PC suddenly very slow overnight


My PC has been running great recently, but yesterday I had some heavy slow down shile playing TF2. It was late, and I thought it might just need a reboot so I turned it off. This morning when I turned it on it seemed fine, until I played TF2 again. I rebooted once more, and this time it took about 10 mins just booting up. When it did, it was running exremely slow. I tried again in safe mode and the same thing happened.

I disconnected the second monitor as well as disbaing SpeedStep in the bios. It was noticably quicker but still slow. I went to safe mode and restored system to a few days ago, now it gets to the login screen quicker but once I login it's again extremely slow to load up.

I'm at a loss to what this could be. I checked the CPU fan and it seems fine, not dust covered either. I only recently got a replacement GPU (7900GT) so I doubt it could be that. Does it sound like I need a new CPU, or could it be something else?

As it is too slow to load any program up I haven't done a spyware check yet, but there doesn't seem to be any new processes running. Could spyware or a virus have this effect?

Thanks for your help
36 answers Last reply
More about suddenly slow overnight
  1. Quick update, just checked the BIOS and CPU is running at 83C (it's an intel core duo 1.8GHz), is that too high?
  2. 83c is way too high. Your CPU is throttling to a lower clock speed. Are you using the stock heatsink?

    If you are check the heatsink and make sure its on right. Seems like you are more than likely having a bad heatsink seating.
  3. :o
    That's hot!
    I think your cooler if FUBAR.
    If the temps do not come down by reseating it, DO NOT use your system until it is replaced!!
  4. Dang... looking at the CPU's that are 1.8ghz, the thermal spec is 60-61C.

    So it was throttling to save itself for sure.
  5. Wow. Double-dang. Check that HSF. Make sure the fan is still running, and that it is properly seated. You may need to remove it, then clean off the old thermal paste and apply a fresh layer before reseating your HSF. Use 91% isopropyll alcohol for the job.
  6. Thanks for all the responses.

    I reseated and its cooled right down to 55C, so that's alright now. It's still very slow on startup though, although I can now start it up so a definite improvement. Do you think it is still throttling? Or is there something else wrong?
  7. @ 1.8ghz... it shouldn't even be idling at 55C, that is if its at idle.

    If it's the stock 4 push pin type... take the time to find a 3rd party HSF, that is non-4 push pin, and has a back plate. Although, you may not like taking your system apart, it is the best bet I think.
  8. What I don't understand is it has run fine with exact same CPU and heatsink for over 2 years. I can understand that before it was reseated it was throttling, but now it is close to 30C lower why is it still? Is there any way the throttling hasn't turned itself off whenit should have? (I know nothing about this sort of thing)

    I have no probem taking apart the PC, but I'm more worried that something else is the problem as the temperature has dropped so much.
  9. Just don't get Thermaltake Big Typhoon!

    They have to be reseated frequently.
  10. Atm, I'm on my E4400 rig. It's OC to 2.8ghz (speed step is off), and its stock is 2ghz, only 200mhz faster then your 1.8ghz CPU.

    The idle temp right now on my chip, is 30C, which is the Tcase (what the bios temp would read) sensor, and my ambient temp is 28C.

    One difference is, I'm using a tuniq HS on it, and it uses a back plate. The 4 push pins are kinda crappy if you ask me, and will break down in time. I still think your idle temp at 55C indicates that 1 of the pins are not secure.

    I guess another thing to think about is why after one night, your CPU temp went to 83C in the first place. I can only conclude that the 4 push pin failed to keep the HS in place.


    After all, the 4 push pins are made out of plastic, which will get brittle in the heat/cold after so long.
  11. Chalk this one up to good old "socket creep". Happens over time in computers when something like a stick of RAM or CPU or Heatsink that shouldnt move, does due to contraction and expansion (heat and cool) and works it's way loose from the motherboard.

    But it's always fun to blame it on a Virus... gives ya an excuse to wipe the system and start fresh... which should be done every so often anyways.

  12. Thanks for the help. I've ordered a new heatsink to be delivered tommorrow so hopefully that will fix the problem, I'll let you know.
  13. Well considering that the CPU allowed itself to run that hot a worst case scenario may be a bad CPU.

