Defective Processor

[EvilAlchemist]

So, I decided to "upgrade" two of my borgs a few days ago and bought two q6600 (I know, I have enough) so i got them in today.

First one did great, running at 3.0 GHz .... 0 problems.

The second one, nadah .... First boot at 3.0 GHz .. i got segfault errors when F@H ran for 5 min.

At first, i thought memory, switched memory, and tried again ... no luck, same error

Then I backed down to stock 2.4, segfault again when F@H ran for 5 min.

Then i switched the CPU and put it in another borg .... segfault ....

I must have gotten a defective CPU ... Am I correct in my conclusion?

 

[Wharf_Rat]

If you send me all your cpu's , Ill test them out for you...

I might need them for a month or more though.

Sounds like you might of found the issue. The other machine you speak of folded ok prior to this chip right?

 

[Connundrum1]

Take the working Q6600 and put it in the rig thats having problems. If it runs fine then RMA the cpu.

 

[EvilAlchemist]

Take the working Q6600 and put it in the rig thats having problems. If it runs fine then RMA the cpu.

That is the only thing i have not done yet .. but i am not sure it is necessary.
If the Q6600 was Incompatible, it would not work in any of my systems.
The fact that the "broken" one failed in two identical machines kinda raises a flag to me.

I sent the "broken" on back so in a week we will see if this solves the problem.

The other machine you speak of folded ok prior to this chip right?

Yes, Folded great before and is folding again on the old chip.

 

[Connundrum1]

Your ramping up big time and your already in my threat list, looks like i'm gonna get a flyby sometime in the near future. hope that new chip works! i hate RMA'ing stuff.

 

[BillR]

That is the only thing i have not done yet .. but i am not sure it is necessary.
If the Q6600 was Incompatible, it would not work in any of my systems.
The fact that the "broken" one failed in two identical machines kinda raises a flag to me.

I sent the "broken" on back so in a week we will see if this solves the problem.

Yes, Folded great before and is folding again on the old chip.

You didn’t mention voltage. Sometimes it takes a slight bump in VCORE to get them where you want.

 

[EvilAlchemist]

You didn’t mention voltage. Sometimes it takes a slight bump in VCORE to get them where you want.

I went from stock 1.262 (i think) all the way to 1.4v no effect

See that is what got me too .. both were G0 stepping.
The first one ran great all on stock ... that is whats so confusing about all this

everytime i would get the segfault error ... or I would get the end unit early error before 1% .... something waws not right ....

anyone else have ideas or had this happen?

 

[Celerator]

I like the idea of switching the CPUs as a final test.
But, yes it sure looks like a bad CPU.

Test, then RMA so you can get that boxen folding again.

 

[Charlie_D]

Not yet, and I really hope I don't; my second Q6600 is currently in the middle of a software install, but I lapped it before I even bothered installing the thing

At least the 120mm fan fits on the heatsink now...

 

[EvilAlchemist]

I like the idea of switching the CPUs as a final test.
But, yes it sure looks like a bad CPU.

i plan to do this in the morning .....

Think I have knocked out everything else ....

 

[Pocatello]

.... segfault ....

I must have gotten a defective CPU ... Am I correct in my conclusion?

Yes.

It is Relic's fault!

 

[BillR]

You may or may not have a defective CPU.

As improbably and absurd as this might sound I just finished a build with my EVGA 680i with a new quad and I have either errors and or lockups. In fact I’d be ashamed to tell you how many VM’s I’ve installed in the past 48 hours only after running prime X 4 for 48 hours.

Keep in mind, this board has had a quad Q6600 G0 same as before for the past year with no issues.

Today, for no other reason for shits and grins I updated the bios which is an update for a later CPU.

After a complete flash and CMOS clear (on Nvidia boards you MUST pull the battery for at least 20 min or the date won’t revert and if the date didn’t revert you didn’t reset the whole CMOS).

It’s been an hour now and it’s still folding away. Did Intel change the core properties of the venerable old Q6600? We shall probably never know, but this worked for me.

Luck man

 

[EvilAlchemist]

Update: I moved the known good CPU to another rig .... everything seems fine.

I had to raise the volatage a little .. but all seems good.

So will just have to wait a week for the other one to come in....