My new C2D system committed suicide...

[Himajin]

I'm wondering if any knowledgable people here might be able to diagnose the problems I'm having with my new Core 2 Duo system.
The symptoms experienced were, in order:

1.) System died suddenly while playing Oblivion (first demanding game tested after building 24 hours previous with no issues)
2.) Utter failure to post or even power up following this.
3.) Boots to XP successfully 12 hours later. XP reports new PCI device, although all devices seem to be accounted for in settings...
4.) While trying to diagnose #3, system dies again
5.) Cleared CMOS, rebooted and started to fix BIOS settings, and system dies a third time.
6.) Once again system fails to power up at all, although LEDs are still lit on motherboard
7.) After turning rear power switch off and on, LEDs also fail to light up anymore

Here are the components:
Core 2 Duo E6600
Scythe Infinity CPU cooler
MSI 975x Platinum Powerup edition
2GB Corsair XMS C5
Gigabyte Geforce 8800GTS (reported on bootup as "Engineering Revision - Not for Production Use," although bought retail...)
3 x PATA drives (legacy from old system)
2 x SATA drives (320, 320) no raid
Enermax Liberty (EL620AWT) 620W
Audigy Platinum (legacy from old system)
Ricoh DVD+RW (legacy from old system)
Case (legacy from old system)
(all at stock)


My initial feeling was that the PSU had failed after the video card upped the juice, although I think 620W was more than enough for this system.
I'm also wary of the PCI device reported after managing to boot to windows again. Could this indicate either the video card or the M/B considering the symptoms?

I'm unfortunately unable to test the parts individually, as my old system is incompatible with most.

If I don't hear otherwise, I'm going to assume that the PSU is at least one of the components at fault, and get a replacement.
I'm a bit worried that there might be other nasties afoot here, though. Any advice would be greatly appreciated.

-=Himajin=-

 

[sdgserv]

Wild guess..Sounds like PS..I have had a few Enermax die on me..My problems were like yours..Nothing made sense, I could not eliminate one thing for another. Would work in the morning then no reboot. Not to say enermax is a bad PS..try something else. Or borrow one from somebody if that is possible. I would do it soon. A bad PS can cause additional problems to MoBo or Vid cards.

There is my worthless opinion..

 

[NulloModo]

Make sure you have the cooler seated properly on the CPU and that your thermal compound is spread evenly.

Other than that, I agree that the PSU not being able to keep up is the only thing that makes sense. While the C2D is very power efficient, the 8800 is a huge power hog, and running 5 hard drives as well may just be pushing the envelope a little, though I would think a good brand 620 watt PSU would be able to handle it.

 

[dVeLoPe]

Try those 2 things said in the previous posts and if that doesn't work try booting up with BARE essentials (remove all but OS hdd) and try booting it up with just the MOBO,CPU,MEM,GPU,HDD

 

[u8myrice]

I had the same thing happened to me with my PD820. I replaced the PSU, nothing else, and it worked like a charm. I'm using the same machine with the new PSU right now and it's running flawlessly.

I would advise you to check your system with another PSU.

 

[Himajin]

Thanks all for the advice. I'll try these suggestions out, and let you know how it goes...

-=Himajin=-

 

[nytrackstar]

try this...unplug the comp from the wall, press the power button to discharge any power left...turn off the psu, remove the cmos battery set the jumper in the "clear" position, leave it all night, thatll make sure the capacitors are fully drained... see if that helps get her up an running again to help futher diagnosis... also, set bios to "safe" settings (reffered to differently with different mobos)

did you run any sort of stress test after putting your system together? (prime95, memtest86)...see if you can get memtest86 (google this) to boot from CD, check for any errors. if u can get this running, this should help eliminating memory as an issue.

although the most obvious culprit might be the psu, i wouldnt rule out all other possibilities so quickly...try and get a hold of a psu tester such as this one (LINK)...make sure all rails are working... if possible try and get multimeter readings of psu rails to make sure they are strong enough...

in regards to the GPU, i would definately report that to GB. there could easily be some kind of conflict between your motherboard and gfx card... honestly, that to me sounds like the most likely reason... since it crashed during a video intensive activity... it could have easily f'd up your motherboard some how... do u have any other gfx card that you could stick in temporarily?


if none of the above helps, i'd remove everything from the case and install it on an anti-static bag on the mobo box (or something similar). BARE mininum config...and see if you can get her to boot. if you can, try the things i mentioned above.

it seems to me that you can get her to boot every now and then, means theres signs of life... dont be discouraged if you try the techniques above and she wont boot, be persistant, i had issues with booting, tried resetting cmos 7 times (overnight) and on the eigth it finally worked...

but yea to sum up:
main focus should be on vid card and psu
- try easiest first (cmos reset and such before replacing parts)
- contact GB in regards to "engineering revision" message
- bare mininums means: 1 stick of ram, vid card, no hd's, and no pci cards... goal is to see if she can boot, if she does, add one thing at a time... start with the 2nd stick of ram, then a single hd, and on...

