When OVERCLOCKING, how does an increased VOLTAGE harm the CPU?

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

x86

Banned
Oct 12, 2001
397
0
0


<< I can feel a ban comming. >>



Excuse me?! On what charge?! May I ask, who the hell has the power to do that?!?!
 

AGodspeed

Diamond Member
Jul 26, 2001
3,353
0
0


<<

<<
You can't get back at an AT moderator DocDumb@$$. He'll just destroy you. ;)
>>



Oh really? Why is that?
>>



An Anandtech Moderator oversees these forums. What are you going to do to the Moderator Doc? Bad mouth him?
 

Maggotry

Platinum Member
Dec 5, 2001
2,074
0
0


<< May I ask, who the hell has the power to do that?!?! >>


I'd guess the same guy/gal that had the power to change your nick.
 

x86

Banned
Oct 12, 2001
397
0
0


<<

<<

<<
You can't get back at an AT moderator DocDumb@$$. He'll just destroy you. ;)
>>



Oh really? Why is that?
>>



An Anandtech Moderator oversees these forums. What are you going to do to the Moderator Doc? Bad mouth him?
>>



It could have been a MOD aka a MODifier.
 

RaynorWolfcastle

Diamond Member
Feb 8, 2001
8,968
16
81
Nope, Mod is a moderator and pissing them off is a surefire way to get a long-term AT-sponsored vacation.

I would go complain in Forum Issues if I were you

-Ice
 

ST4RCUTTER

Platinum Member
Feb 13, 2001
2,841
0
0
This is just a little prank of an AMD Zealot that can't accept the fact that AMD chips are horrible. As for who did this: Get a life.

I am currently programming a MOD to rechange my user name.



<< I can feel a ban comming. >>

Excuse me?! On what charge?! May I ask, who the hell has the power to do that?!?!



I can assure you, writing any kind of script, hack, or "mod" to give you administrative access into this forum's server is a quick way towards an IP ban. You earned the ire of the others by trolling the forums under the pretense of being a victim of AMD's products. Life has a funny way of forcing you to reap what you sow...




 

AGodspeed

Diamond Member
Jul 26, 2001
3,353
0
0


<<

<<

<<

<<
You can't get back at an AT moderator DocDumb@$$. He'll just destroy you. ;)
>>



Oh really? Why is that?
>>



An Anandtech Moderator oversees these forums. What are you going to do to the Moderator Doc? Bad mouth him?
>>



It could have been a MOD aka a MODifier.
>>



What?!?! When people use the term "mod" or "MOD" on these forums, they mean moderator, as in an Anandtech Moderator. An Anandtech Moderator can ban you from these forums if they want to. They can lock your threads (like the thread you created about your water cooling experience with your Dad's Athlon) or they can simply change your handle name (in your case, you have now been dubbed DocDumb@$$).

There's nothing you can do.
 

x86

Banned
Oct 12, 2001
397
0
0


<< This is just a little prank of an AMD Zealot that can't accept the fact that AMD chips are horrible. As for who did this: Get a life.

I am currently programming a MOD to rechange my user name.



<< I can feel a ban comming. >>

Excuse me?! On what charge?! May I ask, who the hell has the power to do that?!?!



I can assure you, writing any kind of script, hack, or "mod" to give you administrative access into this forum's server is a quick way towards an IP ban. You earned the ire of the others by trolling the forums under the pretense of being a victim of AMD's products. Life has a funny way of forcing you to reap what you sow...
>>



True true true. That ban wouldn't have any effect on dynamic IP would it? :)
 

x86

Banned
Oct 12, 2001
397
0
0


<<

<<
It could have been a MOD aka a MODifier.
>>



What?!?! When people use the term "mod" or "MOD" on these forums, they mean moderator, as in an Anandtech Moderator. An Anandtech Moderator can ban you from these forums if they want to. They can lock your threads (like the thread you created about your water cooling experience with your Dad's Athlon) or they can simply change your handle name (in your case, you have now been dubbed DocDumb@$$).

There's nothing you can do.
>>



Oh really??? Tinker Tinker Tinker...
 

x86

Banned
Oct 12, 2001
397
0
0


<< Nope, Mod is a moderator and pissing them off is a surefire way to get a long-term AT-sponsored vacation.

I would go complain in Forum Issues if I were you

-Ice
>>



Why is that?

Just because of this stupid name change, this is probably going to be the most replied to forum in on anandtech. That is ironic in a way.

The mod must really like AMD. :)
 

RaynorWolfcastle

Diamond Member
Feb 8, 2001
8,968
16
81
I really, strongly suggest that you stop threatening the moderators and the forums in general if you EVER want to post here again.

Go complain in forum issues and hope for the best, otherwise you WILL GET BANNED PERMANENTLY... *cough* Orcish *cough*

I tell you this as a well-wisher

-Ice
 

WilsonTung

Senior member
Aug 25, 2001
487
0
0


<< SCREW YOU! For your information, I have not been trolling, just stating my opinion. >>



You should have been banned for your persistent use of Ad hom attacks. I guess the Moderator is just playing with you at this point.

