PC Freezing

Well upgrading the BIOS would be the first thing that I'd have missed - so I'll try that when I get home.

In XP - I installed the Audio driver from the disc, when they finished, the XP driver installer came up for the audio - everything looked OK, but at the end it said something along the lines of Driver Invalid. I'm not going to worry too much about XP since I know that I'll have to run the repair option on it and start again anyway. I haven't noticed any freezing on XP - but I haven't really used it as much since I prefer Vista as my main OS at home.

Regarding the Audio driver, I just remember there being two steps for HD Audio.  There's a Microsoft HD Audio base driver (or something like that) that I had to install first on XP.  That could have been specific to the HP and Alienware laptops I was using, though.

Yeah, try the BIOS.  I just updated mine and it was like 12 versions old.  And my laptop is less than a year old.

<-=+=->

Hello Ryan,
A RAM check may be in order.
Check here for the goods on 64-bit Vista. I'm running it and it's great: 8GB of RAM at my fingertips and it's slightly faster than 32-bit.
http://mistywindow.net/?p=8

Note my links in a comment at the bottom re MS's hardware compatibility link.

I might give it a shot - I'd have to find some free software to merge two small empty partitions (without stuffing up the data on a third partition) to be able to install it. The only thing putting me off is that I'd have to download stacks of MBs of x64 software that I don't have in my 'Software Archive' folder.

Does software use more than one CPU core in 32bit Vista? Or do you need to have 64bit for each core to be used?

32 bit will use all your cores.

For a good utility for partitions, use Paragon's Partition Manager.  It's awesome.  There's a trial version.

<-=+=->

I've found the BIOS files on the XFXforce website - but am not exactly sure which one to go for - I'd rather go for the smaller on, and am not 100% sure why there's be another BIOS update that's 74mb???

Forgot picture:
Capture.JPG

Tried the nVidia version but the updater software said the ID's didn't match - so it looks like I'm D/L'ing a 74mb update now.

Wow.  74MB is a HUGE BIOS update.  I mean HUGE.

I think there's no doubt that you've got to go with P04 since that's the latest.  I'd recommend rebooting and looking at what type of BIOS you're running.

Also, it may be thatyou ahve to run the XFX BIOS update and then the Nvidia update.  But, check out the release notes first before you do any of that.

I'd think that the .49MB download would be the correct one, too.  Make sure you read their documentation.

<-=+=->

Well 73mb of it were the Mobo System Tools that I already had, which I installed instead of the ones I already had and then ran the tool to use the 1mb bin file to update the BIOS, which was successful. Since it was late I just did a Shutdown after it was finished and went o bed, so I'll see if it makes any difference tonight when I get home from work.

That makes sense.  I was seriously wondering about a 74MB BIOS update.

Let me know how it turns out.

<-=+=->

BIOD update hadn't changed anything in Vista Ultimate x86. I turned it on and walked away (PC doinging nothing besides startup). Came back to find the PC frozen again (systray clock not usually 17 mins slow)  :)

I've been using XP for the past 4 or so hours without a hickup - there have been reboots that installations, etc required, but thats it.

I installed the Paragon trial - but can't get it to merge the 2 partitions. Whats the point of having a 30-day trial with limited features? You either have 30 days to try everything, or as much time as you want to use limited features? Running the 'Redistribute Free Spaces' says that I can add up to 158 GB to my C drive, but when I click Next no volumes are listed for me to pinch space from.

I have C: E: and F:
C is a boot partition - so can't really touch it, although I can repair the BCD from the Vista disk if necessary. E drive is empty. F drive has crucial data that I don't plan on moving as it would take quite a lot of time.

I want to merge C: and E: and presever F: so that I can try Vista x64 on the merged drive, since 16GB isn't really enough for it. It was enough for older XP installations from 2 PC's ago. Other partitions was used for TV Tuner Card recording.

No success with the sound car in XP either. All sounds come out fine from the Realtek HD AQudio app, but media programs and Windows Sound control applet still report no sound device installed and don't play music. System sounds alos don't work - use PC speaker. Device manager showed the Realtek device with no issues and a Multimedia Controller with issues - no driver found. I forced the latter to find drivers from Vista 0 and it turned out to be the TV Tuner card controller. There is also a 'PCI Device' and 'Other PCI Device Bridge' coming up as uninstalled with no available drivers. I thought that's what the XFX motherboard disc was for - sigh.

I'd like to test to see if gtames in XP cause freezing - but not while the audio isn't working. It's looking like I'll be reinstalling both XP and Vista again, but I'll save that until a later time this week after you guys post anything else. I'd also like to have the partitions fixed before then.

If I remmeber correctly, booting up with the Vista disc provides the option to test your RAM, so I might give that a go tomorrow.
 Downloaded Realtek drivers from XFXforce.com.

I'd definitely check RAM. I had problems recently with RAM in Vista 64-bit which didn't manifest themselves with 32-bit.
Different problem to yours, but ....
Neither the Vista test nor memtest showed any problems, but removing RAM modules one by one isolated one defective module.
MegaGenius garycase came up with an explanation here:
https://www.experts-exchange.com/questions/23298887/RAM-OK-in-32-bit-OS-but-fails-with-Vista-64-bit-Why.html

HTH

That's a good point.  I had an Alienware laptop that kept freezing and doing stupid things.  Using the memory diagnostics from Microsoft showed that everything was good to go.  However, once I started swapping RAM, I found that one of the two sticks of RAM was bad.

Alienware shipped me a new stick of RAM and I installed it.  Within a week, the memory was bad again.  I went through three sticks of RAM before I sent it in for a motherboard replacement.  As it turns out, one of the memory slots were bad and was putting out too much voltage which was frying the RAM.

Regardless of what you discover now, you might want to see what happens when you move the RAM to different slots.

<-=+=->

Also, keep in mind that garycase (who is, by the way, a MegaGenius) was explaining why a memory problem would be more pervasive in a 64-bit OS.  The main problem here is with a 32-bit OS.

<-=+=->

Well it took over 4 hours of use of XP x86 - but it finally froze for the first time last night - and also again today twice a lot sooner while trying to respond to forum posts. I had an awesome post lined up for some other guy's question and looked up at the screen to notice I just wasted 30 mins of checking prices for nothing since I couldn't hit Submit. Grr. On the laptop now while the Vista mem check is running - guess I'll cancel it and take out one stick.

The RAM was the last item I needed to build the PC, for which I had to wait a week for some to get instore. So not looking forward to having to replace bad memory (or worse the motherboard). You'd like to think there are high quality control standards on new parts like this, sigh   :|

RAM is currently in slots 0 and 2, I'll try slots 1 and 3 and see if there's any miraculous difference. Still no joy with audio in XP.

And yes, Gary's definatly a legend here when it comes to hardware - he's probably answered more of my Q's here than anoyone else.  :o)

For the record, I have 2x 2gb sticks of Corsair DDR2 RAM @ 1333MHz
http://www.centrecom.com.au/catalog/product_info.php?cPath=115_23&products_id=30041

That's good stuff, man.  But, hey... my Alienware was flakey and those are high-dollar machines.

<-=+=->

Well I've moved the RAM to the other slots, and I've taken out my TV Tuner. I was also able to fix up a USB header cable for the back of the PC. XFX gives you 'cards' that go into the PCI bays (not slots) so that you can use all the Front Panel Headers (ie USB and 1394) if your case doesn't have those ports on the front. The Antec 900 has only 2 USB ports at the front - so i can send two of the headers to the back. Also turned a few fans up, although I think I might turn one back down when I put the TV tuner back in).

I decided the best test would be to boot Vista up and see if I can set a Dreamscene wallpaper. Normally this would freeze the PC no matter how long it was on for, plus I wanted to see what was new in content packs 2 and 3 (had only seen pack one before). And it worked. And some of those DS movies are quite nice - using one now.

I'll continue to test the PC out - either by leaving it unattended and checking it later on, or by launching some game (loving the graphics in Colin McRae DiRT, FSX, TDU, Tm United, etc).

So I'll see how this goes - and quickly hit Submit before it does get the chance to freeze   :o)

I read recently that it's best not to use the default pair of slots recommended with most motherboards for 2 x 2GB of RAM, if they tell you to use the black ones, use the yellow instead. Don't knw how reliable that info is.

<quote>
Also, keep in mind that garycase (who is, by the way, a MegaGenius) was explaining why a memory problem would be more pervasive in a 64-bit OS.  The main problem here is with a 32-bit OS
</quote>

Yeah, I know. I was the questioner and I'd been beating my brains out for a week before finding the answer.

As I said in my previous post, mine was a different problem, but it indicates that memory testing isn't always the answer to bad memory.

I've now got 8GB of cheap A-data RAM and it's rocketing along with 64-bit Linux and multiple virtual machines - luxury!

:o)

Well after posting that last comment I walked out, and came back in a few mins to find the PC frozen again.  :(

Since then I've rebooted, put DS back on (gets disabed as a caution), and watched a DVD in WMC to test the Dolby 5.1 speakers I have (very nice). So I'm guess the problem isn't solved yet - I sure wish it was though. I'll leave this on and see how long it lasts.

Going to have some tea now - will be back later.

I decided to try and take out a module of ram to see if either one was corrupt or if the PC would handle 2gb much better than 4.

Got the RAM package out, and noticed that that card folded out. Sure it just says how to insert ram, big deal - but I thought I'd read it anyway just in case. Then right at the end it hits me with the following:

"PLEASE NOTE: Corsair's line of 4GB memory kits are designed for PCs featuring 64-bit operating systems. Installing 4GB onto PCs with 32-bit operating systems is not advised nor supported under warranty."

The good side is that I may have found a fix for this by installing Vista Ultimate 64bit and that I won't have to waste time and money trying to exchange the product.

Bad side is that I have to download everything I have again in 64-bit mode, and I have to acquire a Vista Ultimate 64bit license key (legally of course). Or is a Vista Ultimate license compatible with either mode?

Also, I used to dual boot with XP for the few programs that didn't work in Vista. If for arguments sake I try to dual boot XP 64bit and Vista 64bit, will programs that I used to run on XP 32bit still work on XP 64?

Tried to install Vista x64 twice now. Everything in the first phase completes successfully (in under 9 mins). When the PC restarts though and starts to boot Vista off the HDD, it keeps BSOD'ing with an error like 0x0000007e

I tried booting in Safe mode - and the last driver to appear before the BSOD was crcdisk.sys

Any ideas? Getting quite frustrated now  :(

The Vista Ultimate key is good with 32-bit or 64-bit; you don't need a separate key.  Do you have the media?

Regarding XP64 - application compatability will vary by application, just like with Vista.

Wow... this really blows me away... that RAM wouldn't work with a 32-bit OS.

<-=+=->

Try disabling USB keyboard support:
http://forums.microsoft.com/TechNet/ShowPost.aspx?PostID=510682&SiteID=17

<-=+=->

Disabled USB Mouse and Keyboard support in BIOS, unplugged USB Hub and printer unplugged, no joy.

I'm under the impression that I might have to increase the voltages sent to my RAM - although I'd prefer it if someone knew what I should set it to. In the meantime I think I will remove 2gb, finish install Vista 64, then put the 2gb back in and see what happens.

Man... I think you've either got a bad mobo or bad RAM.  My inclination is always towards the mobo, unless you've bought really crappy RAM, which you didn't.

Go to www.crucial.com, put in your mobo and see if you can't order some really inexpensive RAM - maybe 2x1GB - and then see what happens with the machine.  If it continues to freeze, then your mobo is bad, in my opinion.

Which really blows.

An alternative is to use the System Configuration tool in the Administrative tools to try to determine what it is that's killing the PC.  (That's in Vista).  But, my gut says mobo.

<-=+=->

It's possible, but the mobo is a very good one too -  XFX nForce n790i

I'll see what happens with 2gb of ram in XP 32bit and 4gb in Vista 64bnit over the weekend.

I reduced the RAM to 2gb to finish off the Vista 64 installation. Everything worked fine - used the PC with 2gb for most of last night while installing stuff. Since everything was up and running I decided to put in the rest of the RAM again - and booting Vista 64 I got the same BSOD - 0x0000007e / crcdisk.sys

XP still boots and works fine with 4gb until it freezes, so this makes me think that the hardware works OK.

Either it could be the motherboard or memory issue here. Last time I struggled with an Asus motherboard and replace the same make/model motherboard and everything went well. I even claimed on Vista, memory, drives, etc. even hardware test gave no error or whatever indicated to problem during trouble shooting as you are doing now. 64 bits and 32 bits installation just fine, and then BSOD, so on. The "bad' motherboard will run just fine with 2 G (either 4 sticks of 512 or 2 sticks of 1G), but problem comes when 4 sticks memory insert to all 4 slots. And guess what? I build that board for my kids with 3 G memory and it run just fine since.

That's kind of where my gut's leading me... I think it's the motherboard.  There's no way the OS is going tank just because you increased memory.  It's got to be a hardware issue... I'd replace the motherboard.

<-=+=->

I'm starting to think of that as a real possibility. Although admittedly I'm somewhat in the frame of mind of 'that could never happen to me' even though I've told others to replace such components before with success.  :)

I've put the other stick of RAM in by itself (2gb)  and everything works. I put it in the other slot this time and will attempt to do the same when adding the other 2gb back in. (Was in slots 1 & 3, will try slots 0 & 2 again).

