Does "Burning In" actually change anything?

daos

Senior member
Jan 2, 2003
940
0
0
this is a topic all too familiar in the overclockers world..."how do i burn in?" honestly ive heard many mixed emotions on this but never any hard core evidence to prove either 1)it actually changes anything, or 2) it doesnt change anything at all and is a complete myth.

a good freind of mine has actually told me that if you run a CPU at an overclocked speed for 24-72hours(burn-in) than it can actually change the physical attributes of the chip to allow a more overclockable chip. he says it actually does something to the makeup of the crystaline structure of the circuitry within the chip.

i am actually neutral as of now, but would really appreciate some wise advice here from some of you. once again i thank you all for the time and consideration of a reply.

daos
 

Wingznut

Elite Member
Dec 28, 1999
16,968
2
0
Originally posted by: daos
Does "Burning In" actually change anything?
No.

Originally posted by: daos
a good freind of mine has actually told me that if you run a CPU at an overclocked speed for 24-72hours(burn-in) than it can actually change the physical attributes of the chip to allow a more overclockable chip. he says it actually does something to the makeup of the crystaline structure of the circuitry within the chip.
Your friend is mistaken.

I'm sure pm or Eskimo will get into more detail... But the gist is that overclocking a chip can do either nothing, or damage, to the circuits. Besides that, the cpu's are put through a more rigorous stress test than either you or I (or even your friend) could ever do. Extreme temperatures, voltages, frequencies, and code to stress the different parts of the cpu.

And think about it this way... If it were true that overclocking a cpu could give you a faster cpu, don't you think that AMD and Intel would do it at the factory so that they could sell it at a higher rated speed (and a higher price)?
 

daos

Senior member
Jan 2, 2003
940
0
0
hey wingznut, this is a bit off topic, but i was wondering what did you major in college for? i am getting the urge to go back to school and mircocircuitry sounds really interesting. i am very curious to learn how everything is done, works, etc...

thanks bud.
 

Wingznut

Elite Member
Dec 28, 1999
16,968
2
0
I went to a tech school and studied automotive electronics and mechanics. Then spent the next ten years of my life fixing cars. All of my microelectronics and semiconductor education was self-taught and was driven by the fact that I absolutely despised my automotive career.

Now... I don't think I could be more satisfied with my current job. (Well, a higher stock price wouldn't hurt. :D )

This probably isn't the place to get into more details. So, feel free to send me a private msg/ICQ/AIM/MSN... Whatever works for you. :)
 

Mday

Lifer
Oct 14, 1999
18,646
1
76
The burn in process is a myth. What it's actually used for is to make sure the system is running prior to actually keeping the system. This can be done at the store of purchase to make sure the system works, or at home for a newly built system (in case parts fail and have to be returned).

As for overclocking, wingnut is right, he's not 1337 for nothing. And, they usually clock on the side of caution. There is always some buffer, some say it's 10%, some say 3%, I dont know what that buffer is. Anyway, overclocking does improve performance, but something like 3% or 10% is not going to be noticeable without a benchmark. I am not sure about intel, but rumor has it that AMD has "down" clocked a core due to the demand for a certain speed processor over a processor which is faster.
 

pm

Elite Member Mobile Devices
Jan 25, 2000
7,419
22
81
There is burn-in and then there is burn-in. In semiconductor manufacturing terminology "burn-in" is a stage of the production flow after packaging in which the CPU is placed in an elevated temperature environment and is stressed at atypical operating conditions. The end goal of this is to dramatically reduce the statistical probability of "infant mortality" failures of product on the street. "Infant mortality" is a charateristic of any form of complex manufacturing in that if you were to plot device failures in the y-axis and time in the x-axis, the graph should look like a "U". As the device is used, initially quite a few fail but as time goes on this number drops off (you are in the bottom of the "U" in the graph). As the designed life of the product is reached and exceeded, the failure count rises back up again. Burn-in is designed to catch the initial failures before the product is shipped to customers and to put the product solidly in the bottom section of the "U" graph in which few failures occur. During this process there is a noticeable and measureable circuitry slow-down on the chip that is an unfortunate by-product of the process of running at the burn-in operating point. You put a fast chip into the burn-in ovens and it will always come out of the ovens slower than when it went in - but the ones that were likely to fail early on are dead and not shipped to customers.

