PC slows down and occasional crash

gbu

Posts: 33   +0
I have recently upgraded(almost rebuilt my PC) with new CPU, RAM, Motherboard, PSU etc and so to compliment the new hardware i decided to switch from XP to Win 7(64Bit)

Alls great but i do notice a slowing down if im installing a program or Windows update is running etc

I have 4GB RAM this make:-
h**p://www.microdirect.co.uk/Home/Product/21226/Corsair-memory-4GB-TwinX-XMS2-6400-DDR2-Kit--2-x

ASUS Motherboard
h**p://www.misco.co.uk/applications/SearchTools/item-details.asp?EdpNo=381857&Tab=2&NoMapp=0

Quad Core CPU
h**p://www.misco.co.uk/applications/SearchTools/item-details.asp?EdpNo=371616&sku=Q180280

The lagging isnt to bad but I used to have only 2GB with XP so thought having now 4GB that i should be able to run more programs at once, dont get me wrong i can run other programs but theres a definite pause and lag if i try to run another whilst one is still running.

I have tried removing each RAM stick but its hard to tell as the problem really only occurs(or at least more noticeable) when a big MS update has happened.
There was one yesterday with lots of patches/fixes etc and so it happened again.

Just after(maybe an hour) my pc crashed and i had these 2 messages appear from the Check For Solutions pop-up in Windows 7:-
Unfortunately the Solutions center isnt responding so they have no answers for me

Windows
Problem: Shut down unexpectedly
Files that help describe the problem:
072110-85519-01.dmp
sysdata.xml
WERInternalMetadata.xml
View a temporary copy of these files
Warning: If a virus or other security threat caused the problem, opening a copy of the files could harm your computer.

Windows
Problem: Shut down unexpectedly
Files that help describe the problem:
080310-31012-01.dmp
sysdata.xml
WERInternalMetadata.xml
View a temporary copy of these files
Warning: If a virus or other security threat caused the problem, opening a copy of the files could harm your computer.


I also took these screen grabs from my task manager as the pc was updating, showing CPU and RAM usage, took about 5 minutes for Task Manager to load as everything was so slow;_





Many thanks for looking and even more thanks if you can help me solve this annoying problem :)
 
What is the make and model of your Video card and Power Supply?

Did you use the driver CD that came with the motherboard?

Even when I buy a new device that requires drivers today and the product was released this week, I always go to the manufacturer's Web Site and get the latest drivers, you never know when the CD/ DVD was made and if there are updates available, with some motherboards and Video Cards, there are updated drivers released Weekly till all issues have been addressed.

Go to ASUS and check for BIOS and driver updates, same with your video card.
If the installation is new we might have you start over.

Start with the BIOS update, reset the defaults, once back in Windows, install the latest chipset drivers, do manual Windows update till there are none left, install other drivers (ignore "Windows Update" hardware updates, as in Video drivers, rather get them from the manufacturer's site).

Going to assume for now you were comfortable with building/ upgrading your system.

Let us know.
 
What is the make and model of your Video card and Power Supply?

Video Card is Asus Radeon HD 4350 Silent PCIE HDMI 1GB Graphics Card
h**p://www.amazon.co.uk/Asus-Radeon-4350-Silent-Graphics/dp/B002NQVOKE

PSU is OCZ stealthxstream 500w
h**p://www.ocztechnology.com/products/power_management/ocz_500w_stealthxstream_power_supply

Did you use the driver CD that came with the motherboard?
Yes used the cd to install all drivers etc


Going to assume for now you were comfortable with building/ upgrading your system.

Yep not to bad lol, building was easy just not to sure on configuring :)

Never used drivers direct form the manufactures website before, how can i tell which i ones i have and which are the newer ones?

Thanks for your helpfulness if you can just point me in the right direction for the driver updates that would be great :)
 
Well, Asus web site will have all the drivers for your board and video card; you can check the driver version from the CD.

Judging by the age of the board (not saying it is OLD) there should be BIOS and drivers updates.

