Spread the love! TechSpot Tech Gift Shortlist 2017

Computer sporadically stops responding?

By whenallelsefa1l · 82 replies
Aug 13, 2010
Post New Reply
  1. Sharam

    Sharam TS Rookie Posts: 509

    Hmmm, good find!

    I have seen this on EVGA 780i with the SB HS connected to the others with a flat wide slab of metal but the 680i uses a pipe ... since finding this, I always check the temp and if it seems high, I re-do the whole thing (not an easy DIY)

    Well, let us know how things go after you get them.

    Remember, build the motherboard outside of the case, as in, install the CPU HSF and RAM before putting the mobo in the case, you can even attempt POST with the board out first to ensure everything is working.
  2. whenallelsefa1l

    whenallelsefa1l TS Rookie Topic Starter Posts: 60

    i actually decided to go with the i7-860 655k @ 2.8ghz. It matches the board and came in a bundle on newegg. I'm still getting the same RAM and that same board came in the discounted bundle. I'm excited to get the new components. My motherboard is slowly but surely getting worse, i'm on a friends computer now. i've opted not to use my computer until the new parts come so that i dont damage anything else. I'll probably sell the RAM and cpu on amazon or another website since they seem okay with every test i ran by them.

    I'll make sure to test everything outside of the case that way i can make sure there's no DOA's. I've never gotten a DOA before from tiger direct or newegg though (knock on wood) so i should be okay. I think i may eventually buy an nvidia 470 gpu of some sort to utilize DX11. That will be the day i beat all my friends computers! :p
  3. whenallelsefa1l

    whenallelsefa1l TS Rookie Topic Starter Posts: 60

    Okay so i got all of my new things today.

    Turns out, either the motherboard or the RAM is dead.

    Any way i can figure out which is the dead one? everything lights up except my Tower's LED's (both power and hard drive), and the cables are connected correctly.

    Some little twists that pissed me off were that my CD and DVD drive are incompatible with the EVGA p55v 120-LF-E651-TR motherboard, as is my PCI wireless N internet card. The board only has 2 PCIx8 and 2 PCIx1 slots.

    so thats some more stuff i need to spend money on

    but the true problem, WHAT is wrong here? is the Mobo dead or the RAM? i get no post whatsoever. my CPU cooler lights up along with my GPU. No LED lights on the tower light up, and no response on my screen.
  4. Sharam

    Sharam TS Rookie Posts: 509

    Let's have all the new components one more time,

    EVGA P55V SLI/ EVGA p55v 120-LF-E651-TR (need to check on this, is it a tri channel board?)
    Core i7-860
    RAM "exact model"

    I thought you were going to try the board alone first?
    Disconnect everything from the motherboard, leave the CPU, HSF, RAM and Video in.
    Remove and re-insert the RAM and Video card 2 or 3 times.

    Try to boot with ONLY the CPU, HSF, RAM, Video, PSU and keyboard ... NOTHING ELSE connected to the motherboard ... to start, short the PWR pins with a flat screw driver or something for a second.

    TIll I get the exact model of your RAM and more info on the mobo, use single module in the first Socket with the above test.
  5. whenallelsefa1l

    whenallelsefa1l TS Rookie Topic Starter Posts: 60

    well i bought a multimeter today. i'm going to test the old psu and the gpu.

    the RAM is OCZ Gold 4GB 240-Pin DDR3 SDRAM DDR3 1333 (PC3 10666) Desktop Memory Model OCZ3G1333LV4G

    It came in a bundle with the mobo, so it should be matching even though it is not in the official QVL. i'm only using one module to begin with.
    I have also removed and replaced the RAM and GPU like you have said multiple times last night to no avail. If by tri-channel you mean tri-sli, then no, if tri-channel means RAM, i have no idea, but there are only 4 DIMM slots if that answers the question.

    another thing i forgot to add, i had to use my OLD Ultra LSP 750w psu because it has a 4pin connector needed to power the board that the new C.orsair psu did not. this COULD be an issue but since changing PSU's on my last board made no difference i'm going with it's okay. I'm still going to double check and use the multimeter to check just in case.

    what "PWR pins" is that i should "short" and how would i go about doing this? while the mobo is on and running or while it's off? i dunno if you mean the reset switch on the mobo which i have no idea where is at, or the actual PWR switch pins that are used to turn the computer on.
  6. whenallelsefa1l

    whenallelsefa1l TS Rookie Topic Starter Posts: 60

    update on trichannel:

    Tri-channel memory is not supported by the p55 series.
  7. Sharam

    Sharam TS Rookie Posts: 509

    Your PSU should have the 12V 4 pin as well as the 12V 8pin, I find this very odd with your high quality PSU!!!