There are two mechanisms that cause the circuitry in CMOS - particularly modern sub-micron CMOS - to slow down when undergoing the burn-in process: PMOS bias-temperature instability (PMOS BTI) and NMOS hot-electron gate-impact ionization (known as "NMOS hot-e"). Both of these effects are complex quantum-electrical effects that result in circuitry slowing down over time. You should be able to type either of these two terms into Google to read more about what is actually happening. The end-result is, as mentioned, that the chips will start to fail at a lower frequency than they did before going into burn-in due to the transistor current drive strength being reduced.

There is another use for the term "burn-in" wrt. chips that is used by system builders and that is as a test for reliability and to reduce customer returns due to component failure. This usually consists of putting the system together, plugging it in and running computational software on the system for a period of 24-48 hours. At the larger OEM companies, this is often done at a higher than typical operating temperature.

Some time ago someone on the internet wrote a very factual sounding article on the benefits of running a CPU at a higher than typical voltage for a day or two to improve it's "overclockability". This author wrote some scientific sounding verbage about how NMOS hot-e actually improves the drive strength of PMOS devices as a supposed explanation for why this method works. Reading this particular article and, even worse, seeing people commenting that this was a wonderful article that everyone should follow was the reason why I started posting on Anandtech way back when. The author was wrong on several key points - primarly that NMOS hot-e can occur in electron-minority (hole majority) carrier devices that are biased such as to repell electrons - and I contacted the author with a wide assortment of technical journals showing that he was wrong. He was not particularly open to the fact that he might be mistaken and never remove the article from the website that I'm aware of. Suffice to say, however, that he did not understand basic semiconductor electronics and was wrong.

There is no practical physical method that could cause a CPU to speed up after being run at an elevated voltage for an extended period of time. There may be some effect that people are seeing at the system level, but I'm not aware of what it could be. Several years ago when this issue was at it's height on the Internet, I walked around and talked to quite a few senior engineers at Intel asking if they had heard of this and what they thought be occurring. All I got were strange looks followed by reiterations of the same facts as to why this couldn't work that I had already figured out by myself. Finally, I was motivated enough to ask for and receive the burn-in reports for frequency degradation for products that I was working on at the time. I looked at approximately 25,000 200MHz Pentium CPU's, and approximately 18,000 Pentium II (Deschutes) CPU's and found that, with practically no expections at all, they all got slower after coming out of burn-in by a substanial percentage.

To me there is no doubt in my mind that suggesting that users overvoltage their CPU's to "burn them in" is a bad thing. I'd liken it to an electrical form of homeopathy - except that ingesting water when you are sick is not going to harm you and overvoltaging a CPU for prolonged periods of time definitely does harm the chip. People can do what they want with their machines that they have bought - as long as the aware that what they are doing is not helping and is probably harming their systems. I have seen people - even people who know computers well - saying that they have seen their systems run faster after "burning it in" but whatever effect they may or may not be seeing, it's not caused by the CPU running faster.

Patrick Mahoney
Senior Design Engineer
Enterprise Processor Division
Intel Corp.
 

daos

Senior member
Jan 2, 2003
940
0
0
all i can say is....wow!

pm, your my hero. ;) thats probably the most accurate description ive heard to date. boy i sure do wish i could hang out with some of you guys just for a day or so around the office. :)

one of these days i hope i can work for Intel, and learn all about the different attributes and properties of these technologies. sounds so interesting...im determined to learn more.

if one of you could, please PM me and suggest some really good books, web articles, etc..so i can better my learning on these kinds of things. thanks again and God Bless..
 

pm

Elite Member Mobile Devices
Jan 25, 2000
7,419
22
81
Thanks for the compliments. :) This issue is one that I feel strongly about and I have been posting about for years now.
one of these days i hope i can work for Intel, and learn all about the different attributes and properties of these technologies. sounds so interesting...im determined to learn more.
My goal when I was in high school and college was to work for Intel and things worked out such that I could. One of the members of this forum here at Anandtech applied for a co-op position at Intel, was offered a position and hopefully should be joining my team in the latter half of this year. If you are interested, I would highly recommend applying. When I submitted my resume, I thought that it was a bit of a long shot and Intel called me back in under a week.
if one of you could, please PM me and suggest some really good books, web articles, etc..so i can better my learning on these kinds of things. thanks again and God Bless..
I just started my sabbatical for Intel last week - at Intel every 7 years of working with the company, there is a 2 month paid vacation called a sabbatical - and I'm presently en route to the southern hemisphere for a couple of months. I don't have access to any of the articles that I could recommend for researching circuit reliability effects such as NMOS hot-e and PMOS BTI. Browsing on Google should turn up some good articles on the subjects though.

