will 2volts kill my cpu?

[b0bd0le]

i hope not

[RichBa5tard]

i asume you've got a xp1700/1800+ tbred b? DUT3C or DLT3C.

With normal aircooling: yes
With high end aircooling or h&o: no, unless you run it 24/7
With phase change: it should last a pretty long while, but i wouldn't recommend it.

my 0.02$, i could be wrong. ;)

[BrainStorm]

Sooner or later it will. It's just a question of how long. And that's a question that only your CPU knows for sure. And it isn't going to tell you until it's too late.

But I wouldn't run it at that voltage for long unless you can afford to lose it.

[MCWB]

As RichBa5tard said, depends what cooling you've got. I've had mine at 2 V on good water and now Vapo for the past 3 months or so and it's still alive and kicking. :)

[RichBa5tard]

Originally posted by MCWB
As RichBa5tard said, depends what cooling you've got. I've had mine at 2 V on good water and now Vapo for the past 3 months or so and it's still alive and kicking. :)

True, but i think a XP1700+ DLT3C (1.5v) is much more sensitive to high voltage than a 2100+ (1.65v).


bobdole, if you got a good 92mm fan on your slk900: consider 2v safe for a quick bench, but not everyday use.

Then again, even if it only lives for a couple months, a 50$ cpu is quickly replaced. :)

[b0bd0le]

well, i got a panaflo thats puts out 45cfm somethign like that, is that decent

[felix88]

i wouldn't go over 2v, but with good cooling 2v is ok. pick up a Vantec Tornado and run it on high. :D

[SupaMan]

make sure you count for over/undervolting by the mobo

[TerroH8er]

Originally posted by RichBa5tard
True, but i think a XP1700+ DLT3C (1.5v) is much more sensitive to high voltage than a 2100+ (1.65v).


IMHO i don't really think so, although I could be wrong as I have been in the past. Since the 1700+s are pretty much downgraded 2600+ and 2800+s, I'd think that since you can run lower than spec'd speeds are lower voltages.. so it's basically like running a 26/800+ @ 1700+ speeds with lower voltages, thus meaning that since "originally" they were meant to run at 1.65v, and it's the same silicon as the DUT3Cs, they'd last the same. However, I don't think that it's been proven that DLT3Cs are just downgraded higher-spec'd DUT3Cs..

This post probably doesn't make sense cause it's 2AM here and I had a WHOLE LOT of mountain dew and energy drinks and it's seriously embarassed my judgement hehehe i can only dig into my brain so deep :stick:

[saaya]

hehehe

well my xp1700+ ocs to 2.1ghz with the stock voltage. im bruning it in @ that speed now with the voltage @ 2.1v.
and using the large coolermaster silent cooler it wont even go over 60°C.
i think that the high temps resulting the voltage will reduce the lifespan of your cpu, but dont know how much.

there are hundrets of people running their xps @2.2v with h20 or vapor cooling and they seem to run fine...

i NEVER heard of an xp that died from too much voltage...

[Ewok]

i think people still fail to realise that heat <60c will NOT kill a cpu (hell they are rated to 90c by AMD). What MIGHT kill the cpu is the voltage itself as a result of electro migration, which has absolutely nothing to do with temps or what coolig is being used.

if temps are <60c regardless of cooling, then the only thing thats going to killl the chip too soon is the voltage itself and you wont know that until its too late.

Ive not heard of any cases of the amd chips dying like the p4's do, it happens but not often and only at extreme voltages like 2.1+

[saaya]

well why do you think it happens if you have never heard of it?

...and i think YOU need to realise that you dont know that much about electro migration! ;)

All atoms are moving all the time, the greater the heat , the faster they move. now electro migration happens when the circuit paths are decomposing and two paths short each other or one path gets cut. This happens when the atoms are moving fast enough to break out of their connection with the rest of the atoms. the more current floats through the paths, the more electrons hit the atoms and heat them up.

the hotter the cpu, the faster the atoms are moving ,and the less electrons have to hit them to free the atoms out of their connection with each other, decomposing the paths. no matter how effective the cooling is, even @ -200°C single atoms will heat up enough to leave a path when so many electrons hit them at the same time, that there is no way of cooling them down fast enough to prevent them from migrating. BUT the better the cooling, the more current can a cpu take!

now the P4s current paths are decomposing so quickly, not because they are thinner or of less quality. every cpu has some hot spots, some current paths that are used more then others and thus heat up much faster. the Problem with the P4s architecture is that there are some hot spots that are used so much that they are very sensitive to extra current and decompose much much faster, killing the chip. amd seems to have a better design where the load is not concentrated that much on some single spots, making them more resistant for both, heat and voltage.

and actually most people DO realiza that their cpus are dying. if the voltage is just slightly more then what the curerent paths can take, they will not immediatly short or be cut, but slowely decompose, making the chip crash more and more often and work only at slower and slower speed. this happend to many people i know. their chips started to crash randomly, and the max stable frequency they would operate at would reduce faster and faster until they would not boot or work stable at any speed.

and to answer the question: It depends of the cooling. if your cpu is not running too hot i dont think 2v will kill it. it might reduce its lifespan to 2 or 3 years if its running @ high temps and 2v...