    OK, imagine the board out of the case on the box that it came with (with the anti static bag in between)
    You install the CPU, HSF, one RAM module in the first socket, Video card (haven't checked the mobo yet but I don't think it has on-board VGA, if it does, we try that first) Power supply 24pin and the 4/ 8pin and the keyboard.

    The case front panel are not connected, nothing else is connected to the board either, so, in order to power up the board, using the mobo's manual. You find the pins for the front panel power switch and short them momentarily.

    If you get POST, you enter the BIOS, load the defaults, power down and start adding the rest of the RAM, try to POST, power OFF, add another component, try to POST, power OFF and so on...


    Ahh, your RAM is a single 4GB module with 1.65V rather than the default 1.5V DDR3 spec. If the board is supplying 1.5V, you will need to try different sockets or use a lower voltage DDR3 to get in the BIOS and set the RAM to 1.65V then put the OCZ in.

    Hope this is the case as seen with many boards and high voltage high performance RAM, most boards drfault AUTO setting works with any RAM but again, seen this many times.
  8. whenallelsefa1l

    whenallelsefa1l TS Rookie Topic Starter Posts: 60

    Just to let you know, i am a complete *****.

    The PSU i'm using (the old Ultra LSP) was switched to european as opposed to american. thats why i couldnt boot.

    I have yet to test anything with a multimeter, but i installed windows successfully and got games running and stuff.

    I am still however getting lockup issues.

    Could the culprit be the GPU? seeing as a new PSU did not solve the issue, nor did a new motherboard, CPU, and RAM. The only two remaining entities that i could see causing this are the GPU or the SATA HDD, however, i dont quite understand how an HDD would have anything to do with the 12v pull.
  9. Sharam

    Sharam TS Rookie Posts: 509

    I've seen lock ups with HDD controller but everything lockes up, even the mouse!

    Explain the lock up again please and can you do a CPUZ reading?

    Drives use the +12V rail to spin, hard drives up to 12W and optical drives up to 24W with most brands.
  10. whenallelsefa1l

    whenallelsefa1l TS Rookie Topic Starter Posts: 60

    i havent got time today to do a CPUZ reading, but will most certainly do that by sunday.

    As for the lockups:
    First, i'll be using a program such as Mozilla Firefox, Steam, Xfire, Internet explorer, iTunes, and all of a sudden it will stop responding.
    Once the program stops responding i either hit Ctrl Alt Delete, or i wait. If I wait, explorer will lock up (my desktop/taskbar) and i get the spinning blue circle for the mouse when something is loading in Win7. When i hit Ctrl Alt Delete, i have to hit it multiple times to register, and when it does, sometimes the screen goes black for 10 seconds and then the menu pops up, and when i hit "Start Task Manager" it won't come up unless i do this process a few times. Once it does come up, 3/5 times, it will stop responding as well, and then everything just starts working all at once again.

    It's quite odd. normally my HDD LED will be lit up the whole time through this, with no flashes.
    With my old motherboard however, it was quite the opposite. it STOPPED being lit up and only flashed intermittently, i'd say every 6-12 seconds.

    It seems to be totally fine during games right now though. it's really stable, and the only time i have problems is if my ping is way too high.
  11. Sharam

    Sharam TS Rookie Posts: 509

    OK ... when you installed Windows, did you delete the partition and started fresh?
    Did you at any time before all Windows updates, installing an Anti Virus with updated definitions added a backed up drive, CD/ DVD, thumb drive and such? Trying to rule out possibilities of MBR or system infection.

    On the systems I have worked with, when system freezes with the hard drive lit solid, nothing works, this goes on for 30 to 60 seconds at times and it was related to one of the SATA controllers, the others, no problem. Come to think of it, I never checked the data and power cables .... hmmm ....

    Like you said, if installation, including drivers and all were done right, this would only point to the two components and their drivers perhaps!

