Right now its looking like i can just barely do over 2.0ghz @ 1.11v. I was able to get 2.9 on 1.54v @ 46c on air and am trying one of the "burn-in" ideas that float around. It seemed to work on my x2 3800, so i'll try with this. but does that seem kinda weak that it will only do 2.0 @ 33c? my x2 was able to do 2.2ghz @ 1.07 and it was an 0529
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
Under volted opteron 144 0551
- Thread starter menlatin
- Start date
well its a technique that SOME overclockers use to improve their final overclock. Some people actually gain mhz after running a processor undervolted and stressed for upwards of 24+ hrs. Some people get nothing though, so its just a luck thing. but for those that are familiar with this technique, does it seem weak?
Just trying to inform someone that they have false information, its annoying seeing people believe that burning in will actually give you a better oc, the only thing burning in is for is stability testing, it DOES NOT give you a better oc, its that simple. Goto XS, I dont think your gonna hurt anyones feelings.
- Joined
- Feb 4, 2003
- Messages
- 2,845
Undervolting is certainly not how you would burn in a CPU.
I always jump onto this because it needs repeating so don't take it personally;
Burning in is no more than a process to 'burn in' the contact between your cpu and heatsink. Nothing more. No magic happens inside the cpu that makes it go faster. Burning in works because the contact point is really small considering the amount of heat being put out so the better the contact, the better the heat transfer, the cooler the chip will run, and so the faster you can overclock it.
Because of this; to properly burn in a chip you simply run it as hot as possible for an extended period of time. A cpu will naturally burn in anyway over time due to the same process, but you can accelerate it obviously. Cpu's with heatspreaders on them will see very little affect of burn in due to the much larger surface area.
I always jump onto this because it needs repeating so don't take it personally;
Burning in is no more than a process to 'burn in' the contact between your cpu and heatsink. Nothing more. No magic happens inside the cpu that makes it go faster. Burning in works because the contact point is really small considering the amount of heat being put out so the better the contact, the better the heat transfer, the cooler the chip will run, and so the faster you can overclock it.
Because of this; to properly burn in a chip you simply run it as hot as possible for an extended period of time. A cpu will naturally burn in anyway over time due to the same process, but you can accelerate it obviously. Cpu's with heatspreaders on them will see very little affect of burn in due to the much larger surface area.
spine said:Undervolting is certainly not how you would burn in a CPU.
I always jump onto this because it needs repeating so don't take it personally;
Burning in is no more than a process to 'burn in' the contact between your cpu and heatsink. Nothing more. No magic happens inside the cpu that makes it go faster. Burning in works because the contact point is really small considering the amount of heat being put out so the better the contact, the better the heat transfer, the cooler the chip will run, and so the faster you can overclock it.
Because of this; to properly burn in a chip you simply run it as hot as possible for an extended period of time. A cpu will naturally burn in anyway over time due to the same process, but you can accelerate it obviously. Cpu's with heatspreaders on them will see very little affect of burn in due to the much larger surface area.
that doens't sound right to me at all .. i don't think burning in has anything to do with the HSF ... people burn in their ram
- Joined
- Feb 4, 2003
- Messages
- 2,845
Memory is completely different, and frankly, I've yet to see actual real evidence for burning in of system memory.
Do you really believe a cpu could magically get better over a week of doing something special? Don't you think the cpu manufacturers would do this themselves and better if true?
Do you really believe a cpu could magically get better over a week of doing something special? Don't you think the cpu manufacturers would do this themselves and better if true?
- Joined
- Feb 4, 2003
- Messages
- 2,845
What point are you trying to make with that little comment? You're saying they're ignorant of it? Maybe they know but they don't do it just to get back at us nasty consumers? Maybe they don't want us to know that secretly the very chips right under our noses are capable of much higher speeds but they ironically make us buy faster ones?
Seriously, if there were such a burn in process that physically altered the internals of a cpu that made it go faster, why wouldn't a cpu manufacturer do this on the consumers behalf and make a slower chip faster and sell it for more?
Seriously, if there were such a burn in process that physically altered the internals of a cpu that made it go faster, why wouldn't a cpu manufacturer do this on the consumers behalf and make a slower chip faster and sell it for more?
cpu makers do that exact same thing by selling chips with differenct multipliers .. thus the reason we overclock them....
cpu makers wouldn't burn them in because it takes like 24hours(hours = money) and the results are not certain... but if some guy wants to burn in his chip in hopes of squeezin out a few more hrz then what's the problem...
cpu makers wouldn't burn them in because it takes like 24hours(hours = money) and the results are not certain... but if some guy wants to burn in his chip in hopes of squeezin out a few more hrz then what's the problem...
- Joined
- Feb 4, 2003
- Messages
- 2,845
That guide is talking about burn in relating to CPU's and not ram. I'm not disputing burn in for cpu's, merely the theory behind it. I'd argue that anyone who's seen a decent gain from a 'burn in' would immediately lose it should they clean and reseat their cpu and heatsink.malicious said:
Occam's Razor applies here I feel. What's more likely; a higher overclock is achieved through bettering the known to be critical cpu TIM joint, or that a cpu just miraculously gets better through some crazy electron migration?