    But try the new HSF first and lets see if that helps.
  14. Use programs -> accessories -> system tools -> system restore

    I will restore back to your previous setting and installed programs. Believe me it helps.
  15. Also, in the interest of safety. Go in your bios and set your shutdown temp @70C. It's good to have something in place should this happen again.
    I had a sink come off once while I was at work and when I got home the system was shut down. Checked inside and sure enough, fixed it and never had an issue with cpu or performance.
    Hope this helps!
  16. Your temps still seem a little high (unless your ambient temps are high), but you don't have to guess at whether or not the CPU is throttling. Download a utility such as the "Intel Thermal Analysis Tool" and you can watch your cpu speed and temperature. You may want to boot up, get that running, and then turn your computer case onto it's side so gravity is firmly pressing the heatsink down onto the core, at least if you notice that temps are still high and throttling is occuring.
    After you've eliminated temperatures on your CPU and you are sure that your computer isn't damaging itself, defragment your harddrive. That's just good policy, not a solution for your problem. Next, you need to determine whether or not you have a software problem or a virus problem. Either way, a fresh install, if done properly (remember boot sector viruses), will likely solve your problem.

    If you try to address the possibility of a virus, you can always try the many online free scans, or download a free scanner (AVG has one, for instance), but I never trust those to be effective if they are installed after a virus. The best way is the most difficult - use a second HDD with windows or linux installed and scan from there. If you have a USB flash drive and another computer to work from, you can make a bootable usb windows xp hard drive, but it's not straight forward and it takes some time.
  17. Thanks for all the replies.

    I'm going to try the new heatsink tommorrow, and if that then doesn't make a difference I may have to consider a new CPU. Does anyone have any CPU recommendations for around £120?

    One thing I have just noticed is that in the Nvidia monitor tool it says my disk is running at 100% when trying to load things, is this normal?
  18. 100% disk means virus, failing hdd, corrupted files. Run Hdd checking programs, You can get them free from Your hdd manufacturers site.
  19. Hey what mobo do you have?
  20. It is a...

    Asus P5B (Socket 775) PCI-Express DDR2 Motherboard (MB-151-AS)
  21. So it would have the P965 chipset right? Just guessing as I am at work so I can't get to Asus website.

    If so and you decided to get a new CPU if the temps don't drop you can get a Pentium DC @ 2.2GHz for around that price.
  22. Yeah that's right, it is P965. Had a quick look at that CPU and its only $114 (so around £60) so I could probaby afford a bit more than that.
  23. http://support.asus.com/cpusupport/cpusupport.aspx?model=P5B&SLanguage=en-us

    This is your mobos CPU support listing. You can support any of the new 45nm chips that run cooler and faster.

    A good one in your price range would be the Core 2 E7200


    Its a 45nm part, runs at 2.53GHz and is much faster than what you currently have.

    Now don't forget to update your BIOS as well to the latest one. For your mobo you should be able to load the newest BIOS on a thumb drive, load up the BIOS and start it from there (refer to manual).
  24. I just realized you stated $120 pounds not dollars so here is the best choice in that range:


    The Q6600. Its a 65nm part but its awesome (I have mine OCed to 3GHz) and will make your TF2 experience much more pleasant.

    Trust me on the TF2 thing. I have this, 4GB of memory and a HD2900Pro 1GB and can get 150FPS on a good map and server.
  25. Thanks for your help, I'll definitely consider the Q6600 if the new heat sink doesn't help.

    In fact, I'm quite tempted to get it anyway as mine is quite dated now!
  26. It is a temptress huh..... It sits there taunting you with its promises of fun times and happiness.

    But yea. Its a great chip. Easy OC to 3GHz without changing the voltage and very fast too.
  27. Update: I've put the new cooler in (an arctic cooler) and it's brought it right down to low 30C, and about 38C when both cores are running at 100%. So an incredible difference, but the PC is still not running any faster.

