Okay, if anyone gets this one, you may officially be forever known as the man (or woman). System Specs: P4 3.06ghz MSI 478-socket mobo 1GB DDR PC3200 (Kingston) 80GB Seagate nVidia GeForce 6600GT (71.89) Creative Audigy Z2 Antec TruePower 430W Sony DVD-R / CD-RW Combo Sony DVD-ROM Okay, so the problem is that my computer likes to freeze up constantly. The display freezes (if a sound was playing, it loops random digital glitches). When I try to reboot the computer, it boots up fine but the monitor never goes from standby to active. I know what you're thinking: heat issues. The first thing I did was disable the hardware acceleration. Still locked up. I have yet to enable it and it's still locking up. I also went out and purchased a new video card, thinking maybe my piece of crap FX5600XT had finally bit the dust. Same problem occurred even with the 6600GT installed. I tried running the computer with no video card at all, still the same problem. I removed the drivers and reinstalled several versions, but no dice. So maybe the CPU is overheating? I thought it might be, anyway. I had always flirted with the idea of overclocking, so this was a good opportunity to go out and buy a CoolerMaster liquid cooling system, which I did. It's installed and my CPU temperatures have ranged between 28C and 45C at the time of the lockup. In some cases well below the average operating temperature for a stock processor. It may help to note that I even have both panels off the case and oscillating fans blowing on either side. Still locks up. I thought I might have some bad RAM, but I can't get the computer to go through the entire test without locking up. However, it has gotten as far as 45% and showed no errors to that point. It has locked up at different points in the test every time, which leads me to believe it probably isn't the RAM. I don't think the monitor could freeze the entire computer, so that shouldn't be a consideration. I got the HD through a scandisk and it seems fine (I know scandisk isn't world-renknowned for accuracy). After several attempts, a virus scan came up negative and I ran both Ad-AwareSE 1.06 and Microsoft Anti-Spyware Beta to clean up the system. I tried killing every process I could that wasn't vital to the system...it still froze. When the freeze actually happens, I get absolutely nothing from the activity display LED. It stays dark without so much as a blink. When I boot the computer and the display doesn't come onscreen, the LED practically stays solid. Also odd is the fact that I can go to my computer 15 minutes after one of these freezes and it will boot up fine. It seems the longer I leave the computer powered down, the longer I have before it freezes up again. Another note, I have reset the defaults on CMOS and I have also flashed the BIOS (through the computer and also manually by removing the battery). It almost seems like my motherboard is overheating, but I don't even know if that is something that is significant beyond affecting the temperature of the CPU. Anyway, good luck and thanks for the help.
Does it only happen when you boot into windows? If not, try running memtest or something in dos mode. Try running the 2 sticks independently of each other (if you have 2.) Though I don't see why your video would screw up because of a memory problem, since video cards have their own memory The best I can come up with is there's something with your motherboard like you said, since when it craps out, EVERYTHING craps out and not just a certain thing... and we all know that the motherboard controls everything. edit: might as well try a spare hard drive that you have lying around (hopefully) since you say that the activity LED goes crazy. If you don't have access to one get a cheap 5gb off ebay for $15 and see if it works Well, once you solve it, you'll say at least I got an excuse to buy some new stuff.
Solution probably found in hardware or BIOS setup. Check Wim's BIOS http://www.wimsbios.com/ Possibly you have USB, NIC, and ? sharing resources with your Audigy or something like that going on. Try disabling Com1 and Com2 in BIOS, freeing up IRQ4 and IRQ3 can help your PC to POST (before it even gets to the OS). If you have a USB or network printer, disable your Lpt1 freeing up IRQ7. These type of resource conflicts or BIOS misconfiguration can do just like you say - weird unpredictable lockups ;^( Windows might reassign resources, but IRQ and other conflicts in BIOS should be addressed before Windows is even installed. Well, hopefully you would first flash the BIOS, then load the defaults for maximum reliability, and then if everything stable load the maximum performance defaults, then review each setting individually with the help of Wim (and your mobo manual). Good luck, Regards
Thanks for the replies.. In regards to whether this is isolated to Windows: No, this happens at any point from the time I power on the computer. I tried running memtest86, never could get it to complete a test before locking up. An update on the problem: when I restart the computer after a lockup, it IS going through a full boot into Windows and running fine. I discovered this after leaving it on post-reboot (while the display did not boot up) and then powering down manually. I left it alone and came back (to get the video booted up) and when I booted it up it stated that Windows had not shut down properly. Based on this, I'm almost positive that it is not a problem with the HDD. OGS, thanks for the info and the website. I'll print that up before my PC tanks again (if typing this post didn't take too long) and see if I can't get this set up. On a weird side note, I have a Saitek Gamer's Keyboard and the damn thing won't recognize prior to Windows booting up. My PC recognizes any other USB/PS2 hookup fine, but not this keyboard.. I have to boot up with an old spare in order to tweak the BIOS.
