My CPU just bit the dust :(

[Deinonych]

Today when I was playing Bioshock, my system hard crashed in the middle of the game (fortunately, I had recently saved my game). Subsequent attempts to get my machine back up were unsuccessful. The fans would spin up, but the system wouldn't POST. After swapping out the RAM and the video card, I popped a spare C2D proc in, and sure enough, the system fired right back up.

It seems unusual for a CPU to spontaneously fail like that. My temps have been well within normal range (idle ~40C, load ~50C), and I have good airflow in my case. Any thoughts on what might have caused it to fail suddenly? I'm going to call Intel to RMA the proc, but I want to make sure it doesn't happen again. I can't imagine Bioshock was that hard on it.

Proc is a C2D E6700, at stock speeds.

 

[boglwe]

Wow, I would really like to hear an update on this. Sorry for your loss.

 

[aigomorla]

Originally posted by: Deinonych
Today when I was playing Bioshock, my system hard crashed in the middle of the game (fortunately, I had recently saved my game). Subsequent attempts to get my machine back up were unsuccessful. The fans would spin up, but the system wouldn't POST. After swapping out the RAM and the video card, I popped a spare C2D proc in, and sure enough, the system fired right back up.

It seems unusual for a CPU to spontaneously fail like that. My temps have been well within normal range (idle ~40C, load ~50C), and I have good airflow in my case. Any thoughts on what might have caused it to fail suddenly? I'm going to call Intel to RMA the proc, but I want to make sure it doesn't happen again. I can't imagine Bioshock was that hard on it.

Proc is a C2D E6700, at stock speeds.

list complete hardware.

Mobo, Ram, CPU, (one thats working and not working) PSU.

Otherwise no one can help you debug this.

EDIT: missed the link.

CPU: Intel Core 2 Duo E6700 2667MHz


Motherboard: Asus P5N32-E SLI Plus


Memory: 2048 MB of Crucial Ballistix PC2 6400 DDR2


Video Card: eVGA GeForce 8800GTX


Hard Drive: Western Digital Raptor WD740ADFD 74.0 GB @ 10000 RPMS


Additional Hard Drives: Western Digital Caviar WD5000AAKS 500.0 GB @ 7200RPMS


Monitor: ViewSonic VX2025wm


DVD R/W: Plextor PX-716A


Storage Interface: IDE,SATA,USB2


Sound Card: Creative SB X-Fi XtremeGamer


Speakers: Klipsch Promedia 2.1


Internet Connection: Road Runner Cable @ 4 Mbps


Operating System: Windows XP Professional


Other Components: · Antec Performance One P180B case case
· SeaSonic M12-700 700W PSU
· Atech Flash 9-in-1 universal card reader
· Microsoft Multimedia Keyboard
· Wacom Graphire3 6x8 USB tablet
· Logitech MX518 Gaming Mouse
· CH Products CombatStick USB
· CH Products Pro Throttle USB
· Belkin Nostromo n50
· Canon i950 Photo Printer


im assuming this is your current config. Have you tried poping the CPU in another board? Also what voltage were you running your ram at? You are aware factory default recomended voltage for those ram is too much. Your ram should be no higher then 2.1V for longivity. 2.2V and your asking for it to go out 6-8 months down the road.

 

[Deinonych]

I don't have another LGA775 board to pop the proc in, unfortunately. I am running the RAM at 2.2V - I had stability problems (random lockups, BSODs etc.) running it lower than that. RAM voltage shouldn't affect the proc, though, should it?

[edit] Just to clarify, I had to manually set the timings (4-4-4-12) on my motherboard to get my RAM stable. I haven't tried different voltages. System was unstable with the timings set as "auto."

 

[genec57]

I would have a hard time not believing the factory recommended voltage at 2.2. If they don't know how to spec out their own ram who does? I even spoke with Crucial support a few months ago and was told to set the vdimm at 2.2. That is good enough for me.

 

[chaudx]

The Crucial Ballistix RAM I bought from Newegg said 2.2V was the voltage, and Crucial says that 2.2 is the voltage, so what is wrong with running DDR2 at 2.2V

 

[Markfw]

Me too. I have been running at 2.2 for over a year, no problem (Crucial Ballistic). It says 2.2v, and I set it at that.

 

[needalife]

this is really weird...

 

[Zenoth]

Maybe a short-circuit in the Power Supply.

Never mind, it does power-on ? Hummm ... strange. Still, perhaps insufficient juice from the Power Supply in demanding games who knows ? Not sure, it's the first time I read something like that.

 

[Keysplayr]

Is BIOS flashed to latest? What BIOS version do you currently have?

 

[genec57]

In my experience it is extremely unlikely that the CPU would just fail like that. I think the most likely problem is the mobo - RMA that sucker.

 

[Deinonych]

I'm running BIOS version 0801. There's a newer release (0805) that just came out a few days ago, but I haven't flashed to it yet. I suppose it's possible that the motherboard is the culprit, but I'd like to be sure before I RMA it. Being without a motherboard for several weeks doesn't set well with me unless I'm absolutely sure it was the cause of the problem.

My system has been running fine since I built it 6 months ago. I've had a few video driver issues, but the proc and memory have been trouble free. I ran Orthos for a couple hours a few months back without incident. Very strange indeed.

 

[aigomorla]

Originally posted by: Markfw900
Me too. I have been running at 2.2 for over a year, no problem (Crucial Ballistic). It says 2.2v, and I set it at that.

i am stating something that has happened to me personally, as well as a few of my friends.

We all ran tracers at 2.2V these are the ram sticks which has leds.

In about 6-8 months time, one of my sticks burnt out. Crucial was nice enough to replace it, but the fact remains that these ram sticks are originally intended to be ran at 1.9V which the factory overclocked to 2.2V for the higher bandwith.