Go to www.asus.com, select your region then click on Service at the top and select Support, next page click on "Download product BIOS, Drivers, User Manuals and Utilities here" next page right hand side about half way down, select your product, (do motherboard first then come back and do Graphic Card), then select series and Model, it will ask you for OS first then show you available drivers.
Always start with the Chipset/ INF and all Chipset related drivers then do Windows update, then the rest of the drivers as in Audio and video.

I have the 600W version of your exact PSU, I just don't like the 500W 30 Amps each on the +5.0V and the +3.3V for a PCI-E system, 20A to 25A would've been nough leaving more for the +12V rails. My 600W has 36A on the +3.3V which I love since it is in an AGP based system ;)

Let's see if the BIOS and driver update helps with the symptoms first before we go any further. Read your manual about ASUS EZ Flash2, all you need is the BIOS file extracted to a USB stick preferably 512MB in size at least.
 
Ive checked the drivers and some are out of date so will install those today.

There is a BIOS tab BIOS (3) [BIOS History]

1. Says First release
2. Says Update LAN option ROM.
3. Says Update ExpressGate to v1.4.9.5.

Theres is another tab BIOS-Utilities (1) that says:-
Afudos BIOS update tool V2.38

I dont mind installing drivers but not sure about doing the BIOS?
Do I use all of these?



Outa interest is there a reason why you think a driver(or outdated driver) would be causing this kinda trouble, is it a known symptom of lagging and random reboot to be caused by old drivers?

On another note im wondering if you could be onto something with the PSU, I checked out the event viewer to see if there was any info on the shoutdowns and noticed a lot of Event 41 Kernel Powers errors.
The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.

Now I have no idea what this is or whether its normal although I suspect its not, I have included a screenshot for you to see.

Could it be a faulty PSU or not enough power?
 

Attachments

  • error.JPG
    error.JPG
    81.8 KB · Views: 3
With that motherboard you don't need any the "BIOS Utilities" to update the bios, doesn’t look like it has been any major changes with the BIOS updates, leave that for now.

I basically don't trouble shoot anything till I know the hardware and software and their state, might take a bit longer my way. There is an update to the Chipset drivers, this is an important part of your motherboard that that controls everything, this might not have anything to do with your immediate problem.

Could you give me a reading from CPUZ
under download, get the (.zip, no installation) unzip, run, go to the last Tab (About) save a text file and attach to your next post. This will have some voltage readings as well.
 
Ok have updated the chipset driver(it just ran and didnt need a reboot or anything so presuming it installed ok)

Also updated Graphics Card.

Heres my text file from CPUZ
 

Attachments

  • JAMES-PC.txt
    88.3 KB · Views: 2
Do you have another power supply to test with?

Voltage 4 8.09 Volts [0x91] (+12V)

This is the second system with the +12V reporting 8.9V !!!

See if you can remove some hardware, like optical drives, additional hard drives, even case fans only temporarily to see the difference in a new CPUZ run; try to drop as much load.

If you are comfortable with Volt meters, we can take a reading from one of the 4 pin molex that powers the hard drive, Yellow is +12V, Red = +5.0V and the black is Ground.

Let us know.
 
Ok have disconnected 1 hard drive, 1 dvd drive, 2 case fans a printer and audio speaker connection.

I dont have a volt meter so cant test that part out.

Have attached the new txt file.

Just had a look and its now 7.9

Was you expecting it to go up or down with less hardware connected?

Im now going to re-connect the components and its gone back up to 8.09.
 

Attachments

  • JAMES-PC.txt
    88.3 KB · Views: 1
That is not considered a big change, if the monitor was actual time, it would go up and down. I was trying to drop some load, didn't expect it to go up or down.

There is now evidence that CPUZ might be dropping a digit (the first) need to check the same reading for +12V in the BIOS hardware monitor, as well as another Free monitor.

PC-Wizard from the same guys (CPUID) is free.
Don't try Everest, the trial version doesn't give you the +12.0V reading.

Best would've been with the multi meter.

CPUZ has never been wrong for me, all the instances with wrong readings were with systems that already had some problems, more the reason to test the PSU on its own disconnected from the system.

