Is my opty 175 finally dead?

Well for the last few days overclocks that where once stable (2.8 280x10 1.5 volts) started to become unstable i was doing alot of bios changing at the time so i thought it might just b the bios, so i flashed back to the 623-1's that i had the best luck with and used the computer for a few hrs at 275x10 2.75 at 1.45, changed a few ram settings (i think it was tras from 6 to 8, and DDS from 7 to 8) did a memtest run and everything was fine loaded windows and i got a bluescreen once i got to the desktop, figured ok something is off with the ram timings now (so i set them back to an older stable set of timings that i had, and lowerd my cpu multi) got into windows again and ran OCCT, i hit start and looked at the temps and core 2 was at 65 deg c (this was at 2.0 at 1.3v mind u) and blue screen again right after i saw that, now i cant even load windows, that 65degs on core 2 is about 15 deg higher then i have EVER seen the temp it normally ran about 50 deg MAX at 2.8 and 1.5 volts (nothing was changed cooling wise and the comp hasnt moved in weeks so the heat sink hasn't been jarred)

Pretty much i would love some input as to what u think happened, anything and everything will b greatly appreciated

Duke

PS. this is now the PERFECT excuse for my GF so i can finally order my new parts, (she caught me on newegg.com with a full shopping cart lol, damn her timing was about 3 clicks away from ordering lul) but at the same time i would love it if i could save this system as a back-up

lol what is that 2 much reading for u guys? lol (jp), but i would really love 2 hear what u guys think about this

Duke

Double check the heatsink. Blow it out if need be.

I agree.

Also, your TIM may just be wearing out - that stuff doesn't last forever, you know. The AS5 on my Storm started going bad at 2.5 years and at this point, another year later, my CPU temps are running about 10°C higher than new ...

How long have you been running that system at 2.8 and 1.5v? It is well documented that OC'ing and increasing voltage can shorten the life of CPUs.. maybe it's finally saying, "I've had enough." But yea, like it was mentioned check your HSF for dust build-up, and possibly re-apply TIM.

MY heatsink is clean i just reseated the heatsink about a month ago because of just that reason of the tim being a little old for my taste, when i did that i also cleaned out the heat sink from any dust (brought it down to the shop and blew it out using 220 psi compressed air) all was fine....and it has...err was rather...at 2.8 for probably idk 2 years maybe even more, i never let temps get above 50deg.....why the 65 was a MOJOR shock to me at 2.0 and only 1.3v.....then a BSOD


Duke

There are only a few reasons a CPU temp would jump from 50 to 65°C overnight and I can't think of any having to do with CPU circuitry.

dirty HS
dirty fan
bad/old TIM
bad seat
bad case flow
worn spring (on clamp-ons)
broken frame/HS anchor
bad fan
run-away vCore (board power chips/regulators - does CPU-Z agree with BIOS?)

What did I miss ...?


Edit:

Dukemurmur said:

The only reason i don't put the one there that should b is i can't do it with the heatsink on, other wise i would use it, the board is solid with no flex at all (otherwise i would have figured out a way to make it work)

Duke

Click to expand...

Which screw isn't threre???

I should add board flex to that list ...

lol, read the rest of what was posted in that thread instead of just what u quoted and u would know what screw that screw has never been there for 4 years and as i said there is no board flex, the only thing i can think of is maybe the voltage was WAY off but i looked in bios, but it was set for 1.3v...and that is what occt said it was(i make a habit of knowing what my vcore is in bios and windows) unless for w.e reason that time is was off these boards have been known for overvolting if using the 123% (which i was not) i always use 104% once and a great while 110%, atm i have the system running with ****ty timings on everything, i can't get my old 2..75 stable now (used to b 2.8), i think something is kicking the can and w.e did happen didn't help things...(and yes it passes memtest86 just fine always has, and does atm i don't think it is my ram) think it has to do with the mobo dieing or the IMC/CPU

Duke

But even a bad board wouldn't explain the CPU temp jump - unless you're using the board sensor instead of something like CoreTemp to read the CPU temp ...?

MAYBE, your board started vdrooping? slim but possible.

