PC decides to go sloooow.

Messages
2,748
Name
Julian
Edit My Images
Yes
Looking for a bit of help please.
I'm wondering is there any way to 'stress test' various components on my PC?
Basically it just suddenly becomes unresponsive to the state I have to power it down and reboot it - I cant go the normal route of switching off properly as its too unresponsive so I just have to switch off by holding down the power button.

I've run all my antivirus tests and everything comes up clear ( Eset and Malwarebytes plus a few other toolkits to remove any problems).
Its very odd and random when it happens.
My PC is a AMD Phenom II 3.2ghz with 8gig RAM running Windows 10. I have replaced the hard drive as I thought that may have been the issue.
I was hoping the upgrade to 10 might have solved this problem as it was the same under Win 8.1

Also, sometimes the PC has trouble booting - it will get past the BIOS part of the boot sequence and then will just freeze with a black screen - only powering down will reset it from this state too. Just to note, the PC usually becomes unresponsive while on the internet but that may just be coincidence as that's what its normally used for anyway..

I'm wondering if it could be a RAM problem which is why I thought of any way to stress test individual components? I've stripped everything down and reseated everything, new thermal paste on the processor, disconnected a couple of extra storage drives, sucked out any dust etc.

I'm really at the end of my knowledge here and wondering if anybody has any ideas?

Thanks

Julian
 
I am sure that people with more knowledge than me will be along shortly, wonder if it could be power supply or thermal throttling?
 
Maybe a long shot but have you tried changing the CMOS Battery?
 
I did wonder about the power supply which is why I disconnected two hard drives I used as extra storage (internal drives not external) and that seems to make no difference. Thermal throttling - it runs very cool and I replaced the CPU thermal paste and de dusted everything.
Not tried the CMOS battery yet as it seems to keep time ok when its switched off - I thought the clock usually goes funny when that gets low. Might be worth a try though

Thanks guys - anyone else got any ideas I could try?
 
I would re-do the CPU again, take it out and re seat it, make sure all old thermal pasts is cleaned off and re-do

If that doesn't help do you use on board graphics or a separate graphics card? if a separate graphics card, take it out and use on board for a while and see if that helps.

If the above doesn't work, take out RAM and re-seat it, also there may be a bad stick of RAM so test for that also
 
Do you ever get BSOD's with it? (Blue screen of death).
Do you have a similar PC in terms of spec and age you can use to fault find with components?
Is the system overclocked in anyway?

It's very easy to test the memory. MemTest. Just google it. Burn yourself a DVD/CD and you boot up with it and it will run an intensive memory test giving output as to whether the memory is ok. Any alerts are clearly shown in red and then you know to replace memory.

CPUs rarely go bad unless they have had a hard life with being overclocked or if they have always ran with insufficient cooling. So check the CPU fan is working.

You could try booting into safe mode to see if the problem remains. It does sound more hardware related though. The best thing you can do is to try to boot up with minimal peripherals plugged in. Just bare motherboard, cpu, one memory module and on board graphics if you have it, and see if you still get the problem. Then start adding stuff till you see the problem.
 
I would suspect the RAM, mine is self build and I had to tweak the BIOS to slightly over power the ram when first built to maintain stability.

Later doubled the ram and quite quickly hit a stability issue with random lockups. I stress tested the ram without any issues showing.

Finally did a ram swap out test and determined the original 8GB were at fault but that all the slots were fine. Since just keeping the known good pair total 8GB in all has been to date aok

So by all means stress the ram but not all such tests will reveal a problem.....only an empirical test shows where the problems lays. Is your 8GB one or a matched pair of two that need to run in dual mode sets? How many slots does the motherboard have as in cases where you need ram in pairs the slots are also to be used in set pairs. Give some more details about the PC and perhaps more usefully the motherboard itself?

PS I think it was amongst others this programme I used yo stress test. http://www.passmark.com/products/bit.htm

I also think I used a DOS based tester but that needs you to be able o boot of off a USB stick.
 