A bit of advice - you probably want to leave before things get worse. As much as I dislike you there is no reason why you should be subjected to more humiliation.
 

pm

Elite Member Mobile Devices
Jan 25, 2000
7,419
22
81
Increased voltage leads to three reliability concerns directly and one problem indirectly. Apologies in advance if I have resorted too much to engineering-ese in this post... but it is a complex topic.

The primary reliability concern for high voltages is gate dielectric breakdown. At a high enough voltage the dielectric of the gate of a MOSFET transistor will 'breakdown' essentially like air breaks down in the presence of a high voltage differential between the earth and a stormcloud. In the case of a MOSFET, however, the breakdown leads to a permanent highly resistive path through the FET and the transistor breaking. So, high enough voltages will lead to a dead chip once the breakdown voltage of the weakest FET on the chip is reached.

Two other related concerns are PMOS BTI and NMOS Hot-E. Both are fairly complex topics - in fact I'm still not absolutely certain that researchers entirely understand the causes of PMOS BTI - so I'll gloss over the details as to specifically what happens. From a macro point of view what happens is that higher voltages on the gate lead to increased threshold voltages on the transistor over time. This leads to reduced drive strength (AKA higher channel resistance during "on"), and eventually the reduced current drive will led to a critical path failure.

Lastly electromigration - the physical transport of atoms under constant current conditions - leads to movement of atoms in the wires which eventually thins out the wires so that they become too resistive to carry current (leading to IR voltage drop on the power rails typically, leading to eventual failure).

The indirect source of problems from voltage is, as many others pointed out, heat. Heat affects all of the above issues strongly. They would all have an increased effect if somehow the die temperature were a constant and the voltage were increased, but heat is an additional factor. So increased heat leads to these effects occuring more quickly.



Lastly, I'm not sure if I understand the bulk of this thread. It appears that a user nicked "DocSmarts" has had his nick changed to something less complimentary. If a Moderator truly did change the name of a user, I am not impressed. If a moderator feels that a user is out of line, then those posts should be changed, the user should be warned, or the user should be banned. But changing a users name so that that user is publicly ridiculed is out of line IMO. Moderators have a higher level of responsibility than users on here by definition, and part of their job should be to set an example. Name-calling is fairly childish and should be, in my opinion, beneath a moderators dignity. If someone is out of line - and in this case, I have not seen any other posts so I can't judge - there are better ways to deal with the problem.

Patrick Mahoney
Microprocessor Design Engineer
Intel Corporation
 

ST4RCUTTER

Platinum Member
Feb 13, 2001
2,841
0
0
True true true. That ban wouldn't have any effect on dynamic IP would it?

I'm pretty sure your ISP doesn't have an infinite range of IP's. Banning your IP range i.e. 68.2.14.X would effectively prevent you from accessing the forums. It might ban others using your ISP as well, but chances are it wouldn't be an issue.

edit: As usual pm brings his expertise and evenhanded judgement to yet another thread. If a moderator truly did change DocSmart's username, it's a show of what might be considered vindictive moderation. :(
 

RaynorWolfcastle

Diamond Member
Feb 8, 2001
8,968
16
81


<< Lastly, I'm not sure if I understand the bullk of this thread. It appears that a user nicked "DocSmarts" has had his nick changed to something less complementary. If a Moderator truly did change the name of a user, I am not impressed. If a moderator feels that a user is out of line, then those posts should be changed, the user should be warned, or the user should be banned. But changing a users name so that that user is publicly ridiculed is out of line IMO. Moderators have a higher level of responsibility than users on here by definition, and part of their job should be to set an example. Name-calling is fairly childish and should be, in my opinion, beneath a moderators dignity. If someone is out of line, there are better ways to deal with the problem. >>



Agreed. While I did find it hillarious, I agree that no one should be subjected to this kind of humiliation. This is the reason people like Wingznutz leave IMO.

<-- Wishes all Anandtechers including himself were as civilized as pm

-Ice
 

AGodspeed

Diamond Member
Jul 26, 2001
3,353
0
0


<< Increased voltage leads to three reliability concerns directly and one problem indirectly. Apologies in advance if I have resorted too much to engineering-ese in this post... but it is a complex topic.

The primary reliability concern for high voltages is gate dielectric breakdown. At a high enough voltage the dielectric of the gate of a MOSFET transistor will 'breakdown' essentially like air breaks down in the presence of a high voltage differential between the earth and a stormcloud. In the case of a MOSFET, however, the breakdown leads to a permanent highly resistive path through the FET and the transistor breaking. So, high enough voltages will lead to a dead chip once the breakdown voltage of the weakest FET on the chip is reached.

Two other related concerns are PMOS BTI and NMOS Hot-E. Both are fairly complex topics - in fact I'm still not absolutely certain that researchers entirely understand the causes of PMOS BTI - so I'll gloss over the details as to specifically what happens. From a macro point of view what happens is that higher voltages on the gate lead to increased threshold voltages on the transistor over time. This leads to reduced drive strength (AKA higher channel resistance during "on"), and eventually the reduced current drive will led to a critical path failure.

Lastly electromigration - the physical transport of atoms under constant current conditions - leads to movement of atoms in the wires which eventually thins out the wires so that they become too resistive to carry current (leading to IR voltage drop on the power rails typically, leading to eventual failure).

The indirect source of problems from voltage is, as many others pointed out, heat. Heat affects all of the above issues strongly. They would all have an increased effect if somehow the die temperature were a constant and the voltage were increased, but heat is an additional factor. So increased heat leads to these effects occuring more quickly.
>>



As usual, a very informative and interesting read. Thanks Pat. :)
 

x86

Banned
Oct 12, 2001
397
0
0


<< Increased voltage leads to three reliability concerns directly and one problem indirectly. Apologies in advance if I have resorted too much to engineering-ese in this post... but it is a complex topic.

The primary reliability concern for high voltages is gate dielectric breakdown. At a high enough voltage the dielectric of the gate of a MOSFET transistor will 'breakdown' essentially like air breaks down in the presence of a high voltage differential between the earth and a stormcloud. In the case of a MOSFET, however, the breakdown leads to a permanent highly resistive path through the FET and the transistor breaking. So, high enough voltages will lead to a dead chip once the breakdown voltage of the weakest FET on the chip is reached.

Two other related concerns are PMOS BTI and NMOS Hot-E. Both are fairly complex topics - in fact I'm still not absolutely certain that researchers entirely understand the causes of PMOS BTI - so I'll gloss over the details as to specifically what happens. From a macro point of view what happens is that higher voltages on the gate lead to increased threshold voltages on the transistor over time. This leads to reduced drive strength (AKA higher channel resistance during "on"), and eventually the reduced current drive will led to a critical path failure.

Lastly electromigration - the physical transport of atoms under constant current conditions - leads to movement of atoms in the wires which eventually thins out the wires so that they become too resistive to carry current (leading to IR voltage drop on the power rails typically, leading to eventual failure).

The indirect source of problems from voltage is, as many others pointed out, heat. Heat affects all of the above issues strongly. They would all have an increased effect if somehow the die temperature were a constant and the voltage were increased, but heat is an additional factor. So increased heat leads to these effects occuring more quickly.



Lastly, I'm not sure if I understand the bulk of this thread. It appears that a user nicked "DocSmarts" has had his nick changed to something less complimentary. If a Moderator truly did change the name of a user, I am not impressed. If a moderator feels that a user is out of line, then those posts should be changed, the user should be warned, or the user should be banned. But changing a users name so that that user is publicly ridiculed is out of line IMO. Moderators have a higher level of responsibility than users on here by definition, and part of their job should be to set an example. Name-calling is fairly childish and should be, in my opinion, beneath a moderators dignity. If someone is out of line - and in this case, I have not seen any other posts so I can't judge - there are better ways to deal with the problem.

Patrick Mahoney
Microprocessor Design Engineer
Intel Corporation
>>



The language is all in the writing. You have answered all of my questions quite well! Thank you.
 

x86

Banned
Oct 12, 2001
397
0
0


<<

<< SCREW YOU! For your information, I have not been trolling, just stating my opinion. >>



You should have been banned for your persistent use of Ad hom attacks. I guess the Moderator is just playing with you at this point.

A bit of advice - you probably want to leave before things get worse. As much as I dislike you there is no reason why you should be subjected to more humiliation.
>>



Do moderators usually do such harsh things such as change one's nick name without cause?
 

RaynorWolfcastle

Diamond Member
Feb 8, 2001
8,968
16
81
PM, does overclocking without changing the voltage affect the lifespan of a CPU or GPU? Does the faster switching cause increased wear on the CPU?

-Ice
 

x86

Banned
Oct 12, 2001
397
0
0


<<

<< Lastly, I'm not sure if I understand the bullk of this thread. It appears that a user nicked "DocSmarts" has had his nick changed to something less complementary. If a Moderator truly did change the name of a user, I am not impressed. If a moderator feels that a user is out of line, then those posts should be changed, the user should be warned, or the user should be banned. But changing a users name so that that user is publicly ridiculed is out of line IMO. Moderators have a higher level of responsibility than users on here by definition, and part of their job should be to set an example. Name-calling is fairly childish and should be, in my opinion, beneath a moderators dignity. If someone is out of line, there are better ways to deal with the problem. >>



Agreed. While I did find it hillarious, I agree that no one should be subjected to this kind of humiliation. This is the reason people like Wingznutz leave IMO.

<-- Wishes all Anandtechers including himself were as civilized as pm

-Ice
>>



What does the acronym: IMO stand for?

I cannot agree more that PM is quite well spoken.
 

Ionizer86

Diamond Member
Jun 20, 2001
5,292
0
76
DocDumb@$$, IMO is an acronym for in my opinion. IMHO is also seen, and it stands for in my humble opinion. In addition, PM stands for Private Message or Private Mail, so pm would probably prefer to be referred to by his lowercase handle of pm. ;)