I still have problems with my homebuilt PC

Toggle sidebar Toggle sidebar
J

Jacob Collstrup

Distinguished
Sep 6, 2007
58
0
18,630
I'm writing it in a new topic, since it seems the old thread died out, but its still kinda the same problem.

The old thread is here:
http://www.tomshardware.com/forum/247032-30-problems-homebuild
I couldn't figure the 'Related threads' system

The system is:
MSI P35 Neo Combo mainboard, BIOS version 1.6
Intel Core 2 Duo E4400 CPU
Corsair XMS2 XtremePerformance 2*1GB Memory
Inno3D Geforce 8800GTS, 320MB Ram, PCIe x16
RAIDMAX Power Viking 630 PSU

Yesterday it ran flawlessly for several hours, including roughly 2 hours of multiplaying in Call of DUty 4. No signs of problems as far as I could tell.

Then just this afternoon, when booting up, the just as I got into windows I got the bluescreen and the screen went blank. I don't know if the computer tried to reboot. If it did, then it failed. I tried to turn it on 3 times. And only the third time did it actually wake up again. I just updated the BIOS last monday (7th of january). And the computer still serves me these problems.

Does this mean that its a powersupply problem?

Jacob
 
The bluescreen dissappear very fast, I don't get to see what its about. And the memory was tested last week I think, using memtest x86 from a bootable cd. The test went on for 1½ hours I think and completed 5 tests, no errors was found during this test.

I don't know if this is important, but the computer crashed with a blue screen (Still too fast to read it) and then rebooted, while running the 3DMark 2006 benchmark test.
 
Turn off the reboot on error option. The following assumes Windows XP:

Right-Click the My Computer icon and select Properties.
Click the Advanced Tab.
Under Startup and Recovery, click the Settings button.
Under System Failure, uncheck Automatically Restart.
Click OK.

-Wolf sends
 
What are your CPU temps like? Also, have you fiddled with OCing at all?

When I OC'd my Q6600 from 3.2 to 3.6 just by changing my multiplier from 8 to 9 everyone worked. Ran Prime95 and got blue screen fast. I know my computer is stable at 3.2 so I just dropped it back down and have accepted 3.2 as good.
 
Hmm...I haven't checked the CPU temp just before or after the crash occurs. But usually the CPU temp 28 degrees celcius or thereabouts (82 degrees fahrenheit) and the motherboard is usually at 40 degrees celcius (104 fahrenheit).

The BIOS is running 'Optimized defaults' so I guess it could be overclocking a bit. I know the CPU is set to be running at 2000MHz and most times its running at 2010 or 2011MHz...I don't know if that counts as overclocking.

But now that were talking about temperatures. I'd like to have a program that can keep track of fanspeed, temperature and voltage. I've tried MSI Dual Core Center, but I don't trust it, I've also been adviced not to use it. Will Asus PC-probe work on an MSI mainboard?

Jacob
 
Everybody here pretty much always use Speedfan to monitor temps, fan speeds, and voltages. It can also automatically control your fans and has graphs that let you plot all readings over time so you can keep track of things like temperature when stressing the system.
 
Well...speedfan doesn't seem to be able to read the temperatures right. 'Temp 3' is -128 degrees celcius (-198 fahrenheit). However, there appears to be a 'flame' icon out from what appears to be the GPU temperature (That is; a flame icon next to what I reckon is the graphicscard temperature. Obviously I don't know much about this program! :kaola:

I got to see the blue screen again. Though it didn't appear to be a driver fault, it said something about 'Page error in nonpage -<something>-' This was close to the top of the screen. At the buttom it said something about 'Loading physical memory'. It did mention physical memory.

I freaked out when trying to reboot because the computer kept crashing in windows (no blue screen), it just kept working on some process that wouldn't be done. This was PlexTool. I realized that my USB-stick was in, it creates some sort of virtual cd-rom drive, for some reason I don't understand and this screws up plextools in the windows startup. Anyway, after removing it I got that blue screen I described above. I rebooted and the computer said something about the 'windows\system32\system is damaged or can't be located'. I got angry and rebooted again figuring that if I ignore the problem hard enough it'll go away. As assumed the computer booted up 'correctly' this time.

Seriously I'm lost with these problems. I just don't know what to do about them.

Is it the powersupply thats too weak and makes the computer freak out or is the graphics card running to hot and thus make the computer freak out. Or something completely different?

I checked the 'exhaust port' on the graphics card. There is nothing blocking it. Not that I could see at least.