The meters go on sale here for $9.99 in different stores where I live! New ones are not much higher, not talking bout Fluke and Pro series! I paid around 30 for my digital meter.

Try the BIOS, we need to be sure before going any further.
 
I have also borrowed a volt meter from a friend, hopefully should do the trick.

I can try CPUID as well if you think it will needed?

Ok so do i try with everything connected or do i disconnect then try, dont want to blow myself or PC up lol
 
First test with the system running should be OK if you are careful, just use a spare hard drive power connector, if readings are lower/ higher than what they are supposed to be, then we do it with the PSU alone.

Be back in an hour with instructions how to test the PSU disonnected from the system..
 
If you are comfortable with Volt meters, we can take a reading from one of the 4 pin molex that powers the hard drive, Yellow is +12V, Red = +5.0V and the black is Ground.

Let us know.

Ok so have now tested wit the volt meter and i have +12 Volt and +5v using one of the spare molex connectors.
 
So you mean the +12V and +5V are OK? Is the system any different after BIOS, Chipset and Video driver updates?

Just to make what I said about your power supply clear... I think your power supply choice is great, pointed out the fact that they wasted some output on the +5.0V and the +3.3V ((30A X 5V = 150W) + (30A X 3.3V = 99W) = 249W) then the max combined 5.0V + 3.3V = 152W, this is 97W. That is closer to like 20A each.

E414372E-AC7F-47D0-A8708AE6EAD25F75.jpg

Picture courtesy of ncix.com

+5.0V, +3.3V and the +12V rails max combined output is 481.5W.
481.5W - 152W = 329.5W (max combined for the 2 +12V rails) divided by 2 = 164.75W or 13.73A each, that is still like 27.5A for the +12V.

There should be enough power. Your Q8400’s system power consumption under load is around 210W and at idle is around 135W (total system)
The 4350 is about 160W idle and 195W under load (total system)

Your RAM should be set to 5-5-5-18 @ 1.8V in the BIOS running at stock 800MHz
I would try running them in sync with the FSB at 667MHz for testing purposes.

You have DriveLED, has it reported any issues with the hard drive?
Is your case temperature OK with enough cooling and air flow, especially since the 4350 has passive HS?

There is still a reason CPUZ reads the +12V wrong from the board.

I would download Burnin Test, the trial version allows you to run 15min tests, you can do this back to back after setting your preferences as well as the load, try the default 50% load then go all the way to 100%, if the board has a glitch, it is best to find out now. Select hard drive, RAM, CPU, 2D and 3D and so on.

If all is well, then we might adjust some voltages a bit, the regulators on the board might be asked to deliver certain voltage and it is common/ normal for them to be under or over at times.
 
DriveLED isnt reporting any problems and the fan temp gauges are all running fine and case temp is ok too.

Is there a reason why CPUID is showing a drop in voltage when the gauge is reading it correctly?


Could the RAM be a problem being only 800Mhz should i of gone up to 1066?
im running them in Dual channel in the Dimm A1 and Dimm B1 slots does that make a difference ?

I'll run the burning test tomnight and report back what it finds.
 
Well, your not the first with CPUZ, it gets information from your board. Either CPUZ has a glitch or the board doesn't report back. There are many other monitoring software you can try.

You can try A2, B2, should make no difference, I populate the first sockets by choice.

I meant for your RAM to run at 333MHZ (667) a notch slower than 800MHz to test with "relaxed" timings.

After all tests are done, you can run them at 1066MHz with 2.1V (now it is 1.8V for 800MHz)

Begining my trip to Virginia, be a FEW hours before next reply :)
 
Havent had the time at the mo to run the burnin test, hopefully will in next day or so.

My friend has Everest Ultimate Edition and so ive run the volt test with that.



They read pretty much the same as CPUID but are both different from an actual volt test, could it be cos im testing a free port and not one in use?

So is there no real benefit from running the RAM Dual channel, i put them in that way only cos i thought it was a better way lol but really have no clue as to why or what it means to do so?