So, you've installed each stick of RAM individually and it works, but when you put them in together, it doesn't.  And you've tried installing the pairs in in different slot pairs.  So, it would seem to me that it would be a defective motherboard and not RAM.  The OS definitely isn't going to crash because it has too much memory.

Yeah... I know. I've got the same, "It could never happen to me" attitude....  Kind of like backing up our data.  :)

<-=+=->

Yeah, I think same with you Ryan :o) ..... why did it happen to me ... It took me almost a week long, then new but same make/model board put in and it worked charming. After that, I had enough extra components to build a PC for my kids ... hehehe

Computer is so fun, isn't it?

Tell me about it, I work on an IT Service Desk.

I put two modules of RAM back into slots 0 & 2, got a BSOD. Took out one module from slot 0, PC froze.
I'd say this confirms the motherboard needs replacing. I've since moved the module to a slot 3 since I think that still works.

I can only hope that a new motherboard will definatly fix all problems with this new PC so far. I still think it's a shame I can't use a 32bit OS that just sees 3gb of RAM and leaves things at that.

That's what I'm doing.  I've got a really nice HP laptop with 4GB RAM.  The problem, though, is that Cisco refuses to release their VPN client for x64 Vista, and without that, I'm dead in the water.  So, I'm running i386 Vista, so I see only 3GB RAM.  But that other GB is out there... it's not being used... that bothers me.  ;)

<-=+=->

Well I'll be taking the mobo out of my PC tonight and taking it back to the store tomorrow. I rang up the store today and was not impressed to hear that I can't get an immediate exchange, and that I have to wait at least 2 whole weeks while they send it back to their supplier to look at it. Rubbish service if you ask me. I could understand if it was a non-essential peripheral that I could live without for a little while, but not a motherboard, especially considering how much those one's are worth.

I'll make sure that I convey my dissapointment with them tomorrow - I have a number of projects that I need to work on on my PC (no I'm not talking about simple school assignments that require MS Word - projects that require software like Visual Studio and Sony Vegas Pro).

I'm working in Visual Studio right now... not for the faint of heart.  :)

Sorry to hear about the lousy service - especially on a $350-$450 mobo.  It will be interesting to see what happens when you get the replacement.

Good luck!

<-=+=->

Yes - I'll keep the Q open until I have a working, stable PC   :o)

There's still "at least one week" to go before I should be able to get the motherboard back - I can't imagine how many days it's just sitting around with no one looking at it, but that's another story.

I was just wondering if I had to worry about the thermal paste on the back of the CPU which would've dried up by now. Will it just 'melt' again when I put everything back together again, or will I need to do something else?

I'd go ahead and get another tube of it.  Don't worry about getting all of it off the fan or the CPU, just clean it off as best you can, squirt some more on there and you should be good to go.

<-=+=->

I picked up the motherboard today - and it looks to me as if they've just given me a brand new board and not worried about repairing the old one (as far as I know). Why they couldn't do that 2 weeks ago is beyond me, but anyway.

So now I have 1 motherboard, but I have 2 lots of all the accessories, cables, CDs, manuals, etc that come with it. And 2 boxes.

I got some thermal paste as well for free (after asking) and so I'll see if I can go about putting things back together tonight if I have enough time.

I'll let you know how things turn out.

Please do.

Good to know they replaced the mobo... better to have new anyway.  And always nice to have spare parts and accessories!

<-=+=->

I had spare parts with just the first box  :0)

At least I can chuck out all the spare parts box full of cables and other old stuff dating back from the days of 386's  :)

*GASP!* Throw stuff away?!?  Surely you jest!!!

I finally threw out the box of cables, CDs, etc. for my old 486.  It hurt, but had to be done.  :)

<-=+=->

I quickly put everything back together and turned it on late last night, hoping (but not expecting) that everything would just go smoothly.

During the first phase of bootup - it takes a long time to detect my IDE drives - but both eventually show up successfully. However it gives me the error - "ide channel 0 no 80 conductor cable installed"

I did a quick Google late last night and others reckon that this doesn't impede their bootup procedure - mine doesn't progress beyond showing me the table of all my devices, system busses, IRQs, DMI, etc - the point where it tries to bootup from DVD or HDD.

I'm quite sure the IDE cable (that came with the faulty motherboard) is plugged in properly, exactly the same as last time. Connections are as follows:

Sata A0 - Nothing
Sata A1 - HDD
Sata B0 - HDD
Sata B1 - Nothing
Sata C0 - DVD Burner
Sata C1 - Nothing
IDE Master - HDD
IDE Slave - DVD Burner

The sata setup is probably different to what it was previously in the other PC (most connections were in B and C area, nothing in A), but the IDE is the same as last time.

Any ideas?

Well, I would install basic components with the mobo outside the case for now. Plug the IDE drive only (No DVD burner) during this time see what what happens.

Just found another site that had the following suggestions:

Check IDE connections
Clear CMOS
Unplug USB headers
Replace battery/BIOS chip

Will try everything when i get home tonight

Stop by any electronics store on the way home and get an IDE cable (just in case). Should not connect DVD and hard drive on same IDE cable.

There is only one IDE channel (on the n790i series). I can also try using the IDE cable that came with the new mobo yesterday. There were no problems two weeks ago to this effect. I'm also under the impression that I might have to use the Vista repair console to redo the MBR since the HDDs are on different channels now. So I'll give all that a bash tonight and let you guys know.

Ok - I have removed all errors from the system by resetting the CMOS and changing the IDE cables (although they're the exact same type/brand/# of wires).

Everything was back to normal - and by that I mean back to how things were. I have had the PC freeze on me about 4 times since putting it back together. For the past two days I have:

Repaired the boot manager
Installed SP1 onto Vista x64 (haven't reinstalled Windows again yet)
Reinstall motherboard and gfx card drivers over the top of themselves
Reinstalled Logitech gaming device drivers
Reinstalled Fraps
Reinstalled Test Drive Unlimited (game)
Applied other Important Windows updates.

I am currently only running 1 module of 2gb ram in slot 0.

Assuming the motherboard isn't at fault, there remains the following possibilities as to where the faults lie, and my reasoning as to why it might not be at fault:

RAM          - tried the other module in the previous build with no real difference. Will try a swap again after posting this message to check. If it failed, I would expect the PC to restart, not freeze
PSU          - I have more than enough power (1200w). If it failed, I would expect the PC to turn off, not freeze.
GFX          - If it failed, I would expect the screen to go blank, or produce a random image
Heat        - I don't think things are getting that hot - the fans on the Antec 900 case are set to medium and high (med going in, high going out).
OS           - Have updated Vista to SP1 with no dif. Only once did the PC freeze prior to to installing the DreamScene update, after then it froze a few more times when dealing with the larger movie files. I used System Restore to go back, with no apparent resolution. As much as I don't want to format and reinstall again, I would be willing to do it if I knew there was a real good chance of it fixing something.
Drivers    - Tried multiple versions of motherboard and gfx card drivers in previous build, and reinstalled the latest ones in this build.

What the PC was doing for the last few freezes:
1 - Playing Test Drive Unlimited
2 - Copying DreamScene files over to Windows dir
3 - Dreamscene files again
4 - Using Sony Vegas Pro to work on a video I'm producing.

There have been other odd things since resuming use of this PC, such as Windows Photo viewer not showing any photos that I dbl-click on, Fraps crashes TDU when I try to start recording - goes back to desktop, and occasionally the G25 re-initializes when I launch TDU which means the game doesn't see the controller - have to exit and relaunch it.

That's all I have to report for now - I'm going to swap the RAM with the other stick (still will use slot 0) and see if anything changes.

I swapped the ram - and the PC barely lasted a few mins before freezing, doing nothing special to cause it. So it's *wrose* that the other stick, which I have swapped back into slot 0.

I guess you'll tell me to take the RAM back tot he shop now?

Is there any possibility that the gfx card could be faulty? What would a faulty gfx card produce?


One other new update - I was attempting to play Flight Sim X over a LAN with my younger brother, who has my older PC which works fine. For the first time, the monitor turned off, but the sound continued to work. I went into my brothers room to inform him that my PC had yet again crashed which is why I dropped out, but he said I was still flying as normal, getting further and further away from him.

As soon as I hit the reset button, the video came back fine without any glitches.

You may have already mentioned this Ryan. Have you eliminated a  defective slot rather than RAM?

You can use the memory test built-in Vista to test the mem, also use Reliability & Performance to monitor, collect data, and view report see if you can find any unusual that might cause issue.

I have a brand new motherboard. I'd like to think there's nothing wrong with it. The fact that the PC froze with the old mobo no matter which slot I used also points to the fact that it could be the RAM.

I was told elsewhere (maybe even here) that the Vista memory test wasn't exactly accurate.

Of course, it is just quick test so we could not expect more, but I use it often and several cases that helped very exact. No harm or waste time to try that. Also monitoring and check event from Reliability & Performance (built-in tool in Vista as well) that might give you some input for analyzing the cause.

You know, I had some really strange behavior on a PC once and it was the case.  For some reason (and we really never figured out why), it was shorting.  To test this, you can try taking the mobo and power supply out of the case and connect everything to it and run it that way.  We replaced the case (with the exact same model, no less) and it worked perfectly.

It's a shot in the dark, but it's a test you can run without swapping more parts.

<-=+=->

<quote> I was told elsewhere (maybe even here) that the Vista memory test wasn't exactly accurate </quote>

It was here.

:o)

Yeah, I was chatting to BooMod earlier this morning about the newsletter and such, and he said he had one PC where one side of the case was 'dinted' so that it made contact with the underside of the motherboard, causing problems.

I have the Antec 900 case - and am pretty sure that there's no contact - I have looked there previously

Today I took my RAM back to the store to get it replaced. I was told that I would have to wait 6-8 weeks for a repair or replacement. I told them off big time and they lent me a 1-GB stick of Kingston DDR3 RAM @ 1066mhz.

I've put that in, and the PC is even slower than before as expected. I tried to compile a movie I'm working on in Sony Vegas Pro, which almost always causes the PC to freeze (besides other methods such as playing games, Dreamscene, etc). Without fail, it froze at about 30%


Now what do you think? Where's Gary when you need him?   :o)

Grasping at straws here, but how's the health of your hard drive?  Any bad sectors?

<-=+=->

All HDDs are in good nick, no tell-tale sounds from them, all by Seagate

250GB IDE 16mb cache
320 GB SATA-2 16mb cache
1 TB SATA-2 32mb cache

Well, I gotta tell you - I'm stumped.  I mean, the only thing left to swap out is the CPU.  I don't think I've ever had a bad processor.

Have you?

<-=+=->

I've never ever heard of one being faulty. What about the gfx card? If I had to choose which device would more likely cause the PC to freeze in the way I'm experiencing, I guess I'd go for the CPU before the GFX card.

Just wish I could sell the whole PC and start again. I'd get the exact same specs for the PC, but be getting one that works this time around.

PS - Played Race Driver GRID today - it's truly awesome   :o)

I'd pick an OS that's stable and then put it on eBay.  :)

I don't know, though--- I may go with GFX before CPU.  I mean, I've NEVER, EVER had a bad CPU.  What kind of CPU is it, anyway?  This thread may have that, but it's too darn long to read through now.  :)

<-=+=->

Clutching at straws - have you checked your voltages at the motherboard?

System Details are at the link at the bottom of the question.

Where do you check the voltages (I'm presuming BIOS), and what am I checking for?


Intel Quad Core Q9550 @ 2.83GHz
Corsair DDR3 RAM - 4GB @ 1333MHz
3 HDD's - posted above at http:#21766382
XFX nForce n790i Ultra motherboard
XFX Geforce 9800 GX2 Graphics Card
Thermaltake Toughpower 1200w PSU
Antec 900 case
TV Tuner and Network PCI cards - worked fine in previous machines - no faults in current system - already tried removing them
1 IDE DVD Burner and 1 SATA DVD Burner
MS Entertainment Desktop 8000 Series Bluetooth KB and mouse
Logitech G25 Racing Wheel, Logitech Z-5500D speakers, Logitech 3D Extreme Joystick, MS 360 Controller (not used in this system yet), MS WMC Remote Control

Main Software:
Vista Ultimate x64 (unactivated while problems persist - don't want to activate 50 million times)
Office 2007 - recently installed
Sony Vegas Pro
Virtual CD v9
Nero 8
Some smaller apps that are closed most of the time

Installed Games:
Test Drive Unlimited
Race Driver GRID
Colin McRae DIRT
MS Flight Sim X: Acceleration

This particular install of Vista is somewhat corrupt - Picture viewer isn't working, etc - so my next thing will be to try and install Windows again. Since I'm not currently running 4GB of ram, I might try XP Pro x86 again and see if I can get it to crash or not.

Possibly related info:
When I first setup the PC, when I turned the switch for the PSU on, the system turned on automatically. After getting a new mobo back, it didn't turn on when the PSU switched was turned on. Due to the HDD Connection error previously discussed, I reset the CMOS and BIOS settings back to default  and the behaviour returned.

When the PC freezes, I usually either hit the reset button if I'll use it again, or hold the power button down for 8 seconds. If it's just before bed time I'll turn the PSU switch off, to turn the blue motherboard lights off. I've never used the PSU switch while the PC is on (frozen or not).