Last edited:
Sounds to me like the motherboard is on the way out.
If you can't find any capacitors (especially around video and power supply) that are dodgy (leaking, blown, domed) to replace, then it's a new m/b.
My reason is your comment about getting past bios which, as you know, is the first step and does not depend on software loaded.
It's also possible that you have a power supply unit which is about to fail but I would check the motherboard first.
 
Depends on age, specs.. maybe a time for a upgrade? SSD? best upgrade iv ever done. Im running a intel i5 2500k, 128gb x 2 in Raid0, Windows 10 is also a god send : )
 
It's likely it's the OS. Boot to another OS on a USB stick and use it for a while to eliminate a software problem. It's easy to do.
 
Last edited:
It's a home build about 5 years old I guess. Hard drive is two weeks old. Separate graphics card and runs very cool. System is clean of dust and debris and all vents are clear.
Problem is it can go for a few days without a hitch, other times it will go weird once or twice a night. Sometimes it's not as bad as others when it does it and just seems to slow down a little ie typing seems to lag a bit. Other times (mostly) it's full on slow and unusable do I have to reboot.
Once rebooted it's normally fine.
Don't think it's thermal related as it does it on first boot of the day sometimes so not hot at all.
Mobo is pretty basic one - can't remember at moment - with two ram slots taken up by two 4 gig sticks - the max this mobo will take unfortunately.
 
It's unresponsive as in slow - mouse pointer stutters all over the place etc. It still 'works' but very slow.
Bringing up task manager the processes aren't being used that much.
I've not done a clean instal (yet) as it does it before it gets to the OS sometimes. Don't even get as far as the Windows boot screen.
 
It's unresponsive as in slow - mouse pointer stutters all over the place etc. It still 'works' but very slow.
Bringing up task manager the processes aren't being used that much.
I've not done a clean instal (yet) as it does it before it gets to the OS sometimes. Don't even get as far as the Windows boot screen.
so the CPU and memory graphs aren't particularly high?

try a combo of RAM sticks, one at a time, in each slot etc and see if the stutter goes away?
 
Have you tried a System Restore to a time it was working normally ?
 
Can you boot from a linux live CD (or ISO image written a USB stick may be easier) and see what happens?

If it is a Windows problem, and you ran an upgrade to go from 8 to 10 rather than a wipe and install from fresh then you may have brought whatever problem there was with Windows across (e.g. via the registry). Seeing if a different OS is fine will go some way to eliminating the hardware as the source of the problem.
 
Ok - done memtest and burn in test and not a single error so looks like the hardware is probably OK.
Looks like it may be time for a format and fresh install. Might as well stick an SSD in there at the same time I guess.
Thanks for the help so far folks:)
 
Been on the PC all evening running tests, browsing etc and not failed once tonight - how typical is that!
Seems to be really stable tonight - nothing has changed since yesterday so not sure whats going on
 
Ok - still not right. Ordered a SSD last night and going to do a fresh install of Windows 10 and start from scratch. Just suddenly drops performance for no reason at all.
 
SSD fitted and complete fresh install of Windows 10. All seemed OK yesterday but on first boot this evening it stopped mid boot again - guess this has to be hardware now. Checked all capacitors for signs of bulging etc but all look ok physically. Reseated everything yesterday too.
Done another stress test too and all components pass.
Getting lost as to where to look next - or maybe just admit defeat and get a new PC.
 
I feel for you - been down that road many times in the past.
Ok everything you’ve done has been what I would have recommended.
Since you are on a fresh install on a fresh hard drive, that rules out a lot.
Any win 10 hardware compatibility issues? Check hardware vendors site.
Mobo drivers up to date?
bios?
Take a stick of ram out for a few days - or until you get problems and then swap it over with the other. - as neil said.
You’ve done memtest I see. I used Prime95 in the old days for CPU testing. You could give that or its ilk a try.

Good luck
 
Sorry if I have missed it did you swap out the ram sticks even though they passed a stress test?
 