This is the benchmark from Everest for the Memory:-




Didnt realise you could change then to run at 1066mhz is it safe to do so or would changing them to 2 new sticks that run at 1066 be better?

P.S Many thanks for taking the time to help me, enjoy your trip to Virginia and do not rush to reply back, i can wait :)
 
The 12V supplied to the board goes through many circuits to reach where the monitoring software gets a reading! This could point to a faulty component on the motherboard on it's way out!

Burnin Test on 100% load might finish it once and for all or prove to be just a mis read.
It could also be the PSU as good as it is, there are bad ones out of the box! look in to RMA before it is too late.
The 12V supplied might bend under load, in this case less load than it is supposed to be able to handle (2 X 18A or 27.5A combined)

Dual channel is better than single channel, think of it as an 8 lane highway vs.a 16 lane highway. In this case the difference between 64bit and 128bit.
 
Have now run a few tests with Burnin Test

Ran the first test at 50% and got some fails but looking at the log i think it was for the floppy drive(i dont have a floppy drive installed but it was trying to read it)




Ran again at 100% this time and got another error for the 2D Graphics.





Although no RAM errors where shown i thought id do a separate test just for the RAM and this time did get some errors:-
Error Verifying Data in RAM





Could these be any of the reasons?
 
Can you set the memory to 667 (333MHz) in the BIOS and run another memory test, either Memtest or Burnin Test.

Would be good if we could run each module on it's own.
Then run a 2d video on it's own

You do need to set your prefrences, so, no floppy and so on, think you already got that.
 
I tried the 2D Graphics on its own before i set the RAM to 667 and this time it passed.



Set the memory to 667 in the BIOS and ran the test again, this time it passed:-
 
Like to lnow how the system runs with the memory @ 667MHz, can you post a CPUZ with this setting?

If things are OK, we will set the memory to 800MHz and increase the voltage by 0.05V at a time till it is stable or passes the tests, even with 1066MHz all the way up to 2.3V if need be and play with some other voltages before deciding to change any component.

First lets see if the lag goes away with memory relaxed a bit. Memory voltage is supplied by the +5V rail, that leaves the +12V rail, we will test that using CPU and 3D video test in Burnin Test with 100% load!
 
What was a slightly weird was that there was a MS update yesterday(before I altered the RAM in BIOS) and strangely the lag was almost none existent!
This was the first update ive had since updating the graphics card etc so that may of been the cause.

So now do I leave well alone and hope that the lag has now gone and the update helped or do I continue testing seeing that I am getting errors from burnin test etc?

Heres are the burnin test results with RAM set to 667 and just testing 3D Graphics and CPU at 100% both passed and the CPUID report.

I noticed though you asked for the 3D test and not the 2D test which had the fails on it before?
 

Attachments

  • JAMES-PC.txt
    88.3 KB · Views: 3
  • cpu3d.JPG
    cpu3d.JPG
    31.3 KB · Views: 3
3D will draw more from the +12V rail the reason why I asked to test it. Individual tests passed which is good, but next time you have about an hour, run Burnin Test, disable Network, Floppy, CD and such but leave all other ccomponents checked and test all together with 50% load, do this back to back at least 4 times. The trial version limits to 15min tests.
I use the full version and run tests over night in loops!

Next, increase the load to 75% and run a few tests back to back and then 100% load.

It is OK to get some errors doing tests back to back under high load. If they pass with these settings, then we concentrate on your software and drivers and you should be good to go.

There is still a BIOS update, I like to keep all hardware firmware as well as software patches and drivers up-to-date.
 
Ok am running the test now(still with RAM set to 667)
Have enabled:-
CPU
RAM
3D & 2D
Video
Sound

Will do 4 consecutive cycles then post the report

these were the BIOS updates,

There is a BIOS tab BIOS (3) [BIOS History]

1. Says First release
2. Says Update LAN option ROM.
3. Says Update ExpressGate to v1.4.9.5.

if you think its worth updating i will but have never done a BIOS update before so not sure how and dont want to muck anything up :)
 
Back