Well on the GOOD side of things, i have new parts on the way for a 965 C3 build on an Asus Crosshair III and a 2x2gb kit of Dominator GT cas8 2000mhz, that have hyper IC's, nomnom, lol

So this Opty rig is going to get retired to backup, will drop the clocks back to nearly stock and it will b good enough for that / the random friend that wants to game here also, going to put win xp back on it also so i can run Battle field 2142 online if i want (havnt been able to get it to work on Win7, damn Punkbuster)

Dragon

So you're saying you never dropped in a copy of UBCD and did an extended run of Mez prime to test your CPU? That is just down right sloppy if you ask me. Sounds like there could be a few different problems honestly most of them have already been covered. Most likely you running at a higher voltage for so long has caused the issue. Still though boot up a copy of UBCD at the clocks you're running now Mez prime for like 4 hrs to start with. You pass it, its not your CPU. Have you ever thought that this issue could also be from an aging board? Having to take all that voltage can be rough on boards.

i have ran many other stability tests, Prime 95(blend, large FFT and have even tried small FFT), OCCT, y-cruncher, maxxpi, super pi, need i go on? but im am sure that running it at that voltage had something to do with it degrading if u will, i just don't understand the temp issue, as it has never done it again, and yeah i said i think it is the in the cpu/IMC or possibly the mobo a few posts ago

Dragon

How does it run @ 2.6 ?

atm i have just been running it at stock w/o many problems unless i really stress it for a few hrs with OCCT or similar, i have just been taking it easy on it as i have my new system on the way,it should b here on Tuesday-Wednesday, and i have an old MSI board i might try with this CPU once i get my X4 965 up and running

Dragon

Dukemurmur said:

i have ran many other stability tests, Prime 95(blend, large FFT and have even tried small FFT), OCCT, y-cruncher, maxxpi, super pi, need i go on? but im am sure that running it at that voltage had something to do with it degrading if u will, i just don't understand the temp issue, as it has never done it again, and yeah i said i think it is the in the cpu/IMC or possibly the mobo a few posts ago

Dragon

Click to expand...

Correct but those are windows based apps. To test properly you should honestly use a very stripped image much like the prelinux environment that is use on UBCD. So another question is yes the heat on 1 core did go but have you tested your ram? Now onto the heating up core issue. Am I correct to assume that you've not lidded your opti? If you've never lidded your opti then it could be an issue with the IHS it is a rather crap thermal paste they use on inside. Maybe one core has been off putting more heat than the other all along. At some point along the way say said IHS thermal transfer over that 1 core dropped in C/W rating. Ok now lets also say that your OC + Volts was pressing you're CPU to the max(Not say it was but some CPUs are funny like that) That drop in C/W rating on that one area of the IHS covering said core could have cause a back log of heat into the core. Thus the increase of heat on that core. Which if a back log of heat occurs then you can have serious issues resulting in burning of micro traces inside of the core itself.

Dukemurmur said:

atm i have just been running it at stock w/o many problems unless i really stress it for a few hrs with OCCT or similar, i have just been taking it easy on it as i have my new system on the way,it should b here on Tuesday-Wednesday, and i have an old MSI board i might try with this CPU once i get my X4 965 up and running

Dragon

Click to expand...

Well good luck on the new system. Let us know how the opti fairs in the MSI.

the IHS has been removed since the first day i got the chip as a matter of fact it is right here next to me on the desk atm, just odd that it was higher temps at 2.0 and 1.3v then i have ever seen b4 even at 2.8 plus testing with 1.55v (only ever tried this a few times as it was eveident it wouldnt go over 2.8)

and ty, i hope the new build goes well also lol, and yeah will try to keep ya updated with what happens on the MSI Neo 2 plat board

Dragon

Dukemurmur said:

the IHS has been removed since the first day i got the chip as a matter of fact it is right here next to me on the desk atm, just odd that it was higher temps at 2.0 and 1.3v then i have ever seen b4 even at 2.8 plus testing with 1.55v (only ever tried this a few times as it was eveident it wouldnt go over 2.8)

and ty, i hope the new build goes well also lol, and yeah will try to keep ya updated with what happens on the MSI Neo 2 plat board

Dragon

Click to expand...

Well in this case I am gonna have to lean on the direction you're going it really does look like that core is just trying to eat itself. Sorry for opti is sick maybe it will get better with some benchrest.

Your CPU is de-lidded?

How did you get around having the wrong-sized spacers for your heatsink ...?

I just used bolts on my delidded 939 chips