Attaching DXDiag log file
DxDiag.txt

Checking voltage:
You use a digital multimeter (not analogue they can fry electronics) with the system running and the case open.
"Back probe" the the connections on the power supply plug at the motherboard.

See here:
http://www.lockergnome.com/it/2005/03/02/learn-how-to-diagnose-power-supply-problems/

AFAIK tolerances on an ATX board should be +/- 5% except for 3.3V, which is 4%. I read somewhere that onboard voltage measurements shown in BIOS are inaccurate.

Earth black
3.3V  orange
5V     red & purple
12V   yellow
-12V  blue
-5V     white
PS on  green
Pwr OK  gray

From Scott Mueller's "Upgrading & Reparing PCs":
+/-3.3V    tight tolerance    3.135 - 3.465
+5V          tight tolerance    4.75 - 5.25
+12V        tight tolerance    11.4 - 12.6

If you need elaboration let me know and I'll send a plagiarized picture.  :o)

I'll see if I can pinch/borrow the GFX card out of myold PC tomorrow from my younger brother's PC (which may be hard since I just got Grid and installed it on there last night)  :)

If it came down to it, I guess I could borrow the E6600 cpu from the old PC too. I could put my new CPU in the older PC as a test, but I couldn't put the 9800 in the old PC due to the 450w PSU with no extra plugs.

I'll also be wiping the Vista SP1 OS tonight and will install XP.

3 different sticks of DDR Three :o) RAM have all produced the problem - there is no recovery at all from the freeze

Both Auxilary power connections to the GFX card are made - both plugs light up to proove this.

I haven't upgraded the BIOS on this 2nd motherboard as of yet - but may do it after installing XP (assuming the 74mb file I D/L'ed last time works on XP as well.

One of the 4 pins for the default CPU fan were busted - possibly when dismantling the PC when replacing the mobo, the other 3 are all still in securely. The CPU is very close to the large exhaust fan at the top of the Antec 900 and the other exhaust fan at the back. Both exhaust fans are set to High, the two fans at the front are at Medium.

I've never mucked around with ram voltages and latencies before, nor overclocking, but I'll see if I can find what you're referring to in the BIOS when I get my 2x2gb ddr3 ram back.

It would somewhat funny if my PC was fixed tomorrow and I had to wait 2 months for my non-faulty ram to come back    :o)

The joys of building a new PC - I'm partially glad it didn't happen to a PC that I built for someone else, but still - grr.

Thanks for posting, MegaGenius - I mean Gary     :o)

I was about to suggest seeing if we could get Gary to have a look.
I should have, it may have been the correct answer.

:o)

I wouldn't bother swapping the CPU ... at least not until you've tried a different graphics card.   The one constant in all this has been your 9800GX2 card ... if a different card works perfectly, that's almost certainly the issue.   You've tried other RAM, a 2nd motherboard, etc. => power is always a possibility, but as I noted above, seems unlikely.   It will be interesting to hear the results of using a different graphics card [Although I said I'd install XP, it really doesn't matter whether it's XP or Vista -- I suggested XP because of the know issues in Vista with some graphics drivers => I have them myself from time-to-time].

Will swap the gfx card in about 4-5 hours and let you all know. Nearly finished work    :o)

I've tried putting the my older ASUS 8800 GTS 320mb card in, allowed Vista to install the device and restart - and didn't notice any problems. I was able to render a movie file in Sony Vegas, and play Grid for a lot longer than normal (although after putting my card back in I was able to render a movie there as well - somewhat a fluke I guess).

So I'm thinking 70% / hoping 30% that getting a new GFX card along with reinstalling Windows (which I haven't done yet) will fix things. I'll probably try and get it done during the same period of time that I don't have my RAM, so I'll be relying on the work laptop again for a little while.

Fingers crossed    :o)

... sounds like progress.   Do I correctly infer from yor comment that there were NO lockups with the 8800 GTS installed?  

There was one in Grid, but I'm under the impression that it was a software fault, and that the 1.1 patch is supposed to fix such issues. Apart from that everything else was fine.

Wiped Vista off the face of the earth last night and put XP on + drivers. Last night's session in XP was short, but didn't produce any problems as of yet. Audio drivers installed fine, have to finish installing other devices later on and I will miss SP3 just for the moment while testing.

I'll keep you informed

I also updated the new BIOS firmware last night from Feb 08 to April 08. May/may not be the exact latest version - but I don't have unlimited downloads.

XFX website is down at the moment, so I can't confirm whether or not that's the latest version ... but I'd think it's plenty close enough.   When you decide to install SP3, remember what I noted in my earlier post ... do ALL the updates EXCEPT SP3; then install it from the full installation download rather than Windows Update.   Should work fine with that sequence.

Are you running with the 8800 GTS or the 9800 GX2 ??

GX2

I can't keep my brothers card forever  :o)

... and the system is running nice and stable?   That's good !! :-)

Tests are inconclusive - I didn't make any special use of the graphics card in XP yet, nor have I used it for a prolonged period of time - only about one hour. Previously in Vista I could last longer than that before the system froze.

It will be interesting to see if XP is stable.   When the system freezes, have you seen any messages r.e. the "... Display Driver Stopped Responding and Has Recovered ..." ?  (I forget the exact wording)

That's a fairly well known error with Vista ... and sometimes results in a freeze.   I get the error perhaps once a week lately, but (knock on wood) it hasn't frozen on me since the last Catalyst driver update [Before that it would often freeze completely after that message ... requiring a hard reboot]   NEVER happens in XP with the exact same hardware.   It's very frustrating ... enough so I had decided to switch from my ATI 2600XT to an nVidia card => but in researching this issue I found that the same thing happens with nVidia cards !!   As I noted, the last Catalyst update a few weeks ago SEEMS to have essentially resolved it -- I've only seen the message 2 or 3 times; and the system has never frozen since that update.

I know of that error - and no - I don't get it (was associated with older drivers). The PC becomes completely frozen - like in the days of Win98.

I won't be able to test XP further until tonight, after I get home from work.

Ok - I've had two freezes tonight in XP while playing TDU. The first occurrence was quite weird, though, since the person I was playing online with had the exact same type of freeze at the same time. 2nd time I was on my own in the game.

I'll arrange for the graphics card to be replace at some stage next week, since I still have a few weeks before my RAM comes back, and I'd like to finish something off on this PC before then.

Good idea ... since you can't just "borrow" your brother's card indefinitely :-)

So, at what point is it actually considered "stealing" the video card?   :-)

<-=+=->

On quite a few occasions now, the PC has refused to boot beyond the following BIOS message:

Verifying DMI Pool Data

This also occurred when I had the message that there was no IDE conductor cable installed.
The first time this happened today I hit the reset switch with no success, followed by powering down and turning the PC back on again - which did work.

That's not the case right now, as I'm unable to progress beyond it presently

Can you "pinch" your brother's video card again and try that??

... I would also try a different power supply at this point => that's the one item that (if I remember correctly and didn't miss it on a quick re-scan of the thread) I don't think you've tried replacing here.   Clearly your PSU has PLENTY of capacity ... but that doesn't necessarily mean it's providing good, rock-solid stable power.   Wouldn't hurt at all to try another unit to eliminate it as the issue.   By the way, I'd think any good PSU of 600w or more is fine to try ... 1200w is WAY overkill :-)

I thought that error was based on CMOS or bad HDDs - but my brothers no here today, so I guess it wouldn't hurt...  :o)

Given the number of things you've eliminated by swapping parts ... and the fact this is the 2nd motherboard ... I'm beginning to suspect the PSU as the most likely culprit.   Unless you in fact have a defective 9800GX2 ... which is what using your brother's video card may help check.

I don't think it's the HDD - since I can't even boot from DVD when going into the boot menu   :o(

Getting a screwdriver

How big is your brother's PSU ??   :-)

450w generic (GigaTech) without support for the 9800

<quote>
So, at what point is it actually considered "stealing" the video card?   :-)
</quote>

I think it's when the owner forgets to whom he lent it.

That's the way my son and sons-in-law take ownership of my stuff.

Alan.

Ok - whats the plan of attack:

1: Swap PSUs around - see if my PC works well

2: Swap video card - see if my PC works

3: Swap PSUs around - see if my brothers PC goes bad

I'm trying no 3 atm - have put my PSU in my bro's PC and put it on my desk. I guess the only way to test this is to play Grid for a while and see if there are any crashes...

If you have any suggestions I'll see them come up on my laptop

IF the issue is the PSU, it's likely that it's one of the 12v buses, since it's not consistent.   If that's the case, it would be more likely to exhibit the same issues if you use the 9800GX2 with it.   You could also use the 450w in your PC with your brother's card, to see if that's stable.

Let me suggest again taking the components out of the case and running the system case-less.  I had very similar issues and the case was the culprit - shorting the mobo.

<-=+=->

Brother's PC with all his stuff but my PSU appears to be doing fine.

I'll try and put my 9800 in his PC and see if anything changes

Issue #1 - my GFX won't fit in the old PC - IDE slots get in the way

Issue #2 - the other PSU's motherboard cables aren't long anough to get to the power poerts of my mobo

Put the PSU back in my PC - left 8800GTS in there too

Number of attempts to turn the PC on failed - was on for one second - then turned itself off with no error-beeps.

I followed by removing all PCI cards and put it back together - run the power switch wire that ran via the TV-Tuner card directly into the motherboard, and put it back together.

The machine does turn on, but it's producing a warning message. I decided to write out the whole lot just in case something else jumps out at you. Note that I still am borrowing the 1GB of ram - don't have my 4GB back yet. Also for some reason (likely a loose cable) one of my HDD's isn't showing up - the 320GB SATA one (which incidentally is also the bootup HDD).

I'll fix the HDD up and turn the PC back on. If the warning message continues to appear, I'll load the default CMOS settings again (which I did earlier today when the PC wouldn't boot).

Phoenix - AwardBIOD v6.00PG
Copyright (C) 1984-2003, Phoenix Technologies, LTD
 
(6A61MB03) Release number 811N3P04
NVIDIA BIOS Version: 2.053.C3
Main Processor: Intel(R) Core(TM)2 Quad CPU 2.83GHz(333x8.5), 4 CPU(s)
Memory Testing: 1047552K OK
CPU Brand Name: Intel(R) Core(TM)2 Quad CPU Q9550 @ 2.83GHz
CIE BIOS Supported
EM64T CPU
 
Memory Clock is: 1067 MHz Tcl:7 Trcd:7 Trp:7 Tras:20 (1T Timing) 64bit
FSB Clock is: 1333MHz
IDE Channel 0 Master: ST3250620A 3.AAC
IDE Channel 0 Slave: PIONEER DVD-RW DVR-112D 1.24
SATA 1 (A0): None
SATA 2 (AI): ST31000340AS SD15
SATA 3 (BO): None
SATA 4 (B1): None
SATA 5 (C0): None
SATA 6 (C1): ASUS DRW-1814BLT 1.13
 
Warning! Now System is in Safe Mode.
Please reset CPU or Memory Frequency in the CMOS setup
 
Press F1 to continue, DEL to enter SETUP
04/18/2008-XFX-6A61MB03C-00

Select allOpen in new window

Another update:

Previously I mentioned something about when turning the PSU switch on, the PC automatically turned on - I'm pretty sure that had something to do with the TV-Tuner card - since it's not happening anymore

This time there was no warning message

This time the PC didn't freeze at the DMI Pool Data - however after it it said 'Update Success' (process was immediate - no waiting).

I used an old SATA cable to connect to the 320GB HDD which isn't working - so I'll go back to the original XFX one - not sure if there are any differences in new and older (ie 1.5 years old) ones, but I'll swap that back and hopefully I'll be in XP again.

Gary, are you suggesting that the previous issues with freezing at the DMI Pool Data were related to the 9800 gfx card? I'll continue to use the 8800 in this PC for a few hours before I'll have to put everything back to normal before my brother gets home.

Swapped the SATA cables between the 320 and 1TB drives, and now both work fine - and XP is finally back up and running again - only took 6 hours  :o)

I'll use the PC as is for the next hour, then put the 9800 back in.

PC froze with the 8800 - so maybe its not the GFX card.

This is annoying me soooo much   :o(

But it was stable with the different PSU?  Did you try going case-less?

<-=+=->

I never had the other PSU in my case - it's cables that go into the motherboard weren't long enough.

So I'm left with a PSU or CPU problem perhaps? I might as well take both parts back at the same time and check it out.

I don't really like the idea of going caseless - I've quite sure that nothing is touching the back of the mobo, it's just the screws to the case.

This is indeed a very strange (and I'm sure frustrating) set of problems.   I checked a couple of reviews of your motherboard ... and no only are there no particular issues noted; but it gets VERY high praise for its features and stability.   And since you've had two of them (with identical symptoms) that seems very unlikely to be the issue.

Memory is always a potential culprit ... but this too has been switched around enough to essentially eliminate it.

I didn't find a convenient blow-up of the motherboard, so I'll ask you:  Are there any auxiliary PCIe power connectors on the motherboard?   If so, are each of these and the auxilliary power connection(s) on the graphics card plugged into a different lead from the PSU??

The loss of CMOS configuration is puzzling ... but this could also be an indication of some strange power issue => I'd agree that unless it's a power routing issue (feeding too much power from the same PSU "feed"), the next thing I'd try is a different power supply.