Jacob
 


Is this the same old PSU from your first thread? Have you tested this old PSU? Is the CPU temp look right? If not, re-install your CPU HSF. Good air flow thru the case? MemTest look good? Now, use a different (good) PSU (borrow from a friend if you have too) and narrow down your problem. Than download this tool (Sissoft Sandra) and run your system for at least 2-4 hours. This should tell you how it system perform.
http://www.majorgeeks.com/download92.html
Hope this help...good luck!
------------------
System:
* Intel Core 2 Q6600 * Tt Orb II Blue Cooler * BFG GeForce 8800Ultra 768MB DDR3 * ASUS P5B-VM DO with Tt LANBox microATX case
* Crucial Ballistix, 2x1GB, PC2-6400, CL=4-4-4-12 * Ultra X2 750-Watt with UV & SLI ready * 1.5TB Seagate RAID 0 * 150GB WD Raptor X
* Lite-On 20x SATA/DVD/CD Writer with LightScribe * TrackIR 4 Pro * Cougar HOTAS * Dell UltraSharp 3007WFP LCD (2560x1600 native resolution)
One can never have enough!
 
Its still the 630watt RaidMax PowerViking as in the last thread. I don't have anything else to try with. I suppose I could ask a friend if I could borrow one. Memtest looked good, with no errors after 5 complete tests. The CPU temperatures look fine. just below 30 degrees celcius. But speedfan reports a mainboard temperature of negative 128 degrees celcius!!! That just can't be right no matter what LOL!! :heink:

I don't know what CPU HSF is...I have no clue on that and I couldn't find any sticky in the CPU forum to help me out. So I guess I'd need help to reinstall it if neccesary.

As for airflow...I think my cabinet is good. The previous computer (AMD Athlon 64 3000+) ran very cold. NEVER more than 35 degrees celcius with a Zalman cooler.

I already have Sisoft Sandra, but again I don't know how to use it... :??:

I'm not too good with computers! LOL :kaola:


Jacob
 
Yeah, speedfan usually has one or two temps that are at impossible levels. you can remove them by going into the congiguration settings. What are your voltage readings like in speedfan? Post back all voltages here so we can see if your power supply is possibly dieing.

You need to find out what the bluesceen message says. Go to your start button, right-click on My Computer, select Properties, go to the Advanced tab, and un-check Automatically Restart under Startup and Recovery.

Now if it bluescreens again, the error message will stay up so you can write down what it says and that will help a lot in troubleshooting.
 
Well...the current voltage as reported by speedfan in this moment is:
VCC3V: 3.36V
VCore: 3.50V
VRAM: 2.05V
VChipset: 1.12V
+5V: 5.13V
+12V: 12.23V
VCC1.5V: 0.81V
VSB3V: 3.36V
Vbat: 3.23V

OK...it seems to me, when looking at the above that the VCC is grossly undervolted, it gets slightly more than half of what its supposed to get....is this normal?

Next time I see the blue screen I'll right it down on a paper and post it here.

Jacob
 
Perhaps you should go into the hardware section of your BIOS and verify that VCC1.5V is really reading that low. As a general rule-of-thumb, the actual volts on a power supply should be about +/- 5% so a near 50% drop is not a good sign. To be honest, I'm not sure what VCC1.5V sends power to so I can't say that's definitely your problem, but it's mighty suspicious. Whatever that line is powering may be failing you when it is stressed.
 
I would say swap out your PSU first, as it's not a good idea to have a crappy PSU, Raidmax PSU's are crappy because they are Case bundled PSU's.

Even if that might not fix the problem, it's worth the investment.
Next, I would check your memory, try swapping out the sticks for some slower RAM, as this generally helps a lot of people who get chronic BSOD's. Even though you passed Memtest, that doesn't necessarily mean your memory isn't incompatible with your motherboard. I bought some DDR2 800 OCZ's once for my current rig, no errors in memtest EVER, but I would consistently BSOD. Had to swap them out for some 533's which fixed the problem.

Last thing, I would look into the temp your GPU is running at anything up to about 65C is ok. Any higher, that could be the source of the problem.
 
Ok, the machine acted up again today. First boot and I got a blue screen, it took 3 attemps after that before it finally booted up. I am now writing from this machine, actually all my posts here are written from the faulty machine.

Anyway, I managed to write down the blue screen, but its in danish so I have to translate it. So the wording might not be a 100% correct compared to the error message of an english machine.

Here goes:
--------------------------------------------------------------------------------------------------------------------------------------------------------------
The error might have been cause by the following file: Win32k.sys

PAGE_FAULT_IN_NONPAGED_AREA

If its the first time you see this stop error, you should reboot the computer. If the screenimage is shown again, you should do the following:

Make sure that all new hardware or software is installed properly.

If this is a newinstallation, you should ask the hardware- or software provider, if there is a windows upgrade, that you might need.

If the problem persists you should deactivate or remove hardware or software that has been installed recently. Deactivate some configurations for BIOS-memory, fx caching or -<I don't know this word, directly translated its "shade">-

If you need to use Failsafe mode to remove or deaktivate components, you should reboot the computer and press F8 to choose advanced start settings, and then choose failsafe mode.

Technical information:
*** Stop 0x00000050(0xFE5664DC, 0x00000000, 0xBF802291, 0x00000000)

*** Win32k.sys - Address BF802291 base at BF800000, DateStamp 45f013f6

Starts loading of physical memory
Loading of physical memory done.