If you do some searches on Xtremesystems, you'll see i am not the only one.


Now after my friend and i have both had our loses, we dont keep the ram any higher then 2.1V


I would run a memtest on your ram. See if it passes 2-3 passes. I dont see how the CPU would just die like that, unless you were putting a lot of stress on it. And i dont think your PSU is the culprit either. Thats one hell of a great psu ur running.

 

[nyker96]

You sure it's the CPU? maybe other component is fried!

 

[Deinonych]

Originally posted by: nyker96
You sure it's the CPU? maybe other component is fried!

You'll note from my original post that I put in another C2D (an E4500) and it booted up (and is running) perfectly. When I put the E6700 back in, it wouldn't boot. So, I'm pretty sure the E6700 is dead.

Originally posted by: aigomorla
I would run a memtest on your ram. See if it passes 2-3 passes. I dont see how the CPU would just die like that, unless you were putting a lot of stress on it. And i dont think your PSU is the culprit either. Thats one hell of a great psu ur running.

I ran Memtest for several passes last night. No errors whatsoever. I'm now seriously thinking that the motherboard might have caused the CPU to fail. I might just RMA it just to be sure. Hell, maybe I'll just move up to a P35 board and use my current one in another system (after RMAing it).

[edit] If I wanted to step down the voltage on the RAM, would I not also need to change the timings? If so, what timings should I set if I went down to 1.8v?

 

[bryanW1995]

if you get a new mobo, check out the x38's. I almost wish mine would fail...almost...ok, not really.

 

[aigomorla]

Originally posted by: Deinonych

Originally posted by: nyker96
You sure it's the CPU? maybe other component is fried!

You'll note from my original post that I put in another C2D (an E4500) and it booted up (and is running) perfectly. When I put the E6700 back in, it wouldn't boot. So, I'm pretty sure the E6700 is dead.

Originally posted by: aigomorla
I would run a memtest on your ram. See if it passes 2-3 passes. I dont see how the CPU would just die like that, unless you were putting a lot of stress on it. And i dont think your PSU is the culprit either. Thats one hell of a great psu ur running.

I ran Memtest for several passes last night. No errors whatsoever. I'm now seriously thinking that the motherboard might have caused the CPU to fail. I might just RMA it just to be sure. Hell, maybe I'll just move up to a P35 board and use my current one in another system (after RMAing it).

[edit] If I wanted to step down the voltage on the RAM, would I not also need to change the timings? If so, what timings should I set if I went down to 1.8v?

At 2.05V my ram are at 830mhz. The timings for them are 4-4-4-8 2T. This is keeping my fsb @ 415 with my ram.

I believe most ballastix can do this at that voltage. Make sure you have an active fan on it it. I used a 80mm Fan and attached rubber bands to the corners. And i hang it over the ram sticks. Works absolutely great, and it was dirt cheap.

 

[Sylvanas]

Ouch sorry to hear it if it is indeed a fried C2D, it could possibly be a power supply issue but its odd it would boot fine with your alternate CPU.

 

[Deinonych]

Originally posted by: aigomorla
At 2.05V my ram are at 830mhz. The timings for them are 4-4-4-8 2T. This is keeping my fsb @ 415 with my ram.

I believe most ballastix can do this at that voltage. Make sure you have an active fan on it it. I used a 80mm Fan and attached rubber bands to the corners. And i hang it over the ram sticks. Works absolutely great, and it was dirt cheap.

OK, thanks. What about if I didn't want to overclock them (I'm not overclocking them now)? I don't really care about overclocking -- stability and quietness are more important to me than performance.

 

[ultra laser]

Then run them at their default settings (400, 4-4-4-12) and see how low you can drop the voltage before you start to get memtest errors. When you start getting errors, raise the voltage a bit and see if it can run for a couple of hours without errors. If it can, I would use that voltage. There's no reason to give it more voltage than it needs - especially if you're running at stock speeds. Hell, I run mine at 450mhz at 2.1 with no problems.

 

[Deinonych]

Originally posted by: ultra laser
Then run them at their default settings (400, 4-4-4-12) and see how low you can drop the voltage before you start to get memtest errors. When you start getting errors, raise the voltage a bit and see if it can run for a couple of hours without errors. If it can, I would use that voltage. There's no reason to give it more voltage than it needs - especially if you're running at stock speeds. Hell, I run mine at 450mhz at 2.1 with no problems.

Thanks! I'll give that a try.

 

[aigomorla]

Originally posted by: ultra laser
Then run them at their default settings (400, 4-4-4-12) and see how low you can drop the voltage before you start to get memtest errors. When you start getting errors, raise the voltage a bit and see if it can run for a couple of hours without errors. If it can, I would use that voltage. There's no reason to give it more voltage than it needs - especially if you're running at stock speeds. Hell, I run mine at 450mhz at 2.1 with no problems.

+1 this is great advice.

 

[Deinonych]

So, I was able to set the voltage to "auto" in the BIOS, which was 1.85v according to NVIDIA monitor. I ran Memtest86+ for a few hours this afternoon, and didn't get any errors. So, it looks like I'm good there. Back to the original topic, do you all think the high voltage on my RAM contributed to the processor failure? I spoke to an Intel tech today, and he didn't really have any good ideas why the proc failed.

I'm going to pick up a PSU tester this week to make sure I don't have problems there.

 

[DerComissar]

It's not your hardware, it's that POS Bioshock game. The demo crashed like that on me repeatedly. Read somewhere that it happened to another gamer in Arkansas, after it took out his Q6800, it killed his dog and boiled his goldfish:evil:

 

[Saiyukimot]

Just wonderin, but did you install the latest drivers for your graphics card before playing? People have had problems with not using it...


Dunno how that would affect the hardware mind.