Please HELP!

[EagleOne1962]

I have a pc with the following installed; CPU Q6600@2.4GHZ, MOBO: ASUS P5KPL-AM SE, RAM DDR2 800MHZ, Graphics card ASUS EN9800GT, PSU 500W. All drivers are up to date. I would like to know, why does my games run smoothly for about an hour and then the graphics is getting sticky and the sound stuttering?

 

[Rustyy117]

My guess would be that either your Graphics card or your processor are getting to hot when playing games after a certain period of time, What will happen is when your GPU or CPU get to hot they will underclock themselves in an attempt to cool themselves down.

OR your GPU, CPU or memory (or a combination) are unstable.

Run some tests to see if your computer is stable and cool enough. Use GPU-Z to monitor your GPU's temps whilst running Furmark for 30mins. Use OCCT to stress your CPU and use Real temp to monitor the temps. And use Memtest to test your memory.

If everything is stable and with good temps, then it might be a driver issue

 

[wa1]

Check your temps...

 

[technuttso]

Check your temps. And about your RAM.. u have 2Gb or 4Gb of RAM? Is better with 4Gb, allot better.

 

[apache_lives]

what power supply - 500W could mean geneic aka ~250w which is borderline

test your hardware and see what temps you get, you want to stick under ~68 for cpu and ~88 for video

 

[EagleOne1962]

Hi there, thanks for the info. I ran all the tests and everything looks to be running stable, accept for the CPU that is running a little bit hot-69 degrees. That was only on the first test, after that it dropped down below 68 degrees. Maybe I must also look at a bigger power supply. What driver do you suggest I must look at?

 

[EagleOne1962]

I would just like to say thankyou to everyone elses input. You guys are great!!

 

[Rustyy117]

You should use the latest nVidia driver, All you have to do is go to nVidia's website and download the latest driver for your graphics card and OS.

69 degrees isn't too bad, but I wouldn't let it go any higher. Make sure you run OCCT for atleast 8 hours to ensure stability.

What were your GPU temps like after 30mins of Furmark?

 

[EagleOne1962]

Ok, I've got the latest nVidia driver already installed. The temps stayed around 45 degrees and only went up to 47 degrees

twice then back to 45 degrees. With the cooldown test it went down to 36 degrees. On the graph result of the CPU it said

there is four threads. I'm not sure what to make of it, cause I'm not 100% sure what the correct spec should be. Or maybe

I don't know what to look for. I will make sure to run OCCT for 8 hours or so. Thanks

 

[EagleOne1962]

6206@live.co.za I almost forgot, I've got 4GB ram installed on my PC and it's brand new. That was the first

thing I've changed when the problem started a month ago, along with the motherboard and hard drive. The only old

hardware is my CPU. Is there a way I can include the CPU test results so you can take alook at it maybe?

 

[Rustyy117]

its very unlikely that a hard drive would cause a problem like this, and if your CPU is at stock and under 70C then its also unlikely to be the cause of the problem.

When you installed your new RAM into your motherboard did you manually change the voltage and timings to what they should be?? This is important!!!

You can find the memory timings and voltage infomation on each memory stick.


If you can run 8 hours OCCT then the memory is stable (OCCT CPU test will stress memory aswell).

Also what make and model is your power supply?

 

[EagleOne1962]

no I didn't change the voltage and timings, I thought my PC would detect the settings automaticly. But I'll be sure to check

on this thanks.

My power supply is; Superchannel S500 480W, model-LC-8500BTX

I will only be able to run OCCT on Wednesday when I'm not working.

 

[Rustyy117]

When I installed my RAM into my motherboard, my motherboard set the RAM timings as 6-6-6-16 and the voltage at 1.5V and 1066mhz, When infact my RAM is rated at 7-7-7-21 at 1.6V and 1333mhz. Motherboards are pretty poor at setting voltages and timings
Another example is my motherboard set my Core i7 920's voltage at 1.28V, when its able to run stable at just 1.10V.

I've never heard of that powersupply and apperantly neither has Google (0 results found).

