine after getting all your new parts or did you do a fresh install of Windows ? MoBo change requires that you wipe OS drive and do fresh install.

F. Consider these to aid cooling around RAID card

http://www.newegg.com/Product/Product.aspx?Item=N82E168...
http://www.newegg.com/Product/Product.aspx?Item=N82E168...

G. Have you disabled auto restart after BSOD ?

H. Have you looked at *.dmp files ? Event Viewer Messages ?




m
0
l
a b à CPUs
March 14, 2010 10:09:51 PM

How much thermal paste did you use?

The GTX 260 shoudl idle between 40-55c, with 60c being tolerable.
m
0
l
a c 127 à CPUs
March 14, 2010 10:17:17 PM

logika said:
Thanks for the prompt response!

Yeah, I had constant RAID Array failed with latest Intel Storage Matrix and Drivers. 8.9, 8.8 was better but I still had to rebuild and repair once or twice a month.

Before replacing the CPU fan, I used the Thermal Compound Remover and Surface Purifier on both the CPU and heatsink (Coolermaster V8). I specifically bought the Arctic Silver 5 compound and applied it this time. My case is a Cosmos S from Coolermaster and has plenty of airflow and fans! How come I could be idling at 47C according to motherboard BIOS hardware monitor?


In regards to GPU, I will try GPUZ, please keep in mind the following:

1. I had an identical build and Windows 7 x64 before RMAing all parts and NEVER had that NVidia BSOD with default Windows 7 NVIDIA drivers or Latest x64 bit drivers from NVIDIA.
2. BSOD occurs on both, native WDM Win 7 drivers and latest 196 drivers from Nvidia

What is a good idle temperature for Nvidia GTX260 by EVGA?

THANKS!!!!


The CPU should not be idleing that long and even though AS5 takes a year to cure, should not cause temps to go 47c idle. Normally the cure would take a few c off the normal before the cure sets in because it has better heat transfer properties.

The case is fine but I think you might want to recheck your heatsink and make sure its right.

What kind of RAID are you doing? I have a P35 mobo right now and run a RAID 0. Are you using the in Windows software or do you actually pres CRTL + I to access the chipset RAID configuration? In my opinion, the best one is the non Windows based one. Its like the BIOS. More stable for certain things, like overclocking and such.

Also, what does CPUz report your voltage at? That might be a factor to the CPU being too hot if it is pushing more voltage than the CPU needs at stock, sometimes mobos will put higher than needed at stock when set to default voltage. My mobo did for my Q6600. It would set the voltage at 1.35 when I can run it overclocked to 3GHz at 1.25 or less.



Looks like 50c idle for the GPU and probably 70-75c for a GTX260 under load. If your idle is higher than that then the GPU might be bad or the fan doesn't kick in right when it needs to.
m
0
l
March 14, 2010 11:10:40 PM

Thanks for all the great replies people. I am trying what you have suggested now.

1. I am running RAID 5 with 3 x 750GB WD Caviar
2. I have applied a fine, thin layer of Arctic 5, about a paper thick.
3. My heatsink was installed properly, I am not sure what I would adjust if I were to take if off? It is sitting properly and i tightened it pretty well.
4. I am not sure why to clean NVidia drivers if I reinstalled OS 3 times, clean and same BSOD?
m
0
l
a c 127 à CPUs
March 15, 2010 12:02:47 AM

Hmmmm.....

Did you do the RAID in software or in the chipset? That might make a difference. As I said, its comparable to software overclocking vs hardware overclocking. Software tends to be less reliable.

Have you found out what the temps of your GPU are running at?
m
0
l
March 15, 2010 1:22:46 AM

Ok, have some updates.

I tested the new memory from Corsair with MemTest86+ and got no problems.

I ran GPUZ and received temperatures of 47-50C degrees on GPU Temperature and another GPU temperature was showing 40C almost consistently. Load was always at 0% and fan speed was at 40%. It seems like the temperatures are normal, but the card is BSODing everytime after 10 minutes of loading Windows 7. I do not think it is as much of a software issue as I have:

- reinstalled windows 7, 4 times now, clean format everytime
- tried WDM drivers, newest drivers and everything in between
- i had an identical setup (except the addition of the Adaptec RAID 5805) and not once had the BSOD related to NV files.

Should I just RMA this card ASAP? Seems like a good solution to me.

Also, both CoreTemp and Real Temp are reading about 35C average on the Core i7 CPU while the BIOS is showing 47-50C? Could anyone comment as to why these values are 10C apart?

I was also able to resolve the Adaptec overheating issue by updating the firmware on the unit. My only concern seems to be the dreaded nvlddmkm.sys BSOD everytime...this error seems to be caused by so many things, I spent 8 hours searching the web and the only viable info seems to be to RMA the video.

Thanks once again for all the input, it really has given me hope.

PS: In terms of my RAID, I have built the RAID in Intel Utility before and now that I have Adaptec 5805, I have built the RAID in their Array utility and then fresh installed Windows 7 Ux64
m
0
l
a b à CPUs
March 15, 2010 2:18:30 AM

amd+ati = no stress
m
0
l
a c 127 à CPUs
March 15, 2010 4:19:59 AM

xaira said:
amd+ati = no stress


Thanks for the help. I am sure this solves his issue. Yea one thing.... AMD+ATI /= no stress. What about the people with a ATI HD5K series that had grey screen issues across Windows from XP to 7? I had BSoD in Vista due to the GPU driver being immatur on my HD2900. This is called bad luck, for this guy. 99 times out of 100 there are little to no issues with any hardware.