GL, you can message me if you need further help, i could probably get on the phone with u and walk u thru some steps if u need.

 

[MakubexGB]

Get a PSU tester (~$15) just to make sure. I also think the PSU it's the problem, but it wouldn't hurt to be %100 sure before sending it to RMA. Or also, try powering up another system (even if it's an old one) with that same PSU.

 

[Himajin]

Thanks for the extra replies (especially to nytrackstar for such a detailed post!). This gives me more ammo to work with.

I actually bought a new Seasonic 650W PSU on the way home from work, but I probably won't have time to get back to working on this problem until the weekend. Once I know more, I'll be sure to post an update. For now I'm going to unplug the current PSU to stop any further damage.

Although I hope nystrackstar isn't right, and it is just the PSU, I have a sneaky suspicion more is going on here. I'll get in touch with Gigabyte first of all and see if they know why my card is reporting itself as an engineering revision (a search turned up some other similar stories, but most people just got a firmware update from the manufacturor).

-=Himajin=-

 

[Himajin]

did you run any sort of stress test after putting your system together? (prime95, memtest86)

I ran memtest first thing after building, and the memory checked out okay.

do u have any other gfx card that you could stick in temporarily?

Unfortunately not, all I have is some old AGP cards.
If I still have problems with the new PSU, I'll cross that off the list and take it from there.

-=Himajin=-

 

[nullzero]

It may be your motherboard... I had nothing but problems with the MSI 975x board to the point I had to go out and buy a gigabyte DS3. I had horrible cold boot issues where the system would not turn on for a long time.

 

[nytrackstar]

let us know when youve tried everything i listed, i too hope im wrong and that its just a PSU issue... might be worth it to pick up some ~$20 pci-e card from a local computer store, not just for now but for later on, nice to have a back up incase ure vid card goes capoot (knock on wood!)

 

[Marvelous]

perhaps a mobo problem

 

[sluzbenik]

Gee, I hope you got a good deal on that 8800 then!

You had better really carefully add up your amps. Even 650 watts is probably pushing it with 5 drives and that card. The Enermax combined rating is max 36a. 80% efficiency gives you 28 amps max. 8800xts requires 26a from what I saw, so that leaves you very little headroom for the rest of the system.

 

[Himajin]

Update:
It's not the PSU, unless lack of wattage/ampage is really causing the problems (but I somehow don't think this is the case, as it was running fine for 24 hours before it died, and I've tried running bare bone now as well with no luck).

Anyway, I put the new PSU last night, and everything jumped to life, but it was very short-lived...
Once again the power dropped a few seconds after power-on, although at least this time the system was rebootable straight away, rather than refusing to power on at all.

That leaves the CPU, video card, or M/B.
I'm going to pick up a cheap PCI-E card as suggested above, and round things down as much as I can before I take the parts back for testing/replacement.
(the shop here in Japan I bought from has test systems up and running, so they should be able to confirm the problem and give me a replacement straight away...)

-=Himajin=-

 

[Himajin]

I have now tested a different GPU, memory, HDD, and tried removing the sound card, and I was able to rule those out.
I think I've pinpointed what the problem is, though.

I was wondering why the power was staying on longer with the new PSU, and realized that in between changing these over I'd also replaced the cheap silicone grease that came with the CPU cooler for some arctic silver. I managed to get in the BIOS long enough to confirm the CPU temperature, and right after a cold boot it was already soaring up to 75c.
It then continued up to 93c or so before power cut out like usual. This is with the CPU cooler (a Scythe Inifinity) firmly secured flat on the CPU, and cooler fan attached spinning at around 1200 RPM...
The arctic silver was certainly not overdone, either, as I only put a dot about the size of a grain of rice in the middle, which flattened out to about the size of a nickel.
The cheap silicone grease that was on the CPU before that was also similarly spread.

I think the CPU must be defective in some way, although visually there doesn't appear to be anything wrong with it.
There was no evidence of a fan failure, etc. at the time the problem first occurred.

I'm going to take it back and get the shop test it out for me, and if they have the same problems I should be able to get a replacement.

Thanks all for taking the time to reply. It's good to finally have a solid lead on the cause of the problem!

-=Himajin=-

 

[Jakalwarrior]

Sounds like the IHS isnt making good contact with the core. Technically you violated your warranty but im not an arse like some moral nitpicking people so I say RMA it.

 

[direowls]

Have you stripped the sticker off the bottom of your infinity?

 

[ThreeDee]

Have you stripped the sticker off the bottom of your infinity?

lol , that would be kind of humorous if he didnt ...


easy fix tho ...and I like easy fixes!

 

[jreffy]

Have you stripped the sticker off the bottom of your infinity?

I laugh everytime someone says this, because I was guilty of that on the first computer I built years ago. My friend helped me pick the parts and order it (he was the computer expert at the time) and he was visiting his girlfriend out of town when the parts arrived. So I figured I could put it together myself, from the way he made it sound easy. I did, and it turned on and everything....and then just shut off.

The next day he came home and laughed at me because I forgot to take off the plastic film on the bottom of the heat sink =/