It's possible you have a bad CPU ... but that is (as you've noted) very rare => and I'd expect the failure to be much more consistent and predictable.   With an unstable power rail the power could be fine until the power supply hits some thermal condition (which would be somewhat random in timing); and could then cause a variety of issues ... including freezing the system.

I was reading more about this board ... and was in fact going to ask you if you had the chipset fan working; but you pre-empted that question.   Normally I'd think the motherboard was at fault with the issues you're having;  but the consistency across two different boards ... and the VERY positive reviews of this board ... tend to indicate it's something else.   Problem is, there's not much left to check !!   I'd check the PSU next ... preferably with a different PSU with a different rating (anything above 700w is plenty (even 600 shoud be fine) ... the reason for a different rating is so you'll be operating at a different point on the output curve of the PSU.

... by the way, I looked your picture over carefully and I also don't see any PCIe auxilliary power inputs on the motherboard.

Edited version of the above picture showing my setup and what I have made use of
mobo.jpg

I'm not sure if the store will allow me to change the part over unfortunately - it would have to be something from www.cpl.net.au - there was a Seventeam PSU that I was comparing to the Thermaltake when watching video reviews at youtube by 3DGamerman.

I've updated the BIOS again to the latest available (P05). The BIOS version appears to be reported the same as before in the nVidia CP, but the date has gone from 18/4 to 16/5/08

So we'll see if this magically does anything

btw - you say to swap the PSU - how come nothing adverse happened when I put it in the old PC?

I played GRID on it for over an hour, which would be enough to freeze my PC 2-3 times.

There's no guarantee it's the PSU ... but it's worth checking.   When it was in the old PC you were operating at a different point on its output curve, and almost certainly drawing much less power from the 12v bus.   So IF the issue is being caused by some weird power instability issue I'm not surprised you didn't see if on the other system.

Does the system freeze if it's running a simple stress test ... say Prime 95 in the "Stress Test" mode ??   If you haven't tried that, let it run for a couple of hours.

On a previous occasion I tried to fill up the 2GB of ram I was using in Vista x64 by opening a lot of large programs - was able to get up to %79 RAM usage before it went to the page file. Th freezing occurs nearly like clockwork - between 1 - 2 hours of use. You might be thinking of temperature right now, but if the PC does freeze, and I hit the reset button and continue usage I still get another 1-2 hours of use. Freezes normally occur in games and movie editing software, but that could be coincidence since I'm mainly using those programs a lot at the moment.

I'll try the stress test tonight - do you have a link?

http://files.extremeoverclocking.com/file.php?f=103

Download ... install ... run and select the torture test :-)

Running XP, doing nothing more than listening to music and installing a few items, I could hear a fan turning on/spinning faster for 1 sec then returning to normal. Sounded like the GFX card, so I checked outthe nVidia Monitor to see if it was getting hot.

Turns out that both GPUs are sitting at 110 degrees celcius - which in myh books is way too hot - especially since I'm not doing anything.

If you ask me, it's hot enough to cause a freeze, and enough to be considered faulty considering my setup. The only concern I have is that the PSU fan blows air up into the GFX card (how much or how hot I don't know).

I had the PC on standby for the last 2 hours before returning home, it was on for 30 mins before that, if that has anything to do with it.

I'll shut down for a bit, and come back and run that torture test.

Yes, 110°C is WAY too hot !!!   Period !!  Even if you were gaming your heart out with the most graphics intensive game you have, that's still too hot !!

... I think you've identified the cause of your freezes :-)

FYI, I was curious about what "reasonable" might be for such a powerful dual GPU card's temps.   According to this review, they saw idle temps in the mid-50's, and full load temps of 69 & 70°C
http://www.techpowerup.com/reviews/Zotac/GeForce_9800_GX2/25.html

I suspect your GPU's are simply shutting down from thermal overload (just like your CPU would if it got to that temperature !!).

... Note that since you also noted earlier that "... PC froze with the 8800 ..." this may be related to internal airflow in the case => since it's likely that the 8800 also overheated.

How is the PSU mounted that it blows air INTO the case ??

After shutting down I opened the side cover - and the GFX card was too hot to touch - the PSU on the other hand was as cold as anything. It has a grill and fan on it's top inside the case, plus one at the back.

I got my desk fan out and turned it on high, facing it to the card - which cooled down to normal temps in about 5 mins.

It's winter here - room temp is ~10 degrees celcius. The fans on the Antec 900 are actually making me cold since I'm not using the PC in such a way to heat it up.

The Antec 900 has 4 fans by default, plus allows for another 2 to be added (which I'm not doing). Two fans blow air into the case from the front (have them set to medium speed). Then there is a fan at the back at the top near the CPU that blows air outward - set to high. Then there's the massive 20cm fan at the top the blows air out the top of the case - set to high - and making me a tad more cold  :)

I reinstalled Vista x64 last night (dual boot) because of strange sound problems with XP - not the same problems as mentioned at the start of this thread. This time the sound works, but the Volume slider in Windows has no effect on the volume (SPDIF Digital Toslink). Fortunately my speakers have a remote so I could live with that. But FRAPS also isn't recording sound, and since I knew it did work in Vista - I'm setting that up again. So far there are no indications of overheating again. However, the Vista version of the nVidia CP doesn't seem to include a monitor for temperature, so if the PC does continue to freeze I'll touch the card to ensure if it was that hot again or not.

Latest drivers for everything are being employed in each OS.

Weird !!  The Antec 900 has superb airflow -- and unless it's different than every other PSU I've seen with two fans, the fan inside the case blows air FROM the case through the power supply ... NOT into the case.   So it's difficult to imagine what's causing the GPU's to get so HOT !!   ... especially in such a cold room !!

Is the fan on the graphics card working ??

In that case I'll agree with you that the PSU fan sucks air from the case and blows it out the back.

I would assume the GFX card fan works, since I was hearing it come on/off which prompted me to check the temp in the first place. With so many louder fans on the case it's hard to hear the GFX card one anyway - but I'm fairly sure it works.

The card hasn't got that hot again in the last 7 hours I've had the PC on - was playing a game for about 30 mins and no rise in temperature about 60 degrees. Then again, the PC hasn't frozen during that time either (which now that I think of it is a bit of a record). Interesting  :o)

I'm running the stress test now. The PC is handling it very well. Despite the CPU showing a 100% load, I've been able to continue installing printer drivers and Nero 8 Ultra and work on a document without it being any slower than a normal P4 system.

How long does the test run for though?

... as long as you let it run :-)    I'd let it go for at least a couple hours ... if all is well then, it would most likely run fine forever.

Sounds like your problem has been isolated to an overheating graphics card ... the only remaining question is WHY.   To get as hot as it did, I'd suspect a fan that's not working properly --> next time it hangs notice if the fan on the graphics card is spinning before you shut off the power.

Ah

Well I was away for a fair bit and returned to find the PC frozen with the screen off (since the PC was idle the monitor would've turned off).

For however long it was frozen the case could've cooled it down - I checked and it was only warm to a normal standard.

I've also noted that the DMI Pool hang problem is related to when I have a PCI card in a certain slot. It never used to be an issue, just a recent happening.

The 110°C temp for the GPUs is still the most telling indication of what is probably causing the system to freeze.   The remaining question is WHY !!   Could be a bad graphics card (not likely the GPUs, but possibly a defective fan or fan controller);  might be poor or obstructed airflow (VERY unlikely, as the Antec 900 has one of the best airflow designs you can get !!);  of perhaps you've got something set wrong in some system monitoring software (do you have any hardware control software that allows custom fan speed controls??).

Running out of ideas here :-)    ... if you hadn't already swapped the motherboard, I'd be suggesting you do so by now;  but it simply seems VERY unlikely you've got a bad board, considering the quality of that board and the fact you've had two of them with the same symptoms.

I was curious about how hot my GPU gets (a fanless Radeon 2600XT) ... so I ran rthdribl for 45 minutes.   It stressed the GPU between 90 & 99% the whole time => here's a GPU-Z screen capture I did at the end [53°C with 98% GPU load]:


GPUZ-after-45-Minutes.jpg

... clearly that's not in the same league as a 9800GX2;  but the higher end graphics card should have a higher-end cooler (certainly not fanless !!) that should still keep the temperatures much more reasonable.   Are you watching the GPU temps while you use the system??   GPU-Z is a reasonable little utility to do that [http://www.techpowerup.com/downloads/1104/TechPowerUp_GPU-Z_v0.2.4.html ]

One other thought:  I don't recall discussing this earlier (but this IS a long thread ... so if you've already tried it forgive the duplication) ... if you haven't already done so, go into the system BIOS and set "failsafe defaults" for the performance parameters.   The wording may be different, but most modern BIOS's have something along those lines.   To be TOTALLY sure you've using the conservate settings, a full reset would be even better ==>  unplug the system; remove the CMOS battery;  let the system set overnight (or at least for a couple hours);  then replace the battery; plug the system back in; boot to the BIOS; and set the conservative defaults [often called "failsafe defaults"].   Considering the performance-oriented nature of this board, it's very unlikely that it defaults to conservative settings ... so this MAY have some impact on the reliability you're seeing.  [I'm still bothered by the 110°C temps in the GPU, however ... which it's unlikely a full CMOS reset and conservative BIOS settings will have any impact on.]

Well I've been sitting the deskfan blowing air directly at my GFX card for the past day - have still had 1 or 2 freezes though (card felt normal temp) so saying it's the problem is still only a possibility. I'll have it swapped regardless along with the PSU just in case.

I did a BIOS reset with the old mobo by using the jumper. Numerous times have I loaded the default BIOS settings (for problems relating to bootup).

I was also thinking about getting GPU-Z since I thought it might monitor the temperature (since I can't do that in Vista x64).

Just opened up the app - and low and behold - look at the fan speed indicator:
GPU-Z.jpg

Since the bottom drop down box has two listings (inbuilt SLI) I changed to the other one - and the speeds are shown

Now what?
9800.gif

It's interesting that GPU-Z doesn't show the GPU loading % => it does on all of my systems (may be a function of the "newness" of the 9800GX2's).   The fan speed is very low ... but is probably just because the GPU isn't being heavily utilized.   What happens to the fan speed and temps if you stress the GPU's a bit?   [rthdribl is an easy way to do that:  http://www.daionet.gr.jp/~masa/rthdribl/ ]

I've never seen the Fan speed go over 50% even when just after quitting a game or restarting from a freeze.

Is there one fan per core (2) - or just one fan for the whole card?

My understanding is there's only a single fan on the 9800GX2 (a quick visual inspection should confirm that).   Did you run rthdribl for a while to see what the temps look like?  (20 minutes should be plenty)

Not yet - I'm at work

I took both the GFX card and the PSU back today, so I'm without the PC for another 2 weeks.

I was just catching up with my 3DGameMan subscription and found this, my only interest being that he claims the n790i is 'buggy'

http://www.youtube.com/watch?v=Lro2CxIF4tA&feature=user

Who sells a product when it doesn't work? Especially with that price tag.

Interesting comment r.e. the 790 => certainly conflicts with the reviews I read of your board that really praised its reliability and stability !!   Just shows that different tests can have different results !!

... guess this thread is on hold for two weeks until you get the replacement card & PSU :-)

That's right - back to using the work laptop again

The tech support guys at XFXForce.com say that the motherboard used to be buggy, but from BIOS release P05 and newer these have pretty much all been sorted out

Thought you'd like to know

Parts should return in the next few days. Who wants to place bets as whether or not my PC will work then?  :o)

Well ... the most recent BIOS shown in your screen shot earlier in this thread was P04 => so hopefully you'll find that the P05 BIOS does indeed resolve things !!   No bets, however :-)

... but the first thing I'd do is check the BIOS revision on the board you get back; and update if necessary.   It's actually unlikely that the board will already have the latest BIOS when you get it.

That is an interesting bit of information that I may have overlooked.

^^ Sorry - i can't ignore new features that get added here  :o)

I rang the warranty guys again today and they said I should be able to get the parts in the next 24 hours or so.

They also said that they found no problems with the PSU (that's another $15 out of my pocket) and that they replaced the GFX card. Whether they found problems with it or they just swapped it because I said it got to 110º C hot I don't know.

Of course, it has to arrive in the city on my day off from work when I'm nowhere near it.  :(
When I do eventually rebuild it I'll make sure I have the most up to date BIOS as well. I should probably re-download all the newest drivers again today while I have the university's Internet connection available to me  :o)

I should get my parts back tomorrow - later than I initially expected.

Due to feeling a little giutly of my current stock CPU fan having one of the 4 pins slightly busted and not as good as it could be - I ended up getting a 3rd party cooler - the Thermaltake Max Orb - which yet another device that lights up blue  :o)

Despite the Antec 900 being a very good case for airflow - it does get outdone by this one: http://www.peteredge.orconhosting.net.nz/casefan.html

Outdone?  Maybe just a bit :-)    But I'd take an Antec 900 anyday over that "contraption" --> your airflow is PLENTY for any system you'll ever put in it.   Besides, the guy didn't consider the laws of thermodynamics when he built the case -- you don't want air flowing INTO the case from the top !!

I've been using XP all night without a hitch - games, movie rendering (saving), watching HD TV, etc. A software crash in a certain game prompted me to try it in Vista - where the whole PC forze about 2 mins into the game.

