Jump to content

Question about Expert and dying CPUs


Recommended Posts

i'd venture to guess that 99.99% of dead cpus were overclocked and the voltages way over spec...correct me if i'm wrong. if u overclock dont blame the board for fried components. the expert allows one to overclock and overvolt many diffrent components WAY over spec. just because u can doesnt mean u should. alittle common sense on the users part is required to avoid a fried cpu. thats my opinion :)

 

My CPu was at 1.52 volts under one of the best watercooling systems money can buy. It loaded in the 30's. People run Opteron 146's, especially CABNE, my stepping, at way higher volts. Hell, 1.52v is barely above stock, which is 1.4x on mine. And it's WAY under the 15% golden rule. Nothing was out of spec whatsoever besides my CPU, which was barely.

 

Besides, my chip wasn't overclocked or overvolted when it died. Cmos was set to default.

Share this post


Link to post
Share on other sites

  • Replies 106
  • Created
  • Last Reply

Top Posters In This Topic

You want to send all the pieces off for laboratory examination; it could probably be arranged.

 

you can't scream OMG ITS PSU at the 9lb monster in my machine.
I don't recall screaming?

 

But you are so sure it did and want to open forum challenge somebody to prove the board did not kill your cpu and I would just as easily ask you to prove it did and then when all that gets thru being beat up and around; only some very good forensic engineers might possibly be able to say for sure. So since you know it cannot be just looked at by you or me or anyone else and determine what was first; the chicken or the egg, it is nearly a mute point.

 

What though could be far worse in all this type of posting is if at some time those with money just plain got tired of defaming without proof and went to law about it. See it is one thing for you to say you believe your expert caused the demise of your cpu, but it is entirely a different matter when you make statements that rashly say that the board 'kills' cpu's. See you nor I know what happened. You presume by circumstance what happened. I have by example that my AA0 and AB0 have not killed any of the many cpus I gave them chance to do so with.

 

So the truth is you have one that does not. You have ordered another since one of yours is not dead and you presume it is a revision only that kills and in that I have not seen proof. So there you go; it does not have answer from conversation and surmising. That is only heresay without you having proof. I am not looking for proof since I have no dead cpu or Expert or Venus of any revision.

 

Chicken little said the sky was falling but it never fell. If you have dead Expert board and you want to have it rma'd am sure they will do it. If you have retail cpu, am sure AMD will replace it if you have not removed the IHS on it as well. And that is how it all goes. The sky is not falling and neither have you proven your statement that the Expert killed your cpu nor do you or I have the time and I know I don't have the skills (you may have such skills) for indepth testing at a lab that might say what came first the chicken or the egg.

 

RGone...

Share this post


Link to post
Share on other sites

You want to send all the pieces off for laboratory examination; it could probably be arranged.

 

Tell me where and I will.

 

 

I don't recall screaming?

 

Sorry, wasn't directed at you in general. Just upset how whenever someone says they have an issue with the board its "always" the psu's fault. More upset with the general lack of knowledge that a high 12v single rail psu is ideal with the DFI.

 

 

But you are so sure it did and want to open forum challenge somebody to prove the board did not kill your cpu and I would just as easily ask you to prove it did

 

While I cannot prove beyond a reasonable doubt, I can certainly prove through a preponderance of evidence.

 

Before I had the expert, everything in my system is known good. It had been running for months. I am fanatical about my power regulation and cleanliness, as well as airflow and temps. The power regulation for my system goes as follows:

 

 

Wall -> APC UPS -> Monster HTS3500 Mk III AC conditioner -> Zippy 700w psu with 1% voltage regulation. As you probably know, Zippy is a psu manufacturer with higher quality than even pc power and cooling. They originate in the server market (Zippy/Emacs). So the first thing that everyone looks at when diagnosing problems, the power, was not a factor. Voltages measured with a multimeter were 12.15, 3.35, and 5.15. They varied .001 under load.

 

What is the next thing that people look at? Temps and voltage. Nothing on the motherboard was running out of spec except the chip, which was a matter of percentage points. My chipset never even saw 40C, same with the chip. I had excellent airflow over the ram and all aspects of the board, 2x 120mm and 2x 92mm fans.

 

Shorts? None.

 

My ram was mushkin redline xp4000. It is also known good, and has never been run out of spec.

 

 

On the night that I put the board in, every single componet was verified good. The only componet that changed is the motherboard. The only one. Three days later, the chip is dead, as is the motherboard.