Can I throw a suggestion in please, mine basically went t***s up last week, it would start occasionally , fail to load programs and generally misbehave to such an extent that it very nearly went to the tip and got replaced with a mac!
After 3 days messing about I finally traced the fault to a badly connecting cable ( the connection did not 'snap' into place) between the SATA connection on the MOBO and the SSD , replaced the cable and all is OK
We sometimes miss the obvious
 
trout fisher reminded me I've had a bad sata cable and a bad sata connection on a motherboard once. Try swapping both.
You say you have tried removing some devices to minimise power draw. PSU's tend to lose power over time and I've had odd issues arise. GPU's can draw a lot of power. Do you have an old one to swap with whilst removing most things that draw power again?
 
Last edited:
Start making a log of what you have done and when faults occurred as otherwise you are going to get frustrated. Keep it logical and you will find the problem. The first rule really is to try to boot with MINIMAL components and then add things on. Do you know how to build your machine outside of the case just on a table with a PSU, Motherboard + on board video, 1 stick of ram, cpu (and heatsink fan) and a hard disk? You can just short two pins together to start booting up. Then you can run it for a while until the problem comes. If it does not come, you know it's not one of the things on the table, so you add something on like say a video card. Try it till you see the problem.....keep going like that.
If you get the problem straight from the off, start with ram and ensure you have a known good stick. Then swap out the HDD/SSD/USB stick to a known good one. Then the PSU. If it still causes probs it has to be motherboard or the way the motherboard is configured. Try updating the BIOS. Check for compatible memory for that board too.

EDIT: Yes, also use known good cables :)
 
Thanks folks - a few more ideas to try - will have a play:)
Trouble is it can go days without failing - other times it fails every few hours.
 
May seem a daft question but what would be the effect of a RAM stick breaking down? Would it just reduce its capacity and general performance or would it crash completely?
It's never done BSOD on me if that helps.
 
May seem a daft question but what would be the effect of a RAM stick breaking down? Would it just reduce its capacity and general performance or would it crash completely?
It's never done BSOD on me if that helps.

As mentioned, when I was having issues that turned out to be due to a particular pair of RAM sticks there was no BSOD just random lockups mainly whilst web browsing and only a hard reset solved the issue. A little like your situation it was totally unpredictable as to when it would happen.

I also ran MemTest plus I think PassMark and neither reported any issues RAM or otherwise.

I surmise other folks experience with RAM failures may differ but that was mine. As I had two matched pairs of 4GB per stick I did an elimination test and it was one pair (the very first ones) that were at fault.

HTH :)
 
Removing the graphics card was fun - the onboard doesn't support full HD!

Edit - just had a thought too - occasionally (very occasionally), when using Photoshop it comes up with an error saying insufficient RAM to carry out the operation. I don't remember it doing this before this issue started so I'm guessing this is pointing towards a RAM issue more than anything in light of this.

Hopefully mystery solved:)
 
Last edited:
When things like this happen for me I generally strip stuff out that isn't needed for absolute running.
Firstly take out the extra graphics card and run it on the onboard.
You say you have extra disks? unplug what you don't need for a while, they will be fine.

Also memory as well maybe remove some the Application error you were getting regarding memory was probably related to the graphics card, you mention Full HD maybe also with a very high colour palette? This can cause the graphics driver to complain.
 
So it's not the RAM. Going to borrow my lads graphics card for a few days is the next step - see if it crashes his pc. That will only leave CPU, mobo or power supply.
 
Looking like it could be the graphics card. I borrowed my lads and it seems ok so far. Problem is, he has mine and it's also fine. His card requires the extra power plug to run it but mine doesn't so I'm wondering if it's the mobo not routing enough power to the graphics. His psu is more powerful than mine so it may be his can take the extra power through the mobo.
I think my psu is ok as it seems to run his powered gpu ok (so far).
Hope that makes sense.

Also noticed a couple of nights ago that when I tried printing, it took about 30 minutes to send to the printer - real strange.
Printer is wireless to the router and pc is network cabled to router.

Could this be a mobo data bus problem?
 
Back
Top