What do you think? If not for that event in Vista I'd be cautious to say it was fixed - but I should know more or less for sure by the end of the weekend.

Nope - it froze again in XP - problem still unsolved.

So far we have:

Replaced the motherboard
Replaced the RAM
Remove expansion cards
Checked the PSU - no faults
Replaced the GFX card
Upgraded the BIOS
Tried various drivers and operating systems
Added more case fans
Swapped CPU cooler
Pulled hair out

So now that leaves the CPU as being the only part left unchecked. The day I bought it was a thunderstorm if that has any relevance. If I had a number of tubes of thermal paste I could put my CPU in the older PC and the Core2 into this new PC and see what happens.

Attaching a report from Everest for you to consider while I get some sleep.

Change the extension to .mht


Memo to self - don't upload files to EE - it's very slow
Report.txt

Well, the CPU is, as you've noted, the only thing left to try !!

I'm also perplexed by the CPU info reported by Everest ... it doesn't seem right for that CPU.   Just to be sure I'm not misinterpreting it, run CPU-Z and post what it shows on the CPU tab (the initial display).
http://www.cpuid.com/cpuz.php

Everest said at the start it wasn't up to date and that it wouldn't give accurate CPU info - I'll get CPU-Z when I return hom

While I was out I thought it might be a good idea to go into the BIOS and disable the 3 extra cores of the CPU and see how long I could use the PC before it froze. So I'll see how much I can get done on what's now virtually just a P4 system  :)

If I can't make it freeze I might put another core again but I won't go to the same extent as a memory test - since you can't just take one core back to get replaced  :)

After all that I'll decide if I should swap CPUs with my brother (don't have any thermal paste though) or I'll run CPU-Z when I have the 4 cores turned on again.

Interesting if it runs reliably with only one core !!    Post your CPU-Z results when you get a chance.

Here's the CPUZ report - although it's only showing me running the 1 core
http://ryzza007.googlepages.com/cpuz.htm

Clock & multiplier look good => unlike the Everest report.   Is the system stable with just the one core?
If so, this is getting weirder as you go along !!  Beginning to seem like either a defective CPU ... or an issue with the board actually supporting the Q9550 !!   Wonder what it would do with an E6850 or E8400 :-)  [or a Q6600 -- older generation quad core]   ... wouldn't happen to have access to one of those, would you?

By the way, were you VERY "static sensitive" when you were handling the CPU to install it?  [anti-static wrist strap;  anti-static mat for the motherboard; etc.]

Er - just touch the metal case first, don't shuffle feet, don't touch the underside of the CPU, that sort of thing - it's never failed me before.

PC seems very reliable (admittedly a little slow) with just the one core. I haven't been able to freeze it yet.

I'm a bit more paranoid about static precautions, but as long as you're properly grounded that should be fine.   Just for reference, it's actually best to plug the power supply in to the wall, but turn it off with the switch on the PSU (so there's no standby power).   That way the metal case is well grounded (through the power cord's ground) ... and when you touch it you will be as well.   I also wear an anti-static strap when working on the motherboard (especially when installing a CPU) ... but agree that it's not an absolute necessity if you're careful enough.

With any luck, this whole issue is a compatibility issue with the board and the Q9550 => if so, the next BIOS update may very well make the issue go away :-)  

The two have been compatible for ages - there's a chart on the XFX website.
http://www.xfxsupportb.co.uk/XFX_CPU_GUIDE.pdf

I'm under the impression that the issue is definitely the CPU - no faults at all with the one core. On the next restart I might enable Core #2 as well.

I thought they had indicated there were issues until the last BIOS release [per your note at https://www.experts-exchange.com/questions/23393529/PC-Freezing.html?cid=238#21944530 ] ... but I agree the issue is looking a lot like the CPU.

Paul's words below:

[PAULV 7/7/2008 12:59:23 PM] Hi, thanks for your message and for letting us know. It is true that there were bugs with the 790i mainboards, but since BIOS version P05 and our new release P06 all seems to be ok now. Ideally if this was pre-built then it should have been updated prior to despatching to you, but if not then you will need to do all this yourself, note that all our updates are available from our downloads section or from www.nvidia.com. Hope all goes well, however if any queries or problems please keep us posted. Kind regards Paul

Well ... sure seems like any motherboard issues should have been resolved by the version 6 BIOS you're using.

.... so it's looking more and more like a CPU issue.  You're replaced virtually everything else !!!

By the way, I noticed in your Everest report you're only using one memory module.   Just to confirm that everything's REALLY okay now (with just one CPU core), I'd install two modules (your two Corsair modules so you'll have 4GB)

I'm using the PC with no real noticeable problems on 2 cores atm - and will continue for a week or two to complete some more work - then will take the CPU back for warranty

In other related nVidia news:
http://news.cnet.com/8301-17938_105-9989041-1.html?hhTest=1&part=rss&tag=feed&subj=Crave

I'll also ask a 'related question' to this in order to double the points that you guys can get just as a thankyou.

Here you go guys - I'll close it when you all post there:

http:Q_23561530.html

Extra points not needed -- and "point passing" is against the EE rules.   You should delete the other question before anyone posts.

I see you're right there - I'd just seen it done before a number of times and was possibly told to let it go when starting up as a CV (can't rememebr for sure now) - maybe it's stricter now with the multi-zone posting confusion.  :)

Anyway - the thought is still there.


I'll keep you updated.

This is the 172nd post in this thread :-)  :-)
... I guess it is getting a bit long !!   ... but that's not a problem with me => the goal is to get the issue resolved; and it does seem like you're getting there !!  

I assume Quick_EE gives you that info - or were you that bored to count?

make it 173

I counted :-)   [Sometimes I get curious]  ... now it's 174 :-)

I don't use (never have) Quick_EE or any of the various "aids" that have been available ... I simply respond to those questions I choose to; and then reply when I get an e-mail.   I did, however, just note that if you look at the unanswered questions in the primary zone (PC's in this case) it shows how many posts have been made for each of the still-open questions.   ... and apparently I mis-counted :-)   [It shows 173 -- so THIS post will be #174 ... not my last one :-) ]

You definatly are too bored mate  :)

For all you know the page you looked at hasn't updated itself yet either, in which case now I imagine you'll be trying pretty hard to resist the temptation to recount them again yourself. I used Quickpost a bit and QuickEE once but never did see the need for it.

Sounds as though you're on to it Ryan, but shouldn't you leave this open until you're sure?
Might break some kind of record.
Get a free T shirt.

:o)

I thought this was nothing compared to some of the Q's asked in P&R or the lounge, but there's always room for one more T-Shirt - it's good how they change the design frmo time to time (got one of those new black shirts not too long ago) - and the EE coffee mug is great as well.

Coffee mug??   Now I'm jealous -- all I've got are T-shirts !!   (but plenty of them)

Going through my open questions and noticed this is still open -- any updates??

Um, no

I haven't taken the CPU back since I need it currently and can use it with 2 cores disabled

I will return in it in the next week or so

I will award point today though for you - I was actually planning to do it the other day  :)

Actually there is one update in regard to the GPU reaching 110 degrees celcius

It happened again on the new card - noteably after resuming from hibernation (which I always used to do but hardly do now due to fast bootup speed). I think it is a good possibility that the last time the temp went sky-high I hibernated with XP - so that could be the cause.

I'll have lunch and then read through all these comments again to figure out which ones to accept.  :o)

Ok, not the biggest problem in the world at the moment, but still puzzling and a little annoying:

I got my RAM back today - all 4GBs of it. The sticker on both modules reports DDR3 2GB 1333MHz as expected, but my computer reports that it's only 1066MHz in the BIOS and in Everest 2007.

I know for a fact that it used to show 1333MHz before I sent it away, with an older BIOS version and with the RAM in different slots. I currently have the RAM in slots 1 and 3 (instead of 0 and 2) since the CPU cooler doesn't allow me to place this large module in slot 0 like I was able to do with the much smaller 1GB module earlier.

Any ideas?

Screenie
Untitled.jpg

Post what CPU-Z shows for the SPD for each module (they should be the same) and also what it shows on the Memory tab and on the CPU tab.   On the SPD tab, you have to select each module in the "Memory Slot Selection" drop-down box.

I suspect the FSB:DRAM ratio (shown on the Memory tab) will show what's going on here ... but would like to see everything that's being reported.

I presume you already have CPU-Z ... but if not, it's a free download here:
http://www.cpuid.com/cpuz.php

Here's the info


I'm trying to work out why my modules are being shown in slots 3 and 4 instead of 2 and 4...
Untitled.png

Okay, I don't know for sure what's going on here, but here's an educated guess:

Look at the SPD info:  At a 533MHz clock, the timings are 7-7-7-20.   At the slightly higher clock speeds these become 8-9-9-23 and 9-10-10-26.   If you do the math, you'll see that while the CAS latency is a very tiny bit better at the higher speeds, the total access times are actually best at 533MHz.   So I suspect the SPD default is 533MHz => so that's what the system sets the memory at.   You would potentially get a slight improvement in burst speeds (multiple access from the same column), but basically the memory's performing just fine just as it is.   For example, the CAS latency (1st number) is 13.13ns with 7 clocks at 533MHz;  it's 13.12ns with 9 clocks at 686MHz.   But the OTHER latencies are actually longer at the higher clocks speeds ... they're still 7 clocks at 533MHz, but are now 10 clocks ... or 14.58ns -- 1.45ns longer than at the 533MHz setting.   I suspect the SPD is simply set to default at the best performance for random accesses.

I'll leave it if you say so then - I wonder what changed since I last had the ram on this same mobo - as it was 1333MHz with timings of 7, 7, 7, 21

Thanks

Ok I've got the CPU back, formatted both OS partitions and installed XP and setup most apps on it (and plan to never use it again unless I can't do something on Vista). Today I installed Vista, took out 2 GB of ram to finish it off, then installed SP1 x64. Half way through (at some stage - I wasn't watching) it BSOD'ed and continues to do so - 0x000007e (# of 0's questionable) and Safe mode fails at crcdisk.sys (which I think happened to me before so I'll search this thread for it using Ctrl+F). Nearly got the PC working now - only 4 months after asking this question.

Well I found it above and I'm guessing I have to take out 2GB of RAM again to finish installing SP1 - will try now

Wow .... what a frustrating system !!   Since the last activity on this thread, I've found a few interesting posts r.e. issues with Q9550 CPU's ... apparently these are NOT common with the slightly slower Q9450s.   Not at all sure if this is relevant to your problems, if you happen to have access to a Q9450 I'd give it a try and see if the problem persists.   I built a Q9450 system for a friend a couple weeks ago (on a Gigabyte P45 based board) and it works perfectly.   Used the Q9450 because of the issues I'd read about with the 9550.

Ah so it's not just me then, makes me feel just a tiny bit better :o)

Removing the ram fixed it - I hope I don't have to do it for every update I install - although Windows Updates always fails when checking anyway - never works whether it checks the net or if I tell it to look as my WSUS server. I guess this is case closed then now, assuming it never freezes repeatedly again.

All that said, I did a test to render a video I made in SVP8 and it is unbelievably fast - compared to when I only had one core running with only 1GB of ram.

"... Removing the RAM fixed it ..." ==> Hmmm.   Removing one module will result in the system operating in single channel mode.   This makes it seem very likely that the issue is memory bandwidth.   Have you tried clocking the RAM slower ... perhaps 800Mhz or 1066Mhz ??   You should NOT have to remove a module to do any updates !!

Don't know - but the PC froze in Vista again

I'm downloading AVG 8 and will scan for infections

PC is still freezing

When I got the CPU back, I noticed it had something on one of the contacts - I unsuccessfully tried to blow it off, and decided not to wipe it off. When I first put it in, the PC didn't boot and had no beeps, so I took the CPU, RAM, and all power connectors out of the mobo and took out the CMOS battery and reset the BIOS with the jumper pins and waited 30 mins before putting it back together where it worked again, bringing me to where I am now. The PC has frozen twice in Vista now but not in XP yet.

I think at this point I'd cry "uncle" -- and trash that state-of-the-art motherboard for a P45 or X48 based board !!   Since you already have DDR3 memory, something like a Gigabyte GA-EP45T-DS3R would be a good choice.   There's no guarantee that would resolve this ... but based on everything you've swapped out, I think there's an excellent chance it would do the trick.   It's POSSIBLE ... but relatively unlikely ... that swapping the Q9550 for a Q9450 would help => but I think a more stable motherboard is a much more likely solution.

You have GOT to be frustrated with this !!!

i should have an antec 900 shaped hole in my window you mean - but window's are expensive, lol

I'd like to sell it to be honest and start again, but a) I'd lose a fair bit of money and b) i would hate if it froze for that person too

What would you think of the idea of handing it over to a PC repair shop and telling them to fix it - frankly I doubt they could do anything and would still charge me $100

I seriously doubt they'd find the specific issue here -- you've done a very thorough job of troubleshooting.   This could be (a) an unstable motherboard;  (b) a compatibility issue with the motherboard/DDR3;  (c) a compatibility issue with the graphics card;  or (d)  a power issue.

If I recall correctly, you've pretty much eliminated all of these by swapping parts (I don't recall if you actually swapped the power supply -- and a quick scan of this very-long thread didn't find reference).

One question r.e. its current performance:   Does it also freeze in XP ??