    How is it that you cannot use the new PSU?
  12. whenallelsefa1l

    whenallelsefa1l TS Rookie Topic Starter Posts: 60

    the new PSU did not have a 4 pin. it only had an 8 pin. well. 2 8 pins. plus one of the SATA power connectors was broken from getgo so i just returnehd it. the old one had the one 4 pin i needed for the new motherboard.

    as for updates and stuff, I did completely reformat, took 3 hours, but it reformatted my whole drive. i installed all my device drivers while the system was downloading the 37 updates for Win7, and i did install AVG Free antivirus and updated the definitions in between the driver discs. On the other hand i HAVE been using the same SATA cable since i got the hard drive. when i get home later this week i'll switch out the cable and see what happens, i have a million of the damned things laying all over. The lockups were happening even while installing certain updates though so i would imagine its either the SATA cable, or one of the two components. I did not go anywhere but google and AVG's website during the installation of the antivirus though, so the system should not be infected. It points towards a component as well since its the same problem i've had since the last motherboard CPU and RAM.

    One other question. How in the heck do i test a GPU with a multimeter? lol, if its even possible that is.
  13. Archean

    Archean TechSpot Paladin Posts: 5,690   +96

    You can't, simple solution can be if possible, borrow another graphic card from a friend, if not, perhaps another alternative can be to buy some dirt cheap card, and see what happens.
  14. Sharam

    Sharam TS Rookie Posts: 509

    For GPUs with PCI-E 6 or 8pin, you can take a reading from the plug first then the back using the solder points if visible. I don't think that would prove much.

    You didn't say if you deleted the partition before format. Not that I think it is related, but it is best to be certain about everything, we are down to HDD and GPU and don't need to think what if this and that, in case of MBR infection, even after reformat and install of new AV, system could still be compromised!
    Latest drivers were downloaded rather than using the CD/ DVD?

    These days you do not need to visit certain sites, download files and execute them to get infected, all is needed is to be connected to Internet without even opening a browser. I don't believe we are looking at infections but like to be always 100% sure about things!

    Do test your PSU while the system is running, before Windows loads, you can enter the BIOS so it will sit there, take a reading from all 3 rails using the 24pin then let Windows load, play a DVD and do an AV scan at the same time then take another reading.

    Download BurnInTest™ run a few tests back to back, the trial version lets you do 15 min tests, in preferences, take OFF floppy, CD and such, do the first test with 50% load then increase the load, do few tests back to back.
  15. Archean

    Archean TechSpot Paladin Posts: 5,690   +96

    Or another alternative can be OCCT, its free to use, with which you can thoroughly stress test your graphic card.

    That is a good point Sharam, PSU can be an issue, considering that he has almost changed everything except the few parts.

    Lastly, if you have deleted the old partition, then recreated and reformatted chances are you are clean provided the medium you used for installation is clean, but just to be sure please go through Virus and Malware Removal forums 8 steps as a precaution, and wait for their review.
  16. whenallelsefa1l

    whenallelsefa1l TS Rookie Topic Starter Posts: 60

    I do think it is my harddrive. me and my buddy switched out the cards and it did not change anything. we even hooked his PSU up. I switched out the SATA cable and saw no improvement, and the HDD still takes forever to do things apparently. I'm going to buy a new HDD next paycheck and i'm going to stay far FAR away from Western Digital.
  17. Archean

    Archean TechSpot Paladin Posts: 5,690   +96

    Well you can look at an article about HDD reliability on the link provided here. You'll be surprised about the results.
  18. Sharam

    Sharam TS Rookie Posts: 509

    Don't blame it all on your hard drive, something must have pushed it over the limit, new drive with your old config would've had the same faith most probably due to many reasons hidden in this thread!

    Hope the new drive ends all this and you can get back to uninterrupted gaming.

    Happy fragging ;)
  19. whenallelsefa1l

    whenallelsefa1l TS Rookie Topic Starter Posts: 60

    well i sure as hell hope its not caused by my PSU. I looked at so many videos and i STILL cannot figure out how to test the darn thing with a multimeter. I wish i had someone who friggin knew how to.
  20. Sharam

    Sharam TS Rookie Posts: 509

    Testing the +12V and the +5.0V is easy using one of the large 4pin Molex connector, use the Black as ground, Yellow for the +12V, Red for the +5V. SATA power connector should have the Orange +3.3V as well if it comes straight out of the PSU but the holes are a lot smaller. I use the 24 pin connector.

    With all the above, we are measuring voltage before it goes through the board, not the best way to test but it would tell you enough. The meter should be set to the lowest DC Volt, mine jumps from 2V DC to 20V DC so I have to set it to 20V DC to test all, if I had 5V DC setting, I would have had to use 5V to test the +5V/ +3.3V and set to 20V DC to test the +12V.