 

 

Knowing that the power did not kill the chip and the board, knowing that shorts didnt, knowing the temperatures didnt, knowing the static didnt (wore esd bracelet entier time, plus it worked for three days), knowing that shorts didnt, knowing that everything except the motherboard is verified good, I'd like to heard your suggestion for what killed my chip. After all, you're the certified tech.

 

 

 

and then when all that gets thru being beat up and around; only some very good forensic engineers might possibly be able to say for sure. So since you know it cannot be just looked at by you or me or anyone else and determine what was first; the chicken or the egg, it is nearly a mute point.

 

First, I highly doubt forensic engineers would be needed, as there are no corpses laying on my motherboard....

 

Second, while the chicken and the egg is certainly an interesting paradox, it does not apply. It's not a "what killed what first" kind of thing, chips do not kill motherboards unless they are faulty to begin with, not to mention the fact it worked fine for months previously. so it's not a moot point.

 

 

What though could be far worse in all this type of posting is if at some time those with money just plain got tired of defaming without proof and went to law about it.

 

Good suggestion, I'll consider it.

 

See it is one thing for you to say you believe your expert caused the demise of your cpu, but it is entirely a different matter when you make statements that rashly say that the board 'kills' cpu's. See you nor I know what happened. You presume by circumstance what happened. I have by example that my AA0 and AB0 have not killed any of the many cpus I gave them chance to do so with.

 

in the court of civil law, circumstantial evidence is all you need. While you state that by following the same logic I have, the fact that you have boards which have not killed cpu's proves they don't, you are again incorrect. The law of probability is on my side; componets fail and fail spectacularly sometimes.

 

 

So the truth is you have one that does not.

 

I also have a IBM deathstar that still runs. You gonna put it in your machine?

 

 

You have ordered another since one of yours is not dead and you presume it is a revision only that kills and in that I have not seen proof. So there you go; it does not have answer from conversation and surmising. That is only heresay without you having proof. I am not looking for proof since I have no dead cpu or Expert or Venus of any revision.

 

If you want "proof", however circumstantial it might be, visit another forum or google DFI dead cpu. I suppose I can understand you genuinely thinking there isn't an issue; it's hard to find proof when so many threads are deleted.

 

 

Thanks for the thoughtful reply, however, I'm still waiting on tech support.

Share this post


Link to post
Share on other sites

It's not secret that certain revisions rock and certain ones kill boards. I have ordered another one to tide me over while I rma it; I hope its a good one.

 

Why is everybody in denial?

 

If its not secret why are there threads every week asking the same question over and over again? Our answet is always the same - don't be an idiot with your gear and you should be fine. Funny how I haven't seen any of those people come back claiming their cpu is toast.

 

If you would be so kind as to let us know what these revisions are that eat cpus, that'd be swell. :rolleyes:

 

And by the way, if/when there is actual proof (other than random people claiming they are master overclockers and this board eats cpus like delicious chocolate covered peanut clusters dipped in molten butter) then we'll believe it. Quite honestly I've seen maybe 3 people claim this has happened, so if you'd provide links to this 'proof' we'll certainly take a look at it. :)

 

One more thing, just because you have a good psu doesn't mean that its not the problem. You can't just assume that a 700 name-brand power supply won't fail, even PC P&C 1000 watt models fail (just not often). Just throwin that out there. Chances are you're right, however it doesn't hurt to swap it out now does it?

Share this post


Link to post
Share on other sites

@computerpro3

 

On your suggestion I checked the XS forum about your post.

 

http://www.xtremesystems.org/forums/showpo...642&postcount=1

 

http://www.xtremesystems.org/forums/showpo...80&postcount=15

 

http://www.xtremesystems.org/forums/showpo...34&postcount=20

 

You run your CPU at 3.1GHz then blow the processor up.

 

You come over to DFI-Street to inflict some PWNage.

 

From your first post at XS...

"The machine was running for 3 days PERFECTLY, at 3.1ghz with 33C under load @ 1.53v. Yesterday, I rebooted it, and it hung at the post screen."

From your post here at DFI-Street...

Nothing was out of spec whatsoever besides my CPU, which was barely. Besides, my chip wasn't overclocked or overvolted when it died. Cmos was set to default.

 

Shame on you! There's no need to lie about your setup unless you're trying to hide something.

 

I think you just lost all credibility with this forum and perhaps the XS forum if they see this post.

 

BTW This is called PWNage!

Share this post


Link to post
Share on other sites

@computerpro3

 

On your suggestion I checked the XS forum about your post.

 

http://www.xtremesystems.org/forums/showpo...642&postcount=1

 

http://www.xtremesystems.org/forums/showpo...80&postcount=15

 

http://www.xtremesystems.org/forums/showpo...34&postcount=20

 

You run your CPU at 3.1GHz then blow the processor up.