Every part has been replaced under warranty (motherboard, ram, gfx card, psu, and now cpu).

It used to freeze in XP - although not as much as Vista x64. It hasn't frozen yet, but I wouldn't be too surprised if that changes.

Actually the PSU wasn't replaced as there were no faults found with it

At this point I'd do one of two things [actually I'd probably do (a) -- and if that didn't make things stable, then do (b) ]:

(a)  Run XP and see if it's stable there.   If so, it's likely a driver issue that will -- with time -- be resolved;  so in a few months you can update the BIOS; get the latest Vista drivers; and likely run stable in Vista as well.

or

(b)  Swap the motherboard for one that uses an Intel chipset (they tend to be the most stable).   I have both P35 and P45 based systems -- and they've been rock solid.   I built a P45 based Q9450 system for a friend that's also been exceptionally stable --> he games a bit (8800GT) and runs Adobe products extensively (Premier and Photoshop).   This would obviously cost $$ => but probably no more than you'd pay a shop to tell you they didn't find any problems :-)  [The biggest hassle would be reloading all your software]

... speaking of updating the BIOS => they have a newer BIOS (P06) for your board.   You may want to give that a try before doing anything else.

Got P06 a while ago before sending back the CPU

Well ... then it's back to steps (a) and (b) above :-)

PC froze in XP

I'll try and sell the PC before going for any new hardware - I don't need 2 mainboards

That's better than creating a hole in your window with the Antec 900 :-)

... Considering how much you'll likely lose in a sale, you may want to think about that.   You can always sell the mainboard on e-bay ... and it's VERY likely that with a good P45-based motherboard all will be well.   On the other hand, I can certainly understand your frustration ...  you may just never want to see "that" system again -- no matter WHAT mainboard is inside it !!   (... in that case selling it is a far better option than a hole in your window !!)

It's too bad you're using DDR3 ... the mainboard options are somewhat limited by that => but this is an excellent board if you should decide to give it a try (I've used the DDR2 version of it for a couple machines):  http://www.gigabyte.com.tw/Products/Motherboard/Products_Overview.aspx?ProductID=2851

Yeah it does look alright, although you'd be pressed for room if you went with 2 GFX cards in crossfire (which would prompt me to get the newer ATI HD 4870 X2 card that currently at the top of the market last time I checked. Thing I hate most about ASUS/Gigabyte mobo's are their stupid model names. How much easier is it to remember n790i Ultra than GA-EP45T-DS3R (rev. 1.0)  :)

Of course someone has to buy my PC first and hopefully it works fine for them with their setup.

I've had a look here (http://www.centrecom.com.au/catalog/default.php?page=1&cPath=36_62) and have seen 3 different DDR3 Gigabyte mainboards, and apart from the number of SATA ports and possibly the speed of the PCI-E ports and the heatsink I can't see a lot of differences.

On the subject of SATA ports, the person who designed that obviously didn't try to connect one in a hard to reach/see location in a PC - it's harder than IDE just about.

One is the economical P45-based board I suggested. Another is the "extreme" version of that board ... which has more support for overclocking and supports a 1600MHz FSB. The last is the top-of-the-line X48 based board ... an excellent board with Intel's high-end chipset.

On both of the P45-based boards, the 2nd PCIe x16 slot runs at x8 (although it's physically an x16 slot) ... and in Crossfire mode both will run at x8. On the X48 board, these slots both run at a full x16.

There are a few other minor differences (as you noted -- different # of SATA slots, etc.).

Thanks for the tip

Mainboard     390.50     http://www.centrecom.com.au/catalog/product_info.php?cPath=36_62&products_id=30609
RAM                266.20     http://www.centrecom.com.au/catalog/product_info.php?cPath=115_23&products_id=31967
GFX Card      716.10      http://www.centrecom.com.au/catalog/product_info.php?cPath=26&products_id=34060
CPU               690.80      http://www.centrecom.com.au/catalog/product_info.php?cPath=27_28&products_id=34161
Case             167.00      http://www.centrecom.com.au/catalog/product_info.php?cPath=125_126&products_id=26837
PSU               293.70      http://www.centrecom.com.au/catalog/product_info.php?cPath=74&products_id=25527
Total             2524.30
Currently I have my PC listed for $2500 ONO and I'd honestly be surprised if anyone went for it. I'd spend a few extra hunder $ to go brand new.
While I could reduce that total by going to other stores, I half-considering paying an extra $75 for them to build it so that if I do have problems, the ownus is on them to fix it. My current PC has all it's part purchased from one store at various transactions except for the RAM which they didn't sell.

NICE system :-)

... and I'm sure it will be MUCH more stable than what you've got now. Good luck with your listing. ... of course if you don't get any "bites", you should be able to sell your mainboard for ~ $200 => so you could replace just the mainboard and still have a very nice system.

No bites so far, but I have been playing around in the BIOS
In the CPU options page - I retried the trick of disabling the 3 extra cores. Before getting the new CPU, this improved the amount of time I could use the CPU before it crashed, but yesterday it had no effect. I went back into the same page and changed the settings for 3 options above where you can disable the cores. I think I changed them from Disabled to Enabled, but it could be the other way around. I also re-enabled all the cores. After saving the options, instead of just rebooting the system completely turned off straight away and then turned back on - most likely to apply one or more of the changes. (I'm at work now but you could check the n790i manual online to see what I actually changed).
This did seem to make a notable difference. Whilst using Vista x64 I was able to perform a number of various activities for over an hour without it freezing. It was only when I launched a game that it froze 15 mins later. I'll have to use the OS again tonight to see if it does freeze with no games running though.
 
PS - Congrats on your Zone Advisor promotion Gary
 

HJ This log as shown in Vista x64:

Logfile of Trend Micro HijackThis v2.0.0 (BETA)
Scan saved at 7:02:24 PM, on 18/09/2008
Platform: Windows Vista SP1 (WinNT 6.00.1905)
Boot mode: Normal
 
Running processes:
C:\Program Files (x86)\Common Files\VideoMate\ComproRemote.exe
C:\Program Files (x86)\Common Files\VideoMate\ComproSchedulerDTV.exe
C:\Program Files (x86)\Digsby\digsby.exe
C:\Program Files (x86)\Xfire\xfire.exe
C:\Program Files (x86)\NaturalPoint\TrackIR4\TrackIR.exe
C:\Program Files (x86)\Internet Explorer\iexplore.exe
C:\Program Files (x86)\IEPro\MiniDM.exe
F:\Software Archive\System Tools\GPU-Z.0.2.4.exe
F:\Software Archive\System Tools\SIW\siw.exe
F:\Software Archive\System Tools\HiJackThis_v2\HiJackThis_v2.exe
 
R1 - HKCU\Software\Microsoft\Internet Explorer\Main,Search Page = http://go.microsoft.com/fwlink/?LinkId=54896
R0 - HKCU\Software\Microsoft\Internet Explorer\Main,Start Page = http://www.gmail.com/
R1 - HKLM\Software\Microsoft\Internet Explorer\Main,Default_Page_URL = http://go.microsoft.com/fwlink/?LinkId=69157
R1 - HKLM\Software\Microsoft\Internet Explorer\Main,Default_Search_URL = http://go.microsoft.com/fwlink/?LinkId=54896
R1 - HKLM\Software\Microsoft\Internet Explorer\Main,Search Page = http://go.microsoft.com/fwlink/?LinkId=54896
R0 - HKLM\Software\Microsoft\Internet Explorer\Main,Start Page = http://go.microsoft.com/fwlink/?LinkId=69157
R0 - HKLM\Software\Microsoft\Internet Explorer\Search,SearchAssistant = 
R0 - HKLM\Software\Microsoft\Internet Explorer\Search,CustomizeSearch = 
R0 - HKCU\Software\Microsoft\Internet Explorer\Toolbar,LinksFolderName = 
F2 - REG:system.ini: UserInit=userinit.exe
O1 - Hosts: ::1 localhost
O2 - BHO: IE7Pro - {00011268-E188-40DF-A514-835FCD78B1BF} - C:\Program Files (x86)\IEPro\iepro.dll
O2 - BHO: Adobe PDF Reader Link Helper - {06849E9F-C8D7-4D59-B87D-784B7D6BE0B3} - C:\Program Files (x86)\Common Files\Adobe\Acrobat\ActiveX\AcroIEHelper.dll
O4 - HKLM\..\Run: [Adobe Reader Speed Launcher] "C:\Program Files (x86)\Adobe\Reader 8.0\Reader\Reader_sl.exe"
O4 - HKLM\..\Run: [NaturalPoint] "C:\Program Files (x86)\NaturalPoint\TrackIR4\TrackIR.exe"
O4 - HKCU\..\Run: [Sidebar] C:\Program Files\Windows Sidebar\sidebar.exe /autoRun
O4 - HKCU\..\Run: [WMPNSCFG] C:\Program Files (x86)\Windows Media Player\WMPNSCFG.exe
O4 - HKUS\S-1-5-19\..\Run: [Sidebar] %ProgramFiles%\Windows Sidebar\Sidebar.exe /detectMem (User 'LOCAL SERVICE')
O4 - HKUS\S-1-5-19\..\Run: [WindowsWelcomeCenter] rundll32.exe oobefldr.dll,ShowWelcomeCenter (User 'LOCAL SERVICE')
O4 - HKUS\S-1-5-20\..\Run: [Sidebar] %ProgramFiles%\Windows Sidebar\Sidebar.exe /detectMem (User 'NETWORK SERVICE')
O4 - Startup: Digsby.lnk = C:\Program Files (x86)\Digsby\digsby.exe
O4 - Startup: Xfire.lnk = C:\Program Files (x86)\Xfire\xfire.exe
O4 - Global Startup: ComproRemote.lnk = C:\Program Files (x86)\Common Files\VideoMate\ComproRemote.exe
O4 - Global Startup: ComproSchedulerDTV.lnk = C:\Program Files (x86)\Common Files\VideoMate\ComproSchedulerDTV.exe
O8 - Extra context menu item: E&xport to Microsoft Excel - res://C:\PROGRA~2\MICROS~1\Office12\EXCEL.EXE/3000
O9 - Extra button: IE7Pro Preferences - {0026439F-A980-4f18-8C95-4F1CBBF9C1D8} - C:\Program Files (x86)\IEPro\iepro.dll
O9 - Extra 'Tools' menuitem: IE7Pro Preferences - {0026439F-A980-4f18-8C95-4F1CBBF9C1D8} - C:\Program Files (x86)\IEPro\iepro.dll
O9 - Extra button: Research - {92780B25-18CC-41C8-B9BE-3C9C571A8263} - C:\PROGRA~2\MICROS~1\Office12\REFIEBAR.DLL
O10 - Unknown file in Winsock LSP: c:\windows\system32\nvlsp.dll
O10 - Unknown file in Winsock LSP: c:\windows\system32\nvlsp.dll
O10 - Unknown file in Winsock LSP: c:\windows\system32\nvlsp.dll
O10 - Unknown file in Winsock LSP: c:\windows\system32\nvlsp.dll
O10 - Unknown file in Winsock LSP: c:\windows\system32\nvlsp.dll
O10 - Unknown file in Winsock LSP: c:\windows\system32\nvlsp.dll
O10 - Unknown file in Winsock LSP: c:\windows\system32\nvlsp.dll
O13 - Gopher Prefix: 
O16 - DPF: {D27CDB6E-AE6D-11CF-96B8-444553540000} (Shockwave Flash Object) - http://fpdownload2.macromedia.com/get/shockwave/cabs/flash/swflash.cab
O17 - HKLM\System\CCS\Services\Tcpip\..\{85A8CC1A-C92E-40C5-9C88-B94B8914DB65}: NameServer = 192.168.1.1,4.2.2.1
O22 - SharedTaskScheduler: Component Categories cache daemon - {8C7461EF-2B13-11d2-BE35-3078302C2030} - C:\Windows\system32\browseui.dll
O23 - Service: @%SystemRoot%\system32\Alg.exe,-112 (ALG) - Unknown owner - C:\Windows\System32\alg.exe (file missing)
O23 - Service: @dfsrres.dll,-101 (DFSR) - Unknown owner - C:\Windows\system32\DFSR.exe (file missing)
O23 - Service: @%systemroot%\system32\fxsresm.dll,-118 (Fax) - Unknown owner - C:\Windows\system32\fxssvc.exe (file missing)
O23 - Service: ForceWare Intelligent Application Manager (IAM) - Unknown owner - C:\Program Files\NVIDIA Corporation\NetworkAccessManager\bin32\nSvcAppFlt.exe
O23 - Service: InstallDriver Table Manager (IDriverT) - Macrovision Corporation - C:\Program Files (x86)\Common Files\InstallShield\Driver\11\Intel 32\IDriverT.exe
O23 - Service: @keyiso.dll,-100 (KeyIso) - Unknown owner - C:\Windows\system32\lsass.exe (file missing)
O23 - Service: @comres.dll,-2797 (MSDTC) - Unknown owner - C:\Windows\System32\msdtc.exe (file missing)
O23 - Service: @%SystemRoot%\System32\netlogon.dll,-102 (Netlogon) - Unknown owner - C:\Windows\system32\lsass.exe (file missing)
O23 - Service: nHancer Support (nHancer) - KSE - Korndörfer Software Engineering - C:\Program Files\nHancer\nHancerService.exe
O23 - Service: ForceWare IP service (nSvcIp) - Unknown owner - C:\Program Files\NVIDIA Corporation\NetworkAccessManager\bin32\nSvcIp.exe
O23 - Service: NVIDIA Display Driver Service (nvsvc) - Unknown owner - C:\Windows\system32\nvvsvc.exe (file missing)
O23 - Service: @%systemroot%\system32\psbase.dll,-300 (ProtectedStorage) - Unknown owner - C:\Windows\system32\lsass.exe (file missing)
O23 - Service: @%systemroot%\system32\Locator.exe,-2 (RpcLocator) - Unknown owner - C:\Windows\system32\locator.exe (file missing)
O23 - Service: @%SystemRoot%\system32\samsrv.dll,-1 (SamSs) - Unknown owner - C:\Windows\system32\lsass.exe (file missing)
O23 - Service: @%SystemRoot%\system32\SLsvc.exe,-101 (slsvc) - Unknown owner - C:\Windows\system32\SLsvc.exe (file missing)
O23 - Service: @%SystemRoot%\system32\snmptrap.exe,-3 (SNMPTRAP) - Unknown owner - C:\Windows\System32\snmptrap.exe (file missing)
O23 - Service: @%systemroot%\system32\spoolsv.exe,-1 (Spooler) - Unknown owner - C:\Windows\System32\spoolsv.exe (file missing)
O23 - Service: @%SystemRoot%\system32\ui0detect.exe,-101 (UI0Detect) - Unknown owner - C:\Windows\system32\UI0Detect.exe (file missing)
O23 - Service: @%SystemRoot%\system32\vds.exe,-100 (vds) - Unknown owner - C:\Windows\System32\vds.exe (file missing)
O23 - Service: @%systemroot%\system32\vssvc.exe,-102 (VSS) - Unknown owner - C:\Windows\system32\vssvc.exe (file missing)
O23 - Service: @%systemroot%\system32\wbengine.exe,-104 (wbengine) - Unknown owner - C:\Windows\system32\wbengine.exe (file missing)
O23 - Service: @%Systemroot%\system32\wbem\wmiapsrv.exe,-110 (wmiApSrv) - Unknown owner - C:\Windows\system32\wbem\WmiApSrv.exe (file missing)
O23 - Service: @%ProgramFiles%\Windows Media Player\wmpnetwk.exe,-101 (WMPNetworkSvc) - Unknown owner - C:\Program Files (x86)\Windows Media Player\wmpnetwk.exe (file missing)
 
