Welcome guest. Before posting on our computer help forum, you must register. Click here it's easy and free.

Author Topic: Help narrowing down hardware suspects  (Read 2539 times)

0 Members and 1 Guest are viewing this topic.

Stray Cat

    Topic Starter


    Starter

    Help narrowing down hardware suspects
    « on: May 09, 2010, 02:11:06 AM »
    Hi guys, how's it going?  Not well for me.  Recently built a new HTPC (specs towards the end) and it has been working fine for a week, in I watched a few DVDs and BDs in it, listened to some music, checked out some of the diagnostic stuff it comes with (power usage, temps, etc.) but today, it started acting up.  What it's doing is, in normal mode, freezing after I open certain programs, or freezing a minute or two after it's just running.  I can get it to work perfectly fine in safe mode, and in safe mode with networking, I can do torrents and use firefox, but once I opened two firefox windows and it froze. 
    What I've done so far in trouble shooting this is I removed all recent drivers.  I got a network card today so I downloaded all the programs and such I had on my old computer.  CCCP (codecs), Firefox, utorrent, and some Microsoft updates for DirectX and other programs.  I removed all those, didn't fix things. 
    I started looking into hardware, since I wasn't in the mood to reformat (will if I have to).  What I've done so far on that end is:
    -Monitor system temperatures in the BIOS.  Computer runs at 36 C.  All case fans and CPU heat sink with fan are working great.
    -Checked mother board to see if it had any problems.  It has an onboard LED readout that will give some numbers and the manual has a list of what they mean.  No problem on this front. 
    -Currently running check disk to see if it's the hard drive.  It's about 50% done, been running for a long time now, hopefully it will finish without any problems. 
    EDIT: HDD check done, after like two or three hours.  No problems, so cross that off the list.
    -After that HDD check is done, I was going to pull one of the sticks of RAM, then do the other, to see if one of those is bad.   
    From there, any suggestions on what I should try next?  At this point I still don't know if it's hardware or software related.  Should I pull the CPU, clean it, and reinstall it and the heat sink?  Should I just reformat now and see how things go? 
    Specs:
    Windows 7 Professional 64 bit (legitimate purchase + install, if that matters)
    ASRock P55 Extreme mobo
    Intel i3 CPU
    Radeon HD 5670
    Corsair 650W PSU
    G Skill 2x 2GB RAM
    Let me know if you need any more info.  Appreciate any help I can get. 
    « Last Edit: May 09, 2010, 03:06:41 AM by Stray Cat »

    Calum

    • Moderator


    • Egghead

      Thanked: 238
      • Yes
      • Yes
    • Certifications: List
    • Computer: Specs
    • Experience: Beginner
    • OS: Other
    Re: Help narrowing down hardware suspects
    « Reply #1 on: May 09, 2010, 04:16:34 AM »
    Sounds like either a driver issue, or a bad OS install.
    From what you've said you're already troubleshooting this logically, which is good to see.
    As the HDD has checked out ok, the next thing to do would be to run Memtest on each stick of RAM overnight.
    Check your temperatures just in case - run OCCT as this combines a stress test with temperature monitoring.

    It does sound more driver or OS related than hardware but it never hurts to check everything just in case.
    If the hardware all checks out OK, the next thing would be to reinstall the OS, then load each driver one at a time, checking the PC in between to see if the problem starts happening again.  From what you've said it sounds like a possible network driver issue, but that's more of a guess at this point.
    Did you install the latest Windows updates?

    P.S. welcome to the forums!

    Stray Cat

      Topic Starter


      Starter

      Re: Help narrowing down hardware suspects
      « Reply #2 on: May 09, 2010, 04:32:03 AM »
      Hey Calum.  Thanks for the post and the welcome!  I'm about to sack out, but I discovered something worthwhile so I figured I'd post it before I go to sleep.
      So I was messing around with the RAM.  First, I pulled one out and left the other in.  Turned it on, and it worked perfectly!  I had two firefox windows open, media player playing a song, two diagnostics running, and torrents going and it never froze!  After about thirty minutes of working, I was convinced.  So, I next pulled that RAM and put the other one in, and turned the computer on again.  Same, it worked perfectly for the twenty minutes I left it. 
      So I put both back in, turned it on, and it froze again, same as usual, about five or so minutes in.  I then switched them, turned it on, and it went for about five minutes and froze. 
      I'm very confused by this.  Separately, each work great.  Computer booted up faster and start up programs came on with in seconds running a single stick.  Together, the computer freezes.  Don't know what to make of this.  I figured one would cause the computer to freeze, but neither did.  Only when they are together. 
      Going to have to sleep on this.  I'll try memtest tomorrow to see if it reveals anything about each stick, Calum.