 

You come over to DFI-Street to inflict some PWNage.

 

From your first post at XS...

 

From your post here at DFI-Street...

 

 

Shame on you! There's no need to lie about your setup unless you're trying to hide something.

 

I think you just lost all credibility with this forum and perhaps the XS forum if they see this post.

 

BTW This is called PWNage!

 

How stupid can you possibly be? I stated clearly in both forums that I was running @ 3.1ghz, at 1.52-1.53v. I was also correct in stating that the only thing running out of spec was the cpu, which was barely running out of spec; 1.52v is less than 10%. A mildly overclocked opteron running @ 3.1ghz, 1.52v simply does not blow the motherboard up. If it did you guys would have more problems on your hands than experts eating cpu's.

 

After all, perhaps an AMD Tech said it best; NOT A SINGLE DFI NF4 IS ON THEIR APPROVED LIST. For a reason.

 

I was also correct when I said my chip wasn't overclocked or overvolted when it died. Don't you konw what Checksum error - Default Values loaded means?

 

As for learnign to read my posts more carefully, Maybe these guys can help you out

 

www.hookedonphonics.com

 

And as for credibility; perhaps I should quote the other posts from the many threads saying "DFI Street guys generally don't know about computers?"

 

And btw, heres the link to the whole thread for anyone reading, so they can see other peoples chips that got eaten as well as the funny post that says

 

Although those mods in DFI-street generally don't really know about computer, plus I don't really like their attitude

 

Not to mention the PM I recieved:

 

You are certainly correct 'bout DFI Street... The mod eXRoadie is a ing :b anana:. RGone is really an OK guy but has a lot of friends here so be aware of that as this thread goes on....

 

I can tell you from experience that there is no amount of posting and debate that will get you anywhere at DFI but get you banned. It's not right, it's not fair, but there's no way to change it.

 

I wouldn’t even attempt to get DFI to help with the cost of a smoked cpu, they’re in denial. They have yet to step up to plate since these boards were released and acknowledge that there has ever been a single problem with this board. That’s their decision and it fine with me. Additionally their arrogance defies reason, any problem someone experiences with their product gets tagged as user error or ignorance. If you go to their support site, DFI-Street, it’s run with the very same arrogant fricking attitude.

 

well i talked to dfi today on compensation for my 3200 that the expert board took out.their attitude sucks,must have benn something i did setting up voltages is their attitude.according to them there is no problem with this board doing this.i told them to read the forums and the reply was there was alot of people that dont know how to set up dfi boards correctly.im an A+ certified tech,make my living building computers and how could this happen on auto settings.had to be user error.BS user error.

 

IMHO, DFI did have an answer and a fix for the Expert....it's called the Venus.

 

Make no mistake about it... no way in hell DFI would ever disclosed or admit what went wrong with the Expert, if those issue can be easily overcome and fix with new bios, they would of done it already. For those who don't have issues with the Expert, consider yourself very lucky and let's hope everything still hold true in couple months down the road.

 

So many of the best overclockers in the world have had chips killed by the expert; everyone from Pedro Rocha, to The Stilt, to Hardcoreclocker, etc, etc. They simply don't make user error.

 

I have a feeling that the quotes above are going to be correct; I will be banned and this thread will cease to exist (but I do have jpgs). Yet another problem swept under the rug. You have actually lost most forums respect a long time ago, while I, on the other hand, continue to write sticky after sticky and am considered a go-to guy for information regarding psu's.

 

Please keep in mind that people vote with their wallets.

 

Now that's pwnage.

Share this post


Link to post
Share on other sites

How stupid can you possibly be? I stated clearly in both forums that I was running @ 3.1ghz, at 1.52-1.53v. I was also correct in stating that the only thing running out of spec was the cpu, which was barely running out of spec; 1.52v is less than 10%. A mildly overclocked opteron running @ 3.1ghz, 1.52v simply does not blow the motherboard up.

Dont you have to overclock your htt to 310mhz to get your cpu to 3.1gig ?

 

to me that sounds like your oc'n more then just the cpu.. but what the heck would I know..

Share this post


Link to post
Share on other sites

Dont you have to overclock your htt to 310mhz to get your cpu to 3.1gig ?

 

to me that sounds like your oc'n more then just the cpu.. but what the heck would I know..

 

Come on, you know better than that. The LDT multiplier was turned down to 3x. Technically, that's underclocked. Stock voltages on LDT and chipset too. I doubt the machine would even boot if it was left on 5x 310.....