--
End of file - 7409 bytes

Select allOpen in new window

Did a Memtest and halfway through Test #4 it froze (with no errors so far) - so it's not software and not memory - I'm thinking mobo or CPU.
 
If I can't sell it I'm determined to diagnose the fault and fix it asap

In XP - I installed and ran Seagate's SeasTools for Windows
Did the first few tests on all my HDDs (yes they're all Seagate ones) and all tests so far have passed - I'll do a few more tomorrow for the sake of it).
What's more insteresting is that I started those tests at around 7pm and it's now nearly 12pm - that's a very long time with no freezing. I would be interested to see if I turned of Aero in Vista and used the very old Classic theme if the length of time before freezing increases dramatically. Unfortunately I can't put it to good use since Seatools doesn't run on Vista x64.
What apps can you think of that will test the motherboard, the CPU, and also the VGA RAM (linked to games and Aero)? A person who services our Toshiba laptops at work mentioned that faulty RAM in the graphics card can cause issues but it's very hard to test it. I'm not 100% certain why the PC freezes more often in Vista than XP, in games more often than office software, in video editing apps, in YouTube.com, in Memtest (no OS loaded - very basic drivers if any), and sometimes completely at random.
Further down the track, I will uninstall the drivers for my motherboard and see if that makes any difference. I may do that same for the video card. Only problem is that I probably won't be able to test it too much by playing a game when the PC is in that state - but we'll see. Plase continue to post your ideas.

I've been finding the PC much more reliable than at previous times, even with all 4GB of ram and all four CPU cores enabled - as long as:
I use XP in preference of Vista (haven't tried turning off aero yet)
& I don't play new games (I was able to play an old NFS game (Porsche 2000) with not adverse effects whilst using only the keyboard, in opposition to a new game like TDU along with a G25 (steering wheel) and TrackIR device.

You guys still around here or should I create a Related Question and start again?

I'm still here, but I'm just a bystander.
:o)

If Gary can't help I certainly can't!

At the risk of stating the obvious,from what you describe:
"the PC freezes more often in Vista than XP, in games more often than office software, in video editing apps"

It seems very likely that video RAM is a major factor.

Maybe a new question would get some fresh minds on the job.

Well I turned on my PC this morning - went into XP - opened Outlook to monitor what's going on at work, and copied some stuff to usb, and it's been going strong all day. I was thinking video ram as well - but how would that explain freezing in Memtest? I turned off Aero but Vista still freezes while surfing the net

I'm monitoring your new question on this -- but will refrain from posting unless I see something that "jumps out" as yet-another-thing to try (hard to imagine !!).

One thing I HAVE noticed on numerous systems:  XP is far more stable than Vista -- especially with high end graphics cards (the drivers simply seem to be more mature).   If XP remains stable on your system, you might want to just use it :-)   [If you need to run Vista for testing purposes for developed projects, etc. you could install it in a Virtual PC running under XP => you can even run x64 versions if you use VMWare instead of Virtual PC]

The trouble is as soon as I start using programs/games on XP to make use of the system's high performance it freezes 10 mins later (or sometimes longer - up to an  hour sometimes)

If the system is consistently stable in XP EXCEPT when you play video intensive games, this certainly tends to implicate the video card.   Do you still have access to your brother's card?   It would be interesting to see if, with that card and XP, your system is stable now.

Also [not sure we tested this -- this IS a long thread :-) ] ... start up 4 instances (one per core) of the stress test in Prime95 [http://mersenne.org/gimps/p95v2414.exe ] and let them run for an hour or so ==> if the system doesn't freeze doing that, then it's a fairly safe bet the issue isn't your CPU or memory.

What a thread! Did this ever get resolved?

Anybody ever take a look at northbridge chip temps?

Noticed I'm featured int eh newsletter AGAIN (thanks vallis)

Maybe that'll get more attention to this thread and the problem will be solved soon.
Willcomp - how do you propose i check the temps of the north-bridge chip?

Northbridge should have a temperature sensor. I'd assume that Everest would pick up sensor. If not, try SIW. http://www.gtopala.com/siw-download.html

It's a shot in the dark. I was trying to think of a component that would cause a frozen display other than video card and hard disk. An overheated or faulty video card will definitely cause a frozen display --> as I know from first hand experience on multiple PCs. Your symptoms don't seem consistent with hard disk read failures.

Yeah, I'm happy with the HDD's too - I've done lots or reading and writing to the drives while the PC has been running (movie editing etc) and have run all the Seagate diagnostic tools successfully on each of the 3 Seagate HDDs.

I can only think it's between the motherboard, graphics card, or CPU. I have both SIW and Everest and will check out the temp reading later on tonight.

An interesting read, needing two refills of coffee!..

Anyway, from the perspective of a time served hardware engineer...

Having a "Too big" a PSU can often lead to trouble, as they need a reasonable fraction of the main rail's capacity to be used, so that the secondary rails will have enough energy to regulate correctly.  It is also not unusual for even brand new PSU's to have a small problem on one rail that will not show up on a tester, but give it a dynamic load like a MOBO, and it will misbehave.

For an indipendant memory test (not sure how many GB it can do, but one stick at a time?) try memtest386, you should be able to find an ISO image by googling for it, so as to make a bootable CD, so no OS worries.   Oh, and it's usualy free...   Make sure you get the latest version, at least one of the earlier ones was found not to do a test correctly.  It's a "Propper" memory test, walking one's, zero's, complementary address patterns etc and so forth..  Just like the old days!

As for hard drives, look at Steve Gibson's "Spinrite" tool..  (at www.grc.com) OK, so it's 80USD or so, but it's "the mut's nuts" of drive diag's and data recovery.   Again, you get an ISO and make a bootable CD (or floppy)   It has two main uses, to recover a corrupted drive (what most people would throw away!) or to scan for and identify any potentialy suspect areas of the surfaces, then to mark them as unusable.   Don't take my word for it, look at the testimonials on Steve site.   I have to admit to letting it run over any new drives I buy, and it's suprising what it often finds and flags as suspect.

I used to repair minicomputers years ago in a past job, even rebuilding hard drives (the old rack mounted 14" things!) so I understand memory tests and hard drives.  I always wondered about the PC world's reliance on what the OS and hardware makers gave you for "diagnostics".  As a result, I carry my own copies (legal ones!) of both tools, and they have got me and friends out of "digital holes" on a few ocasions now, saving me way more than the purchace cost of Spinrite in the process.

The only downside for both of them, is the time they need to run.  Even on the fastest CPU's expect them to need hours to run at the very least.   Let Memtest386 do several full passes to be sure.  The record run time for Spinrite in recovery mode, I believe was over 3 months!  But it did the job and recovered the data from an otherwise failed drive.  The running guestimation time for a full diag scan is quite acurate, unless it finds a problem and has to do something with it.

Thanks again guys for an interesting read.  Time to go warm up some water for more coffee!..

Dave B.

Thanks for taking the time to read this massive thread (let alone post in it). Note the 'relevant question' I opened up which continues on from here and has points available until I actually solve this problem.

I have run the memtest iso file previously, but even that froze at about 4% completion on each of the 3 times I tried to run it, which told me that I definitely didn't have software/driver issues.

I'll check to see if the '386' version is any different to the latest version I used. I have to go out for a few hours before I can do that though.

OK Ryan, I obviously missed that detail, as you say, there is a lot of it! ;-)

One idea, if you have a compatable PC to hand, is to test each of the memory sticks in that with whatever diagnostic you choose.  If nothing else, it'll prove one way or the other that there are no bad cells in the memory, but won't prove any access timing problems with the other mobo/ram combination of course.

I only found this thread as it's mentioned in the EE digest mail that arrived today, but have to say it's one of the more interesting questions, by far!...

All the best, going back to lurk mode again.

Dave B

I went searching for a 'Northbridge Temp' reading in SIW and Everest but didn't find one. However - I did notice the unsually high 'motherboard temperature'

See below:

--------[ Sensor ]------------------------------------------------------------------------------------------------------
 
    Sensor Properties:
      Sensor Type                                       Winbond W83627DHG  (ISA 290h)
      GPU Sensor Type                                   Driver  (NV-DRV)
      Chassis Intrusion Detected                        Yes
 
    Temperatures:
      Motherboard                                       90 °C  (194 °F)
      CPU                                               31 °C  (88 °F)
      Aux                                               28 °C  (82 °F)
      GPU1: GPU                                         57 °C  (135 °F)
      GPU2: GPU                                         58 °C  (136 °F)
      Seagate ST3250620A                                25 °C  (77 °F)
      Seagate ST31000340AS                              26 °C  (79 °F)
      Seagate ST3320820AS                               27 °C  (81 °F)
 
    Cooling Fans:
      CPU                                               1205 RPM
 
    Voltage Values:
      CPU Core                                          1.21 V
      Aux                                               2.74 V
      +3.3 V                                            3.26 V
      +5 V                                              5.46 V
      +12 V                                             10.03 V
      -12 V                                             -14.42 V
      +5 V Standby                                      4.95 V
      VBAT Battery                                      2.99 V
      Debug Info F                                      FF 8C FF FF FF
      Debug Info T                                      90 31 28
      Debug Info V                                      97 AB CC CB A5 06 E4 (03)

Select allOpen in new window

Will shutdown the PC, and then will boot up later and monitor the mobo temp to see how long it takes to rise. Remeber I have a lot of cool air passing through this case - 3 intake fans and 2 outtake ones (one being a huge 200m fan)

The + and - 12V readings look a bit wild!   +10 and -14?

How critical are they for things these days?   I know many drives still use the +12V rail for spindle motor power etc, but what still uses the -12V?

Also, if the +5V supply is truly at 5.46V, that too is a "tad high" perhaps.

Any chance of verifying them for real with a DMM?   Branch out an unused power cable for example through a slot somewhere, and probe it as it warms up.

I have had PSU problems in the past where (funnily enough) the +12V went "funny" under load when a CD drive spun up.  The original symptoms were that the system would sometimes not boot.   As it turned out, because the boot drive didn't always get up to speed fast enough, due to the failing PSU.

Dave B.

I'll say that I know next to nothing about system voltages and minor stuff about PSU rails. The PSU had been sent back for warranty previously, and out of all the parts I sent back - it was the only one that wasn't replaced, as it was found to be not faulty. You're probably thinking at this point 'what do they know?'   :)

If any BIOS voltage settings can/should be changed, let me know and I'll see if it makes any difference.

2.5 hours later - I've turned the desktop back on (using laptop until now) and the sensor for the motherboard remains at 90-92 degrees celcius according to the sensor reading displayed in Everest Ultimate 2007 beta

Could the sensor be a dud, freezing my PC when it gets too high? Should I consume another 2 weeks sending it back for replacement again?

Hi..

Hmmm..  I've seen some older systems in the past, that generic diags think there is a sensor present, but in truth there is no sensor, so they read a screwy temperature value.  (In one case, the main board was suposedly running at over 300'C!)