    I'm doing the manufacturers HDD tests now but they've brought up nothing so far, and I'm also running a virus check. If that brings up nothing I'll try another system restore (have tried one already) and then it seems there is no option than to buy a new processor.

    I'd love to hear if anyone has any other suggestions, even if they are unlikely to fix it.
  28. I have also noticed that the CPU usage in the task manager don't add up with the individual usage for processes. For example in the list of processes I am only using around 5%, but it is telling me 50+% is being used. What could this mean?
  29. Might help you did did a screen cap of the taskmanager.

    Sounds like something is using CPU resources. At idle, are both cores down to 0 - 3%? That is how mine is acting.
  30. At idle it's at 0-1%, and then every few seconds it shoots up to 10+% for some reason, but none of the processes show any usage which is strange. If any program is running the usage is around 55-70%, which it never used to be. I've used the intel tool and it says both cores are running at the right speed which is again strange. I'm at a complete loss to what it could be apart from the processor being damaged due to heat.

    The HDD tests didn't bring up anything, and neither did the virus scan.

    Before I order a processor, I always thought that either a processor worked or it didn't. Can they really only "sort of" work? Also, could the hard drive theoretically cause the PC to run this slow even though the checks came up with nothing?

    Thanks for all the help everyone, it really is appreciated.
  31. Let see.. I do recall my dad's system, E4300 (1.8ghz) worked fine for 2 weeks after the build. After that time went by, my dad told me it takes along time for the PC to get into windows. The HD light was always on, indicating something is going on when it trying to read it. I ran Super PI and it was getting the same time running 1MB. So it wasn't the CPU.

    So I did some trouble shooting. At first I thought it was the HD. We order another HD, and a USB exteral hook up for internal HD. I was going to try to make an image, and put it on the new HD.

    At that point when I was examining the old HD, I noticed it was fine. Transfer rates were fine. Then it dawned on me that I didn't check the IDE cable. So I changed the IDE cable, and hooked it up. Same thing. Then I hooked it up to the secondary IDE. Wham, it took off like a bat in hell.

    The conclusion was the primary IDE slot had some kind of cold solder, or possible just something wrong with the channel. It runs fine off the secondary to this day. He still has sata ports to fill up, so I thought I'll just run it that way.

    So from that, my dad's system did run slow because it's primary IDE slot got messed up somehow.

    You prolly should do some test to be sure it's actually the CPU. If you ever tested it on Super PI, that should give some indication that theres a problem with it, if it takes more time then it normally did.

    If you never did that test, you prolly could search to see what other people got at the stock speed.
  32. I've never used Super PI before, but I just did and for the 1M it took 35s. This is about 5 seconds slower than the stock benchmark I found online, is that a significant difference? It doesn't seem like much to me, not enough to cause the extreme slow down anyway.

    I might try switching to the secondary IDE, I'm willing to try anything at the moment!
  33. That seems okay. What CPU do you have? perhaps your memory timings could cause it to run slower.

    I don't see or recall what you have. There are different variants:

    E6300 - Super PI 1MB - 30 sec
    E4300 - Super PI 1MB - 30 sec
    E2160 - Super PI 1MB - 34 sec
  34. Well, I don't quite believe it but switching the HDD to the second SATA slot has cured the problem completely! Thanks so much!

    For the record its an E6300, and I just reran Super Pi and it did 1M in 34 seconds, but I'm just glad it's working again!

    Thanks again to everyone that helped
  35. Wow... you had the same problem my dad's board had, uh? Only difference is his is IDE, he doesn't have any sata drives.

    Well.. glad that got you back up and running. :D

    And as far as the Super PI, it's prolly memory settings. And besides that, that E6300 kicks my old P4 3.0ghz.. if I remember right, it took it 58 secs. :lol:
  36. ^I should test SuperPi on my old system. It has a 3.4GHz Pentium 4 EE the ones with the 2M L3 cache so that extra cache should hellp it out.

    As for the OP, I am glad yours is up and working again. You are lucky though because your mobo is Asus and that one specifically was a good choice as its able to support all the newest CPUs so if you ever decide to upgrade its as easy as flashing the BIOS and dropping the new CPU in.
Ask a new question

Read More