Contact the system administrator or technical support for further information.
-----------------------------------------------------------------------------------------------------------------------------------------------------------

That was it....I hope it makes sense to some of you.

This problem is both scaring me and pissing me off at the same time! :??:

Am I nearing a full system re-install or what?

Is it still the PSU that causes all this fuzz?

I looked into the BIOS today to check up on those voltages that I posted last night. I couldn't find them. And the system health in the BIOS showed all good. (I really hate contradicting info!! LOL)

Jacob
 
This Stop message usually occurs after the installation of faulty hardware or in the event of failure of installed hardware (usually related to defective RAM, either main memory, L2 RAM cache, or video RAM). If hardware has been added to the system recently, remove it to see if the error recurs. If existing hardware has failed, remove or replace the faulty component.

http://www.microsoft.com/technet/prodtechnol/windows2000serv/reskit/w2000Msgs/6093.mspx?mfr=true

Everyone keeps telling you to do something with your RAM.
 


I know 🙁 ...But most people also tell me to change the PSU.

But I don't know what to do about the RAM, I've checked them in the BIOS and they get the right timing and volts according to the BIOS.

I guess the computer both needs new and slower memory and a new PSU...

Jacob
 
You should try one DIMM only in slot 1. In BIOS, adjsut the voltage to 2.1v (for PC6400). Manually set the timings to Mfg. specs. Run some software on the machine, check for stability. If stable, add the second DIMM and check for stability. If the machine bluescreens, try like setting the memory voltage to 2.2v and loosen OR tighten the timings (try both 5-5-5-15 and 4-4-4-12). If It still bluescreen, borrow some other RAM and try it. The error message you are getting is RAM related. A message to Corsair support describing your problem may get you some advice. Have you thought about RMAing your RAM?

http://www.corsair.com/helpdesk/default.aspx
 
I have the Stealthstream 600w sitting here in the retail box for months. I read a good review and bought it cheap for a spare. At least it would be a nice upgrade for you if nothing else.

http://www.sysopt.com/features/cases/article.php/12024_3684541_3

I built my first computer with the help from my former next door neighbor many years ago. He is a mechanical engineer from Denmark and moved back there years ago and I have not heard from him for ages. FTR it was a Pentium 1 with a 100mhz. cpu. with like 32MB RAM. I upgraded to a 133mhz. processor and 64GB RAM (can't remember, but I still have the computer!). My friend built windmills used to produce electricity for a living.
 
I've tried to run it with one mem-block at a time. And they both seemed to work. I didn't stress test the computer for stability back then, so I'll do it again.

But this is the part I can't figure and need help with. According to this document they're supposed to run at 1.9V http://www.corsair.com/_datasheets/CM2X1024-6400.pdf As far as I can tell thats the Mfg. specification. Should I STILL set the voltage to 2.0V or even 2.1V?

Its even worse with the timings, because they're set up in a completely different way in the BIOS. There the timings are set like "CL6" or something like that. And I don't know how to adjust that to something correct or better. I can't even find the specifications for the RAM timing presented in this way.

Its a bit confusing when I'm told to both change the voltage TO Mfg. spec. (1.9V) and told to set the voltage to 2.0...

I know too little about the hardware to know if upping the voltage like that poses a serious threat of burning out the memoryblocks. Thats one of the reasons I'm so cautious with this.


Jacob
 
You most definately need to 'bump up' the voltage to 2.1v. Regarding the timings, it sounds like you have the memory timings set to 'Auto' in BIOS, I don't know, that is what it sounds like. Set the BIOS TIMINGS control to 'manual'. Take it off AUTO and manually set the timings to 5-5-5-15 OR 4-4-4-12. Leave all the other manual memory settings at auto.

So in BIOS...memory voltage is 2.1v. Memory Timings from 'AUTO' to 'MANUAL' with 5-5-5-15 OR 4-4-4-12.

PC6400 is 'manufactured' to run at 1.8v generally. But the different Mfg's give their RAM a rating like guaranteed to run a 800mhz. at 2.1v 4-4-4-12 timings. Perhaps the 1.9v. is just an industry 'standard' you are reading. Your problem is classic RAM related. Applying the correct voltage and timings to your RAM may solve your entire problem and stabalize your system.
 
OK. I'll first try to bumb up the voltage on the memory to 2.1V.

The timings are on BIOS default. But when I change it to manual then the timings are represented in a way that doesn't match those timings you tell me about. I'll write the timings down as they appear in the BIOS, but otherwise leave them as they are for now.

I'm sorry I'm so difficult and helpless, I just don't know much about this whole thing and I'm piss scared that I might ruin something permanently if I set something in the BIOS wrong. 🙁 Thats why I don't proceed with these things untill I'm totally clear and what to do and how to do it. I just don't know how much those blocks can take.

Jacob
 
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom