cpu voltage and overclocking

[pwnt by pat]

i did a quick search and didnt find anything, its also very late and i want to get to bed, so forgive me if this has been asked in the past.

how does the increasing voltage to the cpu affect how much higher it can be overclocked?

thanks

[pip]

it takes energy to flip transistors on and off and the more times it happens the more power is needed, that's about the simplest explanation I can give

[Punisher69]

But more heat is created in doing so

[Avatar28]

it also has to do with electrical noise. As the speed increases, so does noise and eventually the noise will overwhelm the signals within the circuits and cause errors. Adding more voltage increases the "volume" of the signal and thus helps it punch through the noise as it were. That's the real reason I believe.

[Lynx]

Increasing the volume increases the noise levels. I am doing work experience at ARM at the moment and working at bellow the transistor level to get a core to work at a certain process. The real reason is that transistors have a capasitance and so it takes a while to change from a high state to a low state

Therefore the higher the voltage the faster the gates can change states but also the hotter they get which acts as a limit.

[InThrees]

ALSO...

"Transistors" can display two different states: on, and off.

and 'off' state is read when the voltage is below a certain point, and an 'on' state is read when the voltage is above that same point.

Frequently, if the overall vcore voltage is too low, and the chip is overclocked too high, a given transistor won't be able to display enough voltage to be read as 'on', even though it SHOULD be.

Upping the vcore allows for more successful 'on' reads, and the cpu returns to normal operation, instead of returning a bunch of errors or just plain not working. This only works to a point, as you might guess. Running 5 volts through a chip will fry it.

Poor explanation, but hopefully clear enough to help.

[hitechjb1]

This link points to some explanations.

Relationship of clock, die temperature and Vcore (update) (http://forum.oc-forums.com/vb/showthread.php?s=&postid=2026510#post2026510) (page 13)

Relationship of clock, die temperature and Vcore
What is the active power of a CPU at frequency f and voltage V
How to estimate CPU static and active power
(http://forum.oc-forums.com/vb/showthread.php?s=&postid=1709776#post1709776) (page 2)

[ninthebin]

there is a really neat article I have read about this which is pretty simple to follow.

http://hardwareanalysis.com/content/reviews/article/1482/

[RCtruckguy]

inthrees thats THE BEST explaination ive ever seen!! EVER!

~Alex

[dippy_skoodlez]

LOL i know.. GJ inthrees :)

[Xymurgy]

I liked ninthebin's. I guess I just need pictures.

After I read that, I did an 'ohhhhhhhhhhhh.'

[Zuzzz]

I suggest you all read the link ninthebin posted. . it IS the best explaination and certainly makes total sense. remember folks it IS ALL about speed. AWESOME post ninthebin and good looking out. . I think that is by far the best explaination and the simplest. . My WIFE even got the jist of that. THNX . . good post.

(yeah yeah Im a post wh**e. lol)

PS: and yes "I" did learn a lil on that. . that is the whole point. . to learn:D

[dudomann]

WOW! Great article! I learned alot from it.
~Bobb