My guess, is that not all makers fully implement the things the same way.

Do you have a CD with any maker specific tools, diagnositcs, drivers etc for that particular mobo?   If so, see if there is any status monitoring tool of their own that could tell the truth.

It may be worth a scratch arround the bios too, to see what settings there are, that could shutdown things if they get too warm, or otherwise out of tolerance.

I know it's been mentioned before, but you could relax the memory timings (more wait cycles etc) and see if that improves things.

Clutching at straws now, and I don't recall if this has been mentioned.   If you find a CPU Clock "Spread Spectrum" or "EMC", "CE" option, try turning it off, or On, ie, flip the setting...   There is a very slight remote chance, that if something is "on the raged egde" timing wise, it could change things for the better, or worse.  In essance, it could highlight any memory or other timing issues, if after changing that setting, the system behaves in a different way, better or worse.

What that setting does, is it purposefully jitters the clock frequency, so as to "fool" the detectors in product compliance EMC test facilities.   I've never heard of it affecting the basic operation of a system, but there is always a first time.

I'll be quitting for the week soon, but will be back on Monday if anyone has any other comments.

Cheers.

Dave B.

nVidia System Monitor reports that the System Temp is a cool 29 degrees celcius. There goes that theory out the window. I'll check the BIOS stats to be on the safe side.

I'll flip the Spread Spectrum switch as well. I'm pretty sure I've fiddled with all the options in the CPU page - changing one making the system incredibly slow (ie 386 speed) - took 1 min to display the Vista OS DualBoot menu.

I'll also have a look at memory timings, and reduce each related setting by 2 units and see if that makes any difference.

Well I've disabled the CPU Spread Spectrum - but there are also SS settings for "HT", PCIe (SPP and MCP), and SATA Spread Spectrum. Leaving those settings alone for now.

CPU Multiplier currently at 8x - leaving that and the other multiplier settings as standard for now as well.

Changing memory timing settings from 7, 7, 7, 20 to 9, 9, 9, 21

Changing CPU Thermal mode to TM2 only (from TM1 and TM2)

Other CPU Settings are as follows (unchanged):
Limit CPUID MaxVal - Disabled
Intel Speedstep - Disabled
PPM Mode - Native Mode
C1E Enhanced Halt State - Disabled
Executable Disable Bit - Disabled
Virtualization Technology - Disabled

Voltages:
CPU Core - from 2.00000 to 1.18750 (-2 units)
CPU FSB - from 1.1v to 1.1v (doesn't go any lower)
Memory - from 1.500v to 1.500v (lowest)
nForce SPP and MCP are also at their lowest settings (1.3v & 1.5v)

I am disabling 'OS Comp' to see if it makes a dif.
I will also disable to HPET (High Precision Event Timer) Function

BIOS also reports the Board temperature as 29 degrees, and the nForce fan speed at 3880rpm.
I'll save these settings and turn the PC off and go to bed. That should give you some time to make sure I haven't changed any settings to something horribly wrong - and I'll check back here before turning the desktop back on tomorrow.

Until then...
Ryan

I'm currently running Burn-In Test (have been for the past 1.15 hours).
Surprisingly, no freezes at all tonight so far.

The PC is almost coping ok with the torture testing of pretty much every part of the PC (including 3D gfx card test), with no errors to report. I was unable to launch nVidia System monitor after launching the test due to it being too slow, but Everest is monitoring the temps. The motherboard temp went from 88 degrees to 100 in a very short time (if that's accurate or not I still don't really know), and gfx card temps are below 60, while CPU temp is about 52c.

45 mins left until the tests complete.

I did a quick test earlier which failed - simply because the disc in one of the DVD drives was faulty - changed the disc, tested again - no faults to report.

Will run the case bare tomorrow afternoon since I have tomorrow off work.

Burn-In Test results below - still no freezes tonight (despite not making any hardware changes yet):

PassMark BurnInTest Log file  -  http://www.passmark.com
========================================================
 
BurnInTest V5.3 Pro 1022
Logging detail level: Normal
 
******************
SYSTEM INFORMATION
******************
Computer Name: RYAN
 
Date: 10/07/08 23:28:27
Operating system: Windows XP Professional  Service Pack 2 build 2600
Number of CPUs: 1 (4 Core(s)/CPU, 1 Logical(s)/Core)
CPU manufacturer: GenuineIntel
CPU type: Intel(R) Core(TM)2 Quad  CPU   Q9550  @ 2.83GHz
CPU features: MMX SSE SSE2 SSE3 PAE 
CPU1 speed: 2833.2 MHz
CPU L2 Cache: 6 MB
RAM: 2814 MB
Video card: NVIDIA GeForce 9800 GX2 (Resolution: 1440x900x32)
Disk drive: Model ST3250620A (Size: 232.9GB)
Disk drive: Model ST310003 40AS SCSI Disk Device (Size: 931.5GB)
Disk drive: Model ST332082 0AS SCSI Disk Device (Size: 298.1GB)
Optical drive: PIONEER DVD-RW  DVR-112D
Optical drive: ASUS DRW-1814BLT SCSI CdRom Device
Optical drive: VBE HH DVD-ROM DVD-50D SCSI CdRom Device
 
 
**************
RESULT SUMMARY
**************
Test Start time: Tue Oct 07 21:27:19 2008 
Test Stop time: Tue Oct 07 23:27:23 2008
Test Duration: 002h 00m 04s 
Temperature  (Min / Current / Max): 0.0 / 0.0 / 0.0
Temperature  (Min / Current / Max): 0.0 / 0.0 / 0.0
 
Test Name                   Cycles   Operations      Result Errors   Last Error
              CPU - Maths   5670     1.434 Trillion  PASS   0        No errors
               CPU - SIMD   1250     951 Billion     PASS   0        No errors
             Memory (RAM)   2847     1.514 Trillion  PASS   0        No errors
              2D Graphics   49       49679           FAIL   1        No operations reported in timeout period
              3D Graphics   380      761358          PASS   0        No errors
    Disk (C:  Recordings)   28       9.963 Billion   PASS   0        No errors
         Disk (D:  Vista)   10       10.989 Billion  PASS   0        No errors
         Disk (E:  Linux)   31       10.997 Billion  PASS   0        No errors
   Disk (F:  IDE Storage)   1        7.441 Billion   PASS   0        No errors
  Disk (G:  SATA Storage)   2        9.051 Billion   PASS   0        No errors
            Disk (H:  XP)   10       11.156 Billion  PASS   0        No errors
   Disk (I:  1TB Storage)   0        10.050 Billion  PASS   0        No errors
              CD/DVD (J:)   1        2.822 Billion   PASS   0        No errors
              CD/DVD (K:)   2        10.554 Billion  PASS   0        No errors
              CD/DVD (V:)   8        7.181 Billion   PASS   0        No errors
                    Sound   74       124 Million     PASS   0        No errors
                Network 1   125      1.005 Million   PASS   0        No errors
                Network 2   141      1.128 Million   PASS   0        No errors
           Video Playback   2        6768            FAIL   1        Video Codec not supported or audio only.
TEST RUN FAILED
 
******************
DETAILED EVENT LOG
******************
LOG NOTE: 2008-10-07 21:27:19, Status, Main Tests started
LOG NOTE: 2008-10-07 21:27:20, 3D Graphics, 3D test starting: Duty cycle = 50. Test on primary monitor (0). (x,y) = (310,0). WxH = 300x200. 
LOG NOTE: 2008-10-07 21:27:25, Video, Video test starting: Duty cycle = 50. Test on primary monitor (0). (x,y) = (620,200). WxH = 300x200. 
CRITICAL   : 2008-10-07 21:28:45, Video, Video Codec not supported or audio only.
LOG NOTE: 2008-10-07 21:28:45, Video, Monitor: 0
WARNING    : 2008-10-07 23:11:19, 2D Graphics, No operations reported in timeout period
LOG NOTE: 2008-10-07 23:27:47, Status, Test run stopped

Select allOpen in new window

After the tests, Everest reports the motherboard temp as being 94 degress (down from 101 previously).
nVidia Sys Mon reports 32 degrees for it.

94 degrees faranheit = 34 degress celcius - could this be a possible explanation? All other temps in Everest are correctly shown in celcius.

Goodnight for now - going to bed - shame I don't have more time to try and make this PC freeze today.

Hmm...

Interesting the 'C 'F funny.  Guess the Everest tool (US in origin?) is defaulting that reading to 'F regardless of what you tell it to do.  A bug perhaps, is there any newer version?  Sorry, but I don't know that tool, other than from your references..

Also interesting (to me) that there are spread spectrum settings for other parts of the system.  (HT? wonder what that is)  Makes me wonder if that board would ever show up in a ready made system in the EU!  We have some stringent EMC Emisions rules, but that's what the SS options were for, to fool the measuring receivers into reading a lower level, than it actualy is.  Smoke and mirrors, as I think I said once before.

Either way, it sounds like you're getting close to a reliable system now?

Regards etc..

Dave B.

Saw your comment r.e. the 1TB Seagate potentially being the culprit here !!

Wow !!   If that's the case, it's certainly a strange symptom => I've seen many hard drives cause freezes or lockups ... but ALWAYS with some indication of a disk error (or errors).   In any event, it'll be great if that's it -- this problem has certainly gone on long enough !! :-)

... I just ordered a couple of the new 1.5TB Seagates ==> hopefully my system won't start freezing :-)

Well I changed the cable and used a dif sata port on the mobo and turned it on and Windows didn't boot. Didn't really surprise me, I expected I'd have to go into the BIOS to change the HDD priority, but only me 250 and 1024gb drives are there but not the 320 (which holds the Vista partition and boot info).

I'll watch Top Gear Australia for the next hour and try to fix it after.

Man it's all good/bad news here:

Good - Might know whats wrong with the PC
Bad - Means I have to try and move 600gb of data to who knows where
Bad - Now the TV Tuner card has packed up, refusing to work in XP and Vista in either PCI slot
Good - The warranty runs out tomorrow, so I should just be able to get it replaced
Bad - the Top Gear tape screwed up - that's the trouble with VCR's I guess
Good - I can most likely download the episode tomorrow anyway

Now that I've swapped the cable, I have to see if the PC will still freeze (bad HDD) or if it's fine now (bad cable). In which case I'd be deleting a lot of stuff for nothing.

What I don't get, like you Gary, is why a bad HDD causes the system to freeze, when there are no noises, no errors (delayed write fail or other stuff), no issues with the HDD disappearing from My Computer instead of locking up the whole PC, the fact that the drive is 100% storage only, no system stuff or page file on there, and I also don't get why I've been able to copy around 600GB of stuff onto that drive and it's never frozen once during any of those processes.

It does however, explain some of the misleading symptoms that previously pointed to the GFX card

Freezing in games: All games installed on that drive
Freezing in Video editing - all videos on that drive.

Installing a new game onto an old HDD and being able to play for hours fine, followed by playing another game on the bad HDD and freezing 20 mins later is what prompted me to consider it further in the first place.

Ryan.

Do you or anyone you know have a copy of Spinrite?  If you do, it may be worth letting that work on the "suspect" drive and see what it finds about the disk surface and any defects.

With that size of disk drive, they will rely on lots of error detection/correction internaly to make sure your data is read OK.  (The drive will do that all by itself)   If that takes much longer than "normal", Windows can apear deadlocked if that happens when it's way down deep in the Kernal doing low level disk IO stuff.

Just a thought...

Dave B.

See this Dave: (from the other thread on this)
https://www.experts-exchange.com/questions/23758270/Faulty-Hardware-Causing-PC-to-freeze.html?anchorAnswerId=22701669#a22701669

Hi Ryan,

Thanks for that link...

Yes, buggy power distribution can do all sorts of silly things.

Interestingly from what (I think) you said it was a black wire you had to move.  So, ground problems, much like they are in automobiles will create mayhem in PC's, not realy a suprise I guess.    Hope that as you think is/was the cause of the problem.

I'll monitor that other thread too,just out of curiosity.

All the best Ryan.

Dave B

Well with a different cable in Vista and XP froze within minutes - so I'm back to assuming a faulty HDD.

I have 295GB of uncompressed AVI videos (from FRAPS) that I need to shrink down to fit on the other drives so I'm hunting down a batch converter.

Since you mentioned (in the other thread) that it took a LOT of time and dialogue to (hopefully) resolve this, just thought you'd like to know that THIS is the 250th post in this thread !!   The other one only has about 15 or so at this point :-)

Well it ain't over yet - because with all the data moved over and the 1TB drive unplugged, it froze again.
I'm now moving all the date from my 320GB drive to the 1TB drive and will try running without that 0 but I'll need to reinstall an OS since both are on that drive.

You know how I dismissed the motherboard temperature of 90 degrees + as being inaccurately shown in Fahrenheit, I just noticed this (http:#22632316) where it clearly says 90 in Celsius and then 190something in Fahrenheit.

What do you think? Is it worth me trying to replace the motherboard on its own one more time before sending the whole PC back to CPL and paying them a few hundred dollars to fix it? It's funny how each time I ask them about fixing it over the phone they're quick to say a minimum charge of $30 applies, but when I ask what the maximum charge is they stumble and repeatedly say 'ummmm'.

Hear from you later on.