    I was thinking more with your board and CPU working together, the South Bridge and the HS. Still after all these, it is possible that the drive just went south all on it's own. Hard to say from this side of the screen.
  21. whenallelsefa1l

    whenallelsefa1l TS Rookie Topic Starter Posts: 60

    okay that makes sense then cause my multimeter goes from 2 to 20 as well.

    And to me it seems as though the hard drive went on its own only because Western Digital hard drives tend to die easy. And i do move my computer around a lot. My friends live an hour away and i'll put my computer in my car, drive up there, and we'll LAN together for a night or two. might explain a lot actually..
  22. whenallelsefa1l

    whenallelsefa1l TS Rookie Topic Starter Posts: 60

    quick question actually.

    Does having a bad harddrive relate to drops in FPS while in a game? i keep getting like 17-30 fps in game. I'm realllllly hoping its my HDD thats bad.. ugh
  23. Sharam

    Sharam TS Rookie Posts: 509

    About the HDD being bad, don't forget the board and the CPU compatibility issue as well as the reported +12V voltage (your old setup)

    I don't know what game you play and on what server. Many things affect FPS starting with the most important part, the server then the number of players before going to your system.
    Any newer system can handle MP server if set correctly.

    I don't know the location of the server and your ping, if this happens on different servers, do they use custom maps or the standard maps, and does it happen right away or after playing for some time.

    All in all the player's system setup and performance plays a big role as well, I used to run Soldier of Fortune (the original) server 24/ 7 and few other games, one of my servers was not dedicated and I used to play on the same system, this was a PIII Celeron 800MHz over clocked to 1066MHz with 192MB of RAM running on 1.5Mb business ADSL line, 10 slots and we played hours on with either Assassin or CTF mode with no lags or crashes.
    Playing on other servers was different, even with high ping I had decent FPS, and I preferred 30 to 60FPS (as long as you know what you doing with the vid settings). If I had the VSYNC ON, 30 and 60 was perfect.

    I do build and setup gaming systems with attention to the MP gaming performance but over the internet is rather difficult to troubleshoot low FPS; best place would be one of the Forums, perhaps from one of the respected clans!
  24. whenallelsefa1l

    whenallelsefa1l TS Rookie Topic Starter Posts: 60

    well i replaced my HDD and it fixed the lockup problem. This makes me extremely happy.

    However, while playing Call of Duty Modern Warfare 2, i get around 46-56 fps while standing still in a non combat area, then once i enter an area with everyone shooting and a lot of stuff going on i drop below 30 fps. sometimes i'll even hit 19fps.

    My first thought was that my GPU was overheating, but it isnt since it can handle up to 105*C, and runs hot at about 75*C. I'm wondering if my PSU really DID cause any issues. I'm going to have to test it with my Multimeter. If there are any websites that you know of that give good instructions on how to do so it would be greatly appreciated. This could possibly be that my PSU is not giving my GPU enough juice. and thus runs with bad frames.

    I know it's not the servers, seeing as the server only has 12 people total including myself, and it is not run on Dedicated Servers. It's a crappy matchmaking system, but even before it was just lag, not low FPS. I was getting around 60 fps at all times before i got computer issues.
  25. Sharam

    Sharam TS Rookie Posts: 509

    I don't know of any site with instructions but can check.

    Dedicated server is the one that is not generated from within the game, as in, you start a game and in one of the menus you start a server and even play on the same system at the same time. Dedicated is any computer running MP game server and nothing but, this is done by adding switches to the game's executable with certain CFG file that is changed by the server "owner" and it does not run the game itself but a server window which resembles the Command Prompt with most games. The server's upload is of most importance as well as it's hard drive setup, condition and speed.

    Have you changed your settings in the game as well as in your video card's utility?
    Is detailed texturing ON? How about shadows, AA and AF.
    What is your Monitor's refresh rate?

    Without knowing the server's exact setup, it is hard to say if your GPU can’t handle the amount of information or if the server cannot provide the client with enough info based on the ping, upload and it's overall setup. (combat area hits the server hard, as in, sending lot of info to all you 12 players)

    1.2GHz processing power, 387MB of RAM and 360Kb upload should be the minimum requirement of a 12 slot server assuming the server runs at it's best. Keep in mind these figures are for the server only and does not include the system itself "idling" however, most todays computers and connections pass them figures!

    As long as the game is not crashing or you are not getting artifacts, we can assume your card is doing its job, I never setup anything that runs that hot even if it can handle twice the reading you are giving. Most my older setups use third party coolers.

Similar Topics

Add your comment to this article

You need to be a member to leave a comment. Join thousands of tech enthusiasts and participate.
TechSpot Account You may also...