      Calum

      • Moderator


      • Egghead

        Thanked: 238
        • Yes
        • Yes
      • Certifications: List
      • Computer: Specs
      • Experience: Beginner
      • OS: Other
      Re: Help narrowing down hardware suspects
      « Reply #3 on: May 09, 2010, 04:58:33 AM »
      You're welcome, thanks for getting back again so quickly.
      Hm, that's interesting.  Not sure why the issue would not be occurring in safe mode if that's the case, or why the issue would not have shown up at first, that's odd.  Nevertheless, I would be inclined to run Memtest on each stick individually first, then together.  Also, run OCCT with each stick installed, as it will stress both the CPU and the memory.  I would also check the memory timings and voltages in the BIOS, it sounds to me like either the timings are too tight or the voltage is too low, causing this issue only when both sticks are installed.
      I have seen this with a bad memory controller too, which would require replacement of the CPU - that's why I advise running OCCT to stress the relevant components and pinpoint any issues.  At least now it seems you've managed to narrow the issue down to either the memory (settings or hardware), CPU, or the motherboard although that last is much less likely.
      Good luck with it, looking forward to hearing back on this problem tomorrow or whenever you do have time to troubleshoot further.

      Stray Cat

        Topic Starter


        Starter

        Re: Help narrowing down hardware suspects
        « Reply #4 on: May 09, 2010, 04:45:21 PM »
        OK, here's an update. 
        I couldn't get memtest to start up, so I just went with OCCT, which is a pretty cool program!  So using OCCT, I tested each RAM (from here on, calling them RAM 1 and RAM 2) in each slot (bottom slot and top slot).  RAM 1 performed marvelously in both slots, and so did RAM 2.  I did OCCT on RAM 1 in both slots, and RAM 2 in both slots.  Large data stream, for 15 minutes each.  Both worked great, didn't freeze or anything.  CPU temps hit 38 C, MB hit 34 C, so nothing out of the ordinary there.  Along with OCCT, I did som browsing in Firefox, opened up so music files, looked around the computer, etc.  No problems. 
        So, after that, I put both RAM sticks back in, and started up.  It started fine, opened up all programs at start fine, and was working great.  At 6:50 up time, it froze.  Temps were in the normal range (31 C for both CPU and MB), there was never any spikes in memory or CPU (I opened task manager to check this), things were running just as they did with single sticks in, but it still freezes.
        So, I'm taking this to mean both RAM sticks work, both slots on the MB work, but for some reason, these two things don't work well together.  Also, the CPU works fine, as I was maxing it with OCCT on five separate tests (I did RAM 1 in bottom slot twice) and it got hot, but kept on going. 
        Calum, I'd like to try what you mentioned about checking memory timings and voltages.  If you could elaborate more on this or post a link on how to do it, I'd appreciate that.  I've never really messed with BIOS before, but now's as good a time as ever. 

        Stray Cat

          Topic Starter


          Starter

          Re: Help narrowing down hardware suspects
          « Reply #5 on: May 09, 2010, 11:57:37 PM »
          OK, so problem is solved.  Here I was spending most of the day chasing after problems with the RAM when it turns out it had nothing to do with the RAM.  I never would've thought it was the new wireless card I bought, or more specifically the old drivers that came with the CD.  Found a post on some forum talking about how Ralink drivers from pre 2008 would wreck mayhem on computers running Windows 7 64 with dual channel.  Really annoyed right now.  Might try to install the new drivers, then reinstall the card, and see how things work.
          I hate computers :)
          EDIT: http://social.technet.microsoft.com/Forums/en-US/w7itprohardware/thread/9c5fc9c2-e2d1-4b0c-bd37-8619b399f2c9
          There's the link if anyone wants it. 
          « Last Edit: May 10, 2010, 12:11:38 AM by Stray Cat »

          Calum

          • Moderator


          • Egghead

            Thanked: 238
            • Yes
            • Yes
          • Certifications: List
          • Computer: Specs
          • Experience: Beginner
          • OS: Other
          Re: Help narrowing down hardware suspects
          « Reply #6 on: May 10, 2010, 12:16:30 PM »
          That's very odd, to say the least ... I've never heard of anything like that.
          Thanks for posting the solution, you learn something new every day (or so I'm told).

          To answer a query you had, although it's not relevant to your current issue it may be useful in future, memory timings and voltages are all defined by your memory manufacturer.  Each kit will have its own rated specs, and sometimes the motherboard just won't set them correctly.  I'm not familiar with Asrock BIOS layouts, but memory settings are usually in an overclocking menu.  The correct specifications will be in a format such as "1600MHz 9-9-9-24-2T 1.65V".  This would mean the memory's speed would be 1600MHz, the first four memory timings (they are always listed first, but for reference they are CAS, tRCD, tRP and tRAS) are 9, 9, 9 and 28, the command rate is 2, and the voltage is 1.65V.  They're just the recommended specifications, the RAM may work above (better than) those specs and will work at lower specs, but those are the guaranteed specs.  It's worth setting those speeds and timings manually to avoid any possible issues.