Next time, stay out if you can't help.

And to the OP, I would RMA it. Seems like you might have just gotten a bad GPU because if you do a fresh install of Windows 7, then drivers it should not BSoD. In fact you might be one of the few to be able to get a BSoD in 7 so easily.
m
0
l
March 15, 2010 3:35:00 PM

Once again, jimmy and everyone else, big thanks for spending your time to help me!

I just went and bought a 9400 GT from PNY and popped it in. 4 hours now and no BSOD.

I spoke to EVGA rep and they tried to convince me not to RMA the card but test my PSU and MOBO, since this is the second RMA. Sounds a little fishy to me, considering previous EVGA card wasn't doing this and I never got an NV related BSOD prior ro RMA, but did have monitors go into power save mode (reason for RMA in the first place).

I think I am going to RMA the card regardless, considering all is pointing at it. Do you think it could be my PSU as well? How can I test it? Thanks!
m
0
l
March 15, 2010 4:36:31 PM

Hmmm highly unlikely that it could be your psu,not impossible tho,do the rma on that gpu,I believe that you can test your psu by grounding out the middle green to any black,or if you have access to a vom meter test it in situ,OTOH if it were your psu you should have got a message from the gpu complaining about insufficent power,my 8800gt did that once...:) 
m
0
l
a b à CPUs
March 15, 2010 6:40:14 PM

^ - Do not need to short out "I'M OK -Green wire on 20/24 pin connector" line since your computer is booting. But if you have a VOM you can measure the Red/orange wire to black at a molex connect. I always recommend this to verify "software" values for the +5/+12V.

To check your PSU:
Download, install and run CPUID HWMointor. Verify @ idle that the +5/+12 V are nominal, if not you may need a different program.
With HWMonitor running, run Prim95 Blend test. Note Min values for the +5/+12V.
Min exceptable values: +5 => 4.75V, +12V => 11.4 V (thats the spec, I use 11.6 as min). This loads the CPU.
Next run a program to tax the GPU, I use Furmark. again note min +5/+12V

If these voltages are within spec, you are about 95% sure it is not the PSU!!

Aslo not max cpu/Gpu temps as these are the temps you want to look at verse Idle temps.
m
0
l
March 15, 2010 10:32:20 PM

Very cool! Will test my PSU just in case as well.

I have a very annoying problem, don't want to start another thread in software, maybe you can comment.

While booting up Windows 7, my mouse and keyboard even though they are installed and are getting power, take about 5 seconds at login screen to start working. I never had that problem on XP, but it is an issue on Win 7 lately and it extremely annoying. Any comments?

They are both USB, not wireless and no other 3rd party devices except mouse and keyboard are connected to PC.
m
0
l
a b à CPUs
March 16, 2010 2:25:06 AM

It could be background programs that are loading and hoging cpu/resources. Not noticed it on my systems. I usually turn my systems on and go do something, like get a cup of Java, even on my system with a SSD that only takes 30/45 sec to boot.
m
0
l
a c 127 à CPUs
March 16, 2010 9:04:57 AM

dokk2 said:
Hmmm highly unlikely that it could be your psu,not impossible tho,do the rma on that gpu,I believe that you can test your psu by grounding out the middle green to any black,or if you have access to a vom meter test it in situ,OTOH if it were your psu you should have got a message from the gpu complaining about insufficent power,my 8800gt did that once...:) 


A BSoD driver file error from a PSU? IDK. I mean if he is idling or doing something low end or just 2D, it shouldn't load the GPU. Plus he didn't have this problem before hand with the same GPU (GTX 260) that he RMAed since the monitors would go to sleep.

I think its a bad GPU. Hopefully it is and he wont have to change the PSU as well.

As for the mouse/keyboard thing...... I have not had that. I would think it could be the USB driver being slow to load. Are they generic parts or specific brands? Might want to try reinstalling the USB driver.
m
0
l
March 16, 2010 3:57:06 PM

jimmysmitty said:
A BSoD driver file error from a PSU? IDK. I mean if he is idling or doing something low end or just 2D, it shouldn't load the GPU. Plus he didn't have this problem before hand with the same GPU (GTX 260) that he RMAed since the monitors would go to sleep.

I think its a bad GPU. Hopefully it is and he wont have to change the PSU as well.

As for the mouse/keyboard thing...... I have not had that. I would think it could be the USB driver being slow to load. Are they generic parts or specific brands? Might want to try reinstalling the USB driver.



Jimmy man...you rock with your posts.

Yeah, I say bad GPU for sure. I hope the RMAed one will be good and I am sure it will be.

In terms of keyboard and mouse, they are Saitek kb and Dell mouse. I am using generic Windows drivers, just puzzled as to why they would be loading slow on a fresh installed Windows 7 on such a powerful PC. I don't have any other external drives or soft installed on this machine yet. It is a brand new install and lagging...strange isn't it???
m
0
l
!