Share this post


Link to post
Share on other sites

@computerpro3

 

Trust me! If I post a personal attack or make a mistake, RGone or Angry_Games will step in and correct me. I have no allusions that I'm perfect but I have no reason to attack you personally. I only see the facts as given and respond to them. I quote for clarity and provide links when appropriate.

 

http://www.xtremesystems.org/forums/showpo...80&postcount=15

 

It appears the chip is dead. Here's a link, I'm about to pwn the mods on DFI street that claim they don't kill chips.

 

http://www.dfi-street.com/forum/show...373#post441373

 

Grab some popcorn and watch me pwn his butt, I really don't care if I get banned. Lets see how long it takes for them to hide my issue.

You're lucky I have a sense of humor about such things.

 

Stop wasting hard drive space with screen caps. There's no way in hell I will let this thread be deleted or moved. I want to make sure everyone can see it for what it's worth.

 

I will also make sure you don't get banned unless you really flame a forum member or post vulgar language. This thread is priceless so make sure you don't get the thread deleted over at XS so the links stay live.

 

A mildly overclocked opteron running @ 3.1ghz, 1.52v simply does not blow the motherboard up.

1.1GHZ above stock is mild??? If my math is correct that's about a 55% overclock.

 

The Opteron 146 has a stock speed of 2.0GHZ. Anything above the stock speed is overclocking. There's simply no way around it. Or were you lying in your first post at XS and your followup post on this forum

 

And as for credibility; perhaps I should quote the other posts from the many threads saying "DFI Street guys generally don't know about computers?"

Wow, a shotgun blast personal attack on all of the forum members that contribute here.

 

So here's the big question that I hope you will answer honestly. If you hold DFI-Street in such low regard? If you believe that we have no idea what we're doing? If you think there is a problem with the Expert board that DFI is hiding?

 

Why in the world would you even take the time to post something at DFI-Street?

 

Be honest because you've already shown your cards! hint!!! See the top of this post!

Share this post


Link to post
Share on other sites

It could be anything... There are tones of things you cant control that could of took your system down. Surge. Hidden short in system. Overtighten cpu. Static. Its possible you could of done everthing right and everthing went wrong. Thats why you need proof... Just because its dead and you hear a rummor that the board kill's cpu's you jump to conclusions. Even if the board was eating cpu's it would be possible that situation is unrelated. Anybody with expierence with computers would know that. And lots of great OC'rs read both XS and DFI forums.

Share this post


Link to post
Share on other sites

Come on, you know better than that. The LDT multiplier was turned down to 3x. Technically, that's underclocked. Stock voltages on LDT and chipset too. I doubt the machine would even boot if it was left on 5x 310.....

hold on hold on hold on here lol

 

lets see if this makes sense to everyone

 

so

 

you have a cpu that does NOT do 3.1Ghz by default, because the fastest cpu AMD makes is what...2.6Ghz? 2.8Ghz?

 

and your cpu is only a 2.0Ghz cpu by default, correct?

 

and then earlier you say this:

 

Do a search on xtremesystems, the most advanced forum on the net (common knowledge).

 

which we'll assume you are one of those ADVANCED users that is much more advanced than anyone here or any other forum on the net

 

agree so far?

 

ok

 

so

 

lets break down your bits here...

 

you have a cpu clocked to 3.1Ghz, which IS overclocked, yet you THINK it is NOT overclocked because...you lowered the LDT multiplier? An internal link multiplier that has zero to do with manipulating actual cpu speed?

 

Hold up

 

I thought you were 'Xtreme'

 

or maybe you never said that, but you sounded there for a minute like you were telling us you knew about this stuff...but then you say something as incorrect as this? (about the LDT multi being only 3 so "TECHNICALLY" you are underclocked?)

 

I thought you KNEW about this stuff?

 

But what you have shown is that you know zero.

 

 

ah ah ah

 

before you say anything else

 

I advise you to go ask your Xtreme buddies if lowering the LDT multiplier down to 3x TECHNICALLY is not overclocking the cpu...even though your cpu is an Opteron 146, which Rgone and I both have, and they are 2Ghz cpu's by default...200x10.

 

 

 

yes yes...I see...you don't really know what you are talking about, but if you sling enough bits and rumors around and talk like you have a clue as to what you are saying and that you are correct, and you are an expert and we know nothing, others will listen to you and believe you and not buy whatever it is that you bought then destroyed on your own.