USB keyboard support at POST must be specifically enabled in BIOS. Default is to have OS administrate USB keyboard, so no keyboard at POST... Enable BIOS support for boot from USB keyboard, should fix problem... It is good to realize this is not a Windows problem. Trust me, when you get this sorted your Windows should FLY... well, at least be rock-solid :^) L8R
Noctis, can you get your hands on a PSU tester?? They are relatively cheap (about $15.00) and worth their price in gold!! I would love to see if it's the Power Supply Unit causing the issue. Let us know. Thanks!!
if you have tested everything else it does sound like the power supply to me. the only other thing i can think of is either a damaged cpu or a short on the motherboard that happens only when a piece of metal expands with heat and touches where it shouldnt. you say if you leave the pc off longer then it takes longer to freeze up then this definately sounds like something to do with heat somewhere in the system. i am on a situation like that i would consider taking everything out of the case, dissasemble the whole pc, dust it all off with an air duster, reapply some new thermal paste to the cpu and put everything back togeather. if this still dosent solve anything i would think about finding a friend with components you can borrow and test such as cpu, ram and psu. i know from experience though there is nothing worse than buying a new component to find out the old one worked fine
Yea maybe try a new pw supply... but one thing to check is the memory timing in the bios, try setting it to the slowest timing
Okay, another update... As far as the IRQ stuff.. I've never messed with that before. I found myself a bit surprised, but I have no clue how to manipulate those settings. I think my BIOS might be a bit scant on features, unfortunately. I have an MSI mainboard. The only IRQ settings I found that I could set were assigning IRQ numbers to IRQ slots (i.e., IRQ 1/5 is set to auto, IRQ 2 is set to auto, etc, through the 4th IRQ slot). The PSU is maybe two months old. It's an Antec TruePower 430W. I've checked the voltages on it several times and everything seems to be perfectly normal and well within the acceptable operating ranges. Other than that, I flashed the BIOS again and loaded the defaults again. The only setting I'm not sure about is "Default Graphics Adapter" which is set to PCI, but I'm definitely running an AGP card in an AGP slot. I'm almost out of ideas here. I don't have onboard graphics, so I'm trying to get my hands on a PCI graphics card to test that theory. Failing that, I'm going to take my computer in to be thoroughly cleaned by a shop (I don't have time) and I may just have them go ahead and do diagnostic work. This will be my first time in 12 years of computing that I will not have been able to fix something on my own. Ouch. EDIT:: Just a side note, I also monitored the mobo temperature and it is staying between 32-40C, so it is definitely not overheating. EDIT:: Just became a Google Certified Expert on IRQ conflicts. Viewed all the IRQ assignments in Device Manager and no duplicate assignments.
your motherboard http://www.msi.com.tw/program/products/mainboard/mbd/pro_mbd_detail.php?UID=636 Supports AGP 8X interface One AGP slot supports 8x/4x (AGP 3.0) at 1.5V
Another update... I've applied new thermal paste to the card and insured the fan and heatsink are clean and working properly. I attached a heat sensor to the GPU and found that it is well within normal operating temperatures. I'm still trying to get my hands on a PCI Express card to test that theory. Could the AGP slot possibly have some sort of physical damage that is causing a short? I really think this is all going to come back to the video card and/or AGP slot. When I leave my computer off for awhile, I can usually surf the web without it locking up after 20 minutes. But if I jump into a graphics intensive game, it locks up within 5-10, no matter how long the computer has been idle.
take the ram to bestbuy *they sell the same ram* tell them to test it takes all of about 1 min or so to test. i belive thats the problem as i have had the same problems before with some ram
Sorry, I meant PCI not PCI Express. My mobo doesn't support PCI-E anyway. About the memory.. could it be that it locks up when I play a game with a heavy load because it's loading the RAM up and finally hits the bad spot?