So if i were you i'd do the following

1: Set the correct Timings and Voltage for you RAM and CPU.

2: Get a Good quality PSU like a Corsair 550HX or a OCZ ModXtreme.

3: In the morning start OCCT CPU test and then go to work, Assuming you have a 9 till 5 job when you get back it will be done. (don't worry, if it fails or gets to hot, it will shut down the program).

If OCCT runs for 8 hours with no errors then its pretty safe to assume that your memory, motherboard and CPU are stable (and not the cause of the problem). Then its either the Game you play ( for example My PC is stable, but Fallout 3 crashes from time to time, but no other game do this.) A Video driver, Graphics card stability OR your PSU isn't up to the job. I think the latter is the cause to be honest.

 

[EagleOne1962]

I will do just that. I want to thank you again for all your help. I really appreciate it. It is nice to know, that there is still

people out there who cares. You don't know how many people I've asked, not even our PC techs. this side could help me.

Very poor indeed. Thanks again from the bottom of my heart, you're the best!!!

 

[Rustyy117]

N'awww thanks

Let me know how the OCCT CPU test turns out

 

[EagleOne1962]

Sure, I will let you know tomorrow knight.

 

[EagleOne1962]

Hi there, here are the results;

CPU Temp 1 - 69 degrees
" " 2 - 66 "
" " 3 - 61 "
" " 4 - 62 "

GPU Temp - 41 degrees

Vcore - 1.24 average

3.3V - 3.36 average

5V - 4.77 - 4.88 average

12V - 12.88V average

Tell me what you think of this tests.

 

[Rustyy117]

so the test was still running when you got home? Those temps are OK.

I'm not to sure about your PSU, There seems to be some big variations on the 12V and 5V

41 C GPU temp was Idle temp right? Use the OCCT GPU stress test or Furmark and monitor the temps with GPU-Z. I'll run the OCCT PSU test to give you some idea of what someone else is getting, I'm using a quality Coolermaster Real Power M 700W

 

[Rustyy117]

Here's a link to my results for OCCT PSU Test, Only ran it for 14mins because I have to cook the dinner

Dunno why my 12V rail is so low. It was stable though.

The PSU test will stress the GPU and CPU at the same time, abit like if you were playing Crysis on max. And so will make sure everything it stable.


Another question, did you change the timings of the RAM before you ran the test?

 

[Rustyy117]

http://s676.photobucket.com/albums/vv125/iNoXious/

Ooooops forgot the link, to busy eating XD

 

[EagleOne1962]

No, all tests were completed by the time I got home. The 41 C was the result I got. I did not set the timings on the RAM. I

only have the option of auto, 1.8V, 2.2V and 2.8V in the bios. I must be honest with you, this is all new to me. I've never ran

these tests before so forgive me if I throw you of the track a bit. I know a little bit about the hardware but very little about

the software. I'm still learning. So looking at your results I can see a big difference between yours and mine. I won't be able

to tell you if there's a problem or not though. Is there a way I can get this results to you, so you can take a look at them?

 

[Rustyy117]

If you still have the graphs then you can use a image hosting website like photobucket and send me the link (like i did above ^^)

So did you set the OCCT test to run for 8 hours only?

 

[EagleOne1962]

Yes I set OCCT for 8 hours only before I went to work this morning. I leave home for work around 6 o'clock. I work from 7 to 4.

I left everything as is, and pressed the on button to start the testing. I did not change any settings.

I will try to put the graphs on photobucket like you did.

Thanks.

 

[EagleOne1962]

http://i899.photobucket.com/albums/ac198/EagleOne_album/2009-12-15-13h56-CPU1.png

Here is the link to my graphs.

 

[Rustyy117]

Well i dnt know enough about PSU's to tell if thats the problem or not, Sorry : /

Atleast we know that the CPU, RAM and Mobo are stable. : )

And if you say that 41C is as hot as your GPU gets under Furmark or OCCT GPU Test then thats not the problem( unless the GPU is defective in another way)

Does this problem happen in all games you play? or just 1?.

It could well be the PSU, but i just dnt know enough about them