[BrainStorm]

Ewok, I agree with you that 60C will not kill a CPU immediately, although >90C probably will (or at least come close). But with respect to elctromigration, I'm going to have to agree with saaya here. The cooler your cpu is, the less electromigration will occur at a given voltage. But that being said, given the same cooling, 2v will kill your cpu faster than 1.9v, which will kill your cpu faster than 1.8v, etc. The only question is how long will it really last at 2v. The better the cooling, the longer it will last. But no one knows for sure how long for any given cpu. It might last a few days....it might last a few months...it might last a few years. Personally, I wouldn't run it at 2v for 24/7, but that's just me.

[Johnny Knoxvill]

i wouldn't run it on 2V whatever cooling you have, unless you don't mind replacing the cpu a couple of times a year.

[Ewok]

the problem is knowing exactly what voltage the chip is getting.

mines set at 1.93 in bios, reads as 1.84-1.86 in bios/mbm/winbond and measuring off the leg of the old vcore regulator chip shows 1.57v?!?!?

i would have thought measuring off the leg of the old vcore chip where people did vcore mods on the old KR7, KX7 etc would give the accurate voltage but it CANNOT be at 1.57v when im running at 2.5ghz.

so telling exactly which voltage the chip is getting is difficult if not impossible, i still dont know what voltage my chip is actually getting.

[saaya]

why do so many people think that 2v kills an xp in weeks???

nobody has ever lost an xp because of too much voltage, and there are hundrets of people running them at 2.2v+ !!! for months...

[felix88]

Ewok, finding the actual voltage that your chip is getting shouldn't be that hard. you've just got to know where to look.

unfortunatly i don't have the board, or know enough about it to tell you were to measure. i do know that the voltage goes into the chip somewhere and you should be able to measure it there.

[saaya]

so you dont know about it but you you say its easy?

[Ewok]

ive asked in amd forum but nobody seems to want to give me an answer.

i thought measuring off the chip leg that people did vcore mods from would tell me the right voltage but that said 1.57 which cant be right. strangely the same leg on the same chip said 1.82 when it was set at 1.80 in the bios but with a different cpu.

so i have no idea.

[saaya]

i only know that the cpu itself doenst have a single pin to get its juice but dozens... dunno if they are all fed by the same voltage line.

check the pdf files about the board or ask amd or the company that build your board...

:)

[Royal Oaks]

AMD spec sheet says no more than .5V over stock. So, if it's a DLT3C you are really pushing it, and yes that will make the CPU die sooner (not instantly). You also have to remember though that these CPU's are designed to run for years upon years. So, even if you are running that high voltage unless your ambient is really high I think you are alright. I can run 2V through my Palo all day long but that isn't too bad (only .25V over stock). I am also running an SK-7 and Smart Fan II which is much more powerful than the Panaflo. It's all in your temps though, which is the great thing about AMD. If you are getting good temps at 2V I wouldn't worry about it. However, if you are over 50C that is too much for my blood.

[PimpJack]

Originally posted by saaya
hehehe

well my xp1700+ ocs to 2.1ghz with the stock voltage. im bruning it in @ that speed now with the voltage @ 2.1v.
and using the large coolermaster silent cooler it wont even go over 60°C.
i think that the high temps resulting the voltage will reduce the lifespan of your cpu, but dont know how much.

there are hundrets of people running their xps @2.2v with h20 or vapor cooling and they seem to run fine...

i NEVER heard of an xp that died from too much voltage...

Why would u ever want to increase ur voltage to 2.1v when u can do the same at 1.65 or 1.5v.... U can't possibly be helping a chip by pounding it with unnecessary voltage. If I fed u some speed and made u sit still u'd be pretty pissed off wouldn't u?

I'v been running my 2100@2530 and 2.03v for a long time with load temps of around 30-35c it seems to be going strong. I would reccomend everyone ocing high to use 50c as their MAX load temp. My high oc's usually crash if the chip gets above 50c anywayz.
:toast:

[CCW]

dont amd chips use 5v line so measure that?