Why do CPU's suddenly stop working?

[pkomma]

I was running my computer (AMD 1.4 tbird with MSI KT266a mobo) and I shut it down. When I turned it back on again, I got no bios screen, just fans turning. I crack it open, connect the diagnostics and I get 4 red lights of doom (means CPU cannot cycle). How on earth does my CPU go from working to non working within minutes. It had been working for 2 months fine, how does it suddently decide its time to die. Man, i'm frustrated, I hope to god this thing didn't suddenly decide to burn out. All fans/heatsink were untouched for last 2 months of operation. Any advice?

Update:
If the CPU is indeed burnt out (god damn i hate AMD) there's a good chance mobo is burnt out too. How would I go about testing it? Right now I can get diagnostic feed back (I assume good sign, though I think it doesn't neccesarily mean anything) and my network card lights are on. Any way to test other than sticking a new CPU in there?

 

[Rahminator]

Clear CMOS?

 

[pkomma]

Didn't help

 

[veryape]

Have you tried reseating everything? I know it sounds very routine but you need to start at the beginning.

 

[BrunoPuntzJones]

<< Your CPU is dead. It happens sometimes with lower quality CPUs. Not that all AMD CPUs are low quality, but some of them definately are. They burn out just like that. >>

i thought orcs and trolls were different?

 

[Vadatajs]

<<

<< Your CPU is dead. It happens sometimes with lower quality CPUs. Not that all AMD CPUs are low quality, but some of them definately are. They burn out just like that. >>

i thought orcs and trolls were different? >>



He has a valid point. No manufacturing process is 100% defect free. You probably got a bum cpu. Hopefully there is some sort of warranty.

 

[BrunoPuntzJones]

<< He has a valid point. No manufacturing process is 100% defect free. You probably got a bum cpu. Hopefully there is some sort of warranty. >>

understood, poor wording/poor decoding of the message on my part.

i took it to mean they were inferior cpus. he just meant that some runs are less reliable than others, misunderstanding on my part

 

[Vadatajs]

<<

<< He has a valid point. No manufacturing process is 100% defect free. You probably got a bum cpu. Hopefully there is some sort of warranty. >>

understood, poor wording/poor decoding of the message on my part.

i took it to mean they were inferior cpus. he just meant that some runs are less reliable than others, misunderstanding on my part >>



S'coo bro, I wasn't trying to flame you.

 

[John]

<< Why do CPU's suddenly stop working? >>



Sometimes there isn't a reason "why".

 

[Jgtdragon]

Its time to get a faster and newer cpu. Its faith.

 

[pkomma]

Dude, its not friggin fate when my CPU burns out like that. It was a retail CPU, so I think I can exchange it under warranty, but god damn, this is another 2-3 weeks (more 'cause of holiday season) that we can't be out of a computer. This was my Dad's computer and he's currently unemployed and constantly is in need of computer for his job search. God damn, this is an inferior product. For how much faster AMDs go, how on earth can you justify these chips just burning out like that. I was an avid AMD fan, but this is literally my 3rd AMD chip I gotta RMA. Though i'm not eating the cost, I'm sure that there are plenty of people who get stuck with these chips with 0 warranty. Honestly, mebbe Intels jacked prices reflect the need for multiple quantity of AMD chips.

 

[pkomma]

Reseating doesn't matter... this is a CPU clock iniliziation failure, which probably means dead chip.

 

[John]

<< was an avid AMD fan, but this is literally my 3rd AMD chip I gotta RMA. >>



You either have bad luck, or bad installation methods. Hardware can fail at ANY given time. I have seen retail packaged hardware be DOA, and I have seen others fail days, weeks, months, and even years down the road. Ignoranace is bliss sometimes.

 

[pkomma]

I'll be honest, I prolly f'd up on my first chip. Second one was DOA, and this 3rd one stopped operating after a month of decent usage. I dun know, I've never heard of people's CPU's dying after normal operation.

 

[Jgtdragon]

Don't want to start a flame war here, but when is the last time we hear an Intel cpu burn up?
You get what you paid for. Maybe Intel spent more money on their inspection process; therefore, high price on their cpu. I heard more about AMD cpus go bad than Intel.

 

[BeauJangles]

Enough with the Intel vs AMD thing. Don't bother with the "I don't mean to start a flame war here" because obviously you do. Just say that IN YOUR OPINION intel has a better product. Fine. No one should argue with that. It's like saying "no offense but" obviously what you are going to say is offensive, so why bother saying no offense?

grr...

Anyway, did you have any bridges connected? If they were penciled and the tracing wore off the CPU could stop POSTing. Just a thought, good luck bro!

 

[pkomma]

'er, no over clocking or anything... was a good ol' retail amd 1.4 with retail heatsink/fan stuck into mobo.

 

[ST4RCUTTER]

Are you overclocking these chips? I haven't had an AMD chip go bad ever... What was it running at (temp. wise) before it died?

edit: Just beat my post...

 

[pkomma]

If that plastic was left on it, it would have never survived as long as it did. As far as a temperature spike, I couldn't really tell you. The only time I checked temperature was when I was setting it up (month ago) and at the time it was at a reasonable temperature though I don't recall exactly what. I had plenty of fans in there to keep the system it self cool.

Even if a temperature spike occured, aren't these designed to function at temperatures higher than normal anyway? I know they get to 600 fahrenheit with heat sink off. With heatsink on/fan blowing it prolly was around 100ish. What on earth would drive the temperature up to a level where it would be destroyed? I assume it'd require 100ish or more degrees to burn out.

 

[MrHelpful]

The fan could have failed, the heatsink could have loosened in some way... Have you considered those possibilities?

 

[HaVoC]

pkomma

You might consider is that you have:

a) A bad power supply that is burning out the chips with too much voltage. It may seem fine but suddenly have a voltage spike. It's a slim possibility but nevertheless a possibility.

b) Bad A/C power...maybe just on the circuit you've plugged the computer into. Do you have GOOD quality surge protector?

c) Bad motherboard that is causing the chip to burn out.

Seems like any three of these are possibilities. I seriously doubt you got three bum retail chips in a row. (and you hinted the first one was probably user error)

 

[pkomma]

I don't wanna bother with the history, but, this 3rd chip was installed with a new motherboard and a new power supply. Its plugged into an APC battery/surge protector also. As far as fan/heatsink, they weren't touched and the fan is still turning away as it hangs while it tries to boot it self up.

 

[Erasmus-X]

I never in my life had a chip go bad on me under normal usage, AMD or Intel. I did have one DOA chip though, an OEM Celeron 466MHz PPGA from two years ago.

The majority of cases I've seen with dead AMD CPUs is inadequate cooling, improper installation methods, or both. Remember, the Athlon is a very hot-running CPU because of the current it requires. Not only do you need a good HSF for your CPU, but you also need a good central cooling system in your case. It can make quite a difference.

 

[Duvie]

I have built several intel and amd computers since 1998 and I have yet to have one die....

I think you may need to examine the other hardware as well...examine method of installation...watch those voltages and temperature when installing them with the amd chips...it could save a lot of time troubleshooting issues and rma'ing product.

 

[pkomma]

As far as fans:
Retail AMD fan on chip
One Slot exhaust fan
One 8 inch sunon exhaust fan
One 8 inch enlight intake fan

All are hummin away. This CPU was under very normal conditions. I had shut down it while it was running windows fine moments earlier. After I turned it back on it just hung. I don't see how a temperature flux or a voltage flux could occur in such a short time.