 

We see users like you every day...you burn up your rig by overclocking a cpu to 3Ghz+, and a lot of them are just like you...they know zero (because lets be honest here, anyone who claims that they are knowledgeable about overclocking then say something so ignorant as "lowering the LDT multipler to 3x means I am not overclocking" yet you overclocked the cpu from 2Ghz default speed to 3.1Ghz...is about the worst possible type of computer owner, and even worse than that, the worst type of internet/forum user....you are one who spreads lies and rumors without any proof to back it up, and I suppose even worse, you claim to have proof, yet your proof is ignorance and your own fairy-tale made-up overclocking knowledge that is so far from reality as to be less believable than...anything the best sci-fi/fantasy writer could come up with).

 

 

 

 

now

 

 

 

I advise you, seriously, before you post one more word at this forum, to go off and be a good citizen and do some real homework about cpu's and memory and more specifically, overclocking (hey! there's a comprehensive Overclocking Guide in the AMD- Overclocking section of these forums!).

 

 

Because seriously...we have zero tolerance for rumor mongerers like you who make our jobs and our lives difficult.

 

It takes 100 posts to eliminate 1 single rumor post

 

and rgone and I personally do not have time to make 100 posts to correct your ignorance.

 

 

 

 

 

 

now

 

 

 

 

 

to the original poster of this thread

 

 

 

 

you need to read Rgone's posts.

 

I have 3 Expert boards myself, and have had 10+ of them come through my hands in the last 3 months alone.

 

Not a single one of them has killed a cpu unless the user did it himself.

 

 

now, back when these boards were first coming out, someone got ahold of a beta board, a hand-made board that was still being evaluated by Oskar the designer of the board, and the person who got ahold of it was pretty dumb in the sense that the instant he plugged a psu into it, it was all flashy lights etc...which to any intelligent person was a good sign that there was a problem right away.

 

But this person wasn't fazed, and he put a very expensive 4800+ X2 cpu in it

 

and blammo it fried it

 

NOW...NORMAL PERSONS, or persons of halfway intelligence, would have stopped right there and said "oh hell no, this board is headed back to the designer and he's gonna get told that this board has some serious issues"

 

but

 

did this person do such a thing?

 

tsk tsk no

 

what did he do?

 

he immediately dropped an FX-57 into it

 

blammo, fried FX-57

 

 

NOW...ONCE AGAIN...anyone with a brain would have stopped right there

 

 

but did he?

 

no

 

once again, he dropped yet ANOTHER 4800+ X2 cpu into it

 

and once again, it killed it

 

 

 

 

 

 

now

 

 

I ask you seriously

 

 

as this is one of TWO cases I know of that a board killed a cpu WITHOUT any user fault (other than lack of intelligence, as the other case I know of involved YET ANOTHER pre-release beta board), I ask you...would these two cases say to you "OMG THESE BOARDZ KILL CPU'S!!!"?

 

 

No.

 

 

now also keep in mind...we've heard plenty of users cry like women about how they just KNOW Expert boards have killed cpu's.

 

And I tell you honestly, Rgone and I work the OFFICIAL DFI tech support email, as well as this forum, as well as plenty of internal communication with DFI.

 

We've seen no cases where we can truly say that the boards have killed cpu's without the user doing some odd thing (almost always overclocking).

 

Rgone might be able to remember a case or two if there was one, but myself, watching these forums, I myself cannot recall a single event (and if anyone can, please refresh my memory).

 

 

how many Expert boards are there in the wild now?

 

10,000? More? Less?

 

how many Expert boards have killed cpu's with no help from the user?

 

1? 2? 0?

 

what does this tell you?

 

 

It tells you that because one very 'famous' guy at another forum wasted 3 $1,000 cpu's (they were about $1,000 each then) on a single Expert board that WAS NOT FIT FOR ANYONE, IT WAS A BETA PRE-RELEASE TEST BOARD, all of the sudden everyone and their dog has an Expert board that kills cpu's.

 

 

 

It is rumor repetition at it's worst, which we call rumor mongering.

 

 

And what I said earlier is as true as toasted toads...it takes ONE HUNDRED posts from me and/or Rgone to squash a SINGLE rumor post that has no merit.

 

 

that, my friends, is exactly how powerful a single rumor is.

 

 

Now, for something a little bit lighter (but still on topic), please read the excellent post about rumors from our friend soundx98:

http://www.dfi-street.com/forum/showpost.p...389&postcount=1

Share this post


Link to post
Share on other sites

The fact remains and is true for all who walk this path. There will be failures and that is why manufacturers created SPECIFICATIONS. If you choose to operate anything outside of specified parameters then you can blame none other than yourself. Suck it up move on and don't cry to me because I know and understand this

Share this post


Link to post
Share on other sites

Guest
This topic is now closed to further replies.
 Share


×
×
  • Create New...