As far as a more general circuit design overview I would recommend the same book that I have recommended to other people on this forum recently. My favorite book for learning in detail how to design a chip from the circuit perspective is "Principles of CMOS VLSI Design" by Weste et.al.. It is very detailed and yet it is usually fairly readable. It is ridiculously expensive - as most textbooks seem to be - so I would recommend going to your local college library and reading it there. Or you could try ordering it from your local library. I have found it to be the best book that I can recommend to beginners because it covers all aspects of design, doesn't dive in too deep, there are plenty of diagrams and pictures and the authors are pretty good writers and so they make it easy to read.
 

zayened

Diamond Member
Feb 28, 2001
3,931
0
0
Originally posted by: pm
There is burn-in and then there is burn-in. In semiconductor manufacturing terminology "burn-in" is a stage of the production flow after packaging in which the CPU is placed in an elevated temperature environment and is stressed at atypical operating conditions. The end goal of this is to dramatically reduce the statistical probability of "infant mortality" failures of product on the street. "Infant mortality" is a charateristic of any form of complex manufacturing in that if you were to plot device failures in the y-axis and time in the x-axis, the graph should look like a "U". As the device is used, initially quite a few fail but as time goes on this number drops off (you are in the bottom of the "U" in the graph). As the designed life of the product is reached and exceeded, the failure count rises back up again. Burn-in is designed to catch the initial failures before the product is shipped to customers and to put the product solidly in the bottom section of the "U" graph in which few failures occur. During this process there is a noticeable and measureable circuitry slow-down on the chip that is an unfortunate by-product of the process of running at the burn-in operating point. You put a fast chip into the burn-in ovens and it will always come out of the ovens slower than when it went in - but the ones that were likely to fail early on are dead and not shipped to customers.

There are two mechanisms that cause the circuitry in CMOS - particularly modern sub-micron CMOS - to slow down when undergoing the burn-in process: PMOS bias-temperature instability (PMOS BTI) and NMOS hot-electron gate-impact ionization (known as "NMOS hot-e"). Both of these effects are complex quantum-electrical effects that result in circuitry slowing down over time. You should be able to type either of these two terms into Google to read more about what is actually happening. The end-result is, as mentioned, that the chips will start to fail at a lower frequency than they did before going into burn-in due to the transistor current drive strength being reduced.

There is another use for the term "burn-in" wrt. chips that is used by system builders and that is as a test for reliability and to reduce customer returns due to component failure. This usually consists of putting the system together, plugging it in and running computational software on the system for a period of 24-48 hours. At the larger OEM companies, this is often done at a higher than typical operating temperature.

Some time ago someone on the internet wrote a very factual sounding article on the benefits of running a CPU at a higher than typical voltage for a day or two to improve it's "overclockability". This author wrote some scientific sounding verbage about how NMOS hot-e actually improves the drive strength of PMOS devices as a supposed explanation for why this method works. Reading this particular article and, even worse, seeing people commenting that this was a wonderful article that everyone should follow was the reason why I started posting on Anandtech way back when. The author was wrong on several key points - primarly that NMOS hot-e can occur in electron-minority (hole majority) carrier devices that are biased such as to repell electrons - and I contacted the author with a wide assortment of technical journals showing that he was wrong. He was not particularly open to the fact that he might be mistaken and never remove the article from the website that I'm aware of. Suffice to say, however, that he did not understand basic semiconductor electronics and was wrong.

There is no practical physical method that could cause a CPU to speed up after being run at an elevated voltage for an extended period of time. There may be some effect that people are seeing at the system level, but I'm not aware of what it could be. Several years ago when this issue was at it's height on the Internet, I walked around and talked to quite a few senior engineers at Intel asking if they had heard of this and what they thought be occurring. All I got were strange looks followed by reiterations of the same facts as to why this couldn't work that I had already figured out by myself. Finally, I was motivated enough to ask for and receive the burn-in reports for frequency degradation for products that I was working on at the time. I looked at approximately 25,000 200MHz Pentium CPU's, and approximately 18,000 Pentium II (Deschutes) CPU's and found that, with practically no expections at all, they all got slower after coming out of burn-in by a substanial percentage.

To me there is no doubt in my mind that suggesting that users overvoltage their CPU's to "burn them in" is a bad thing. I'd liken it to an electrical form of homeopathy - except that ingesting water when you are sick is not going to harm you and overvoltaging a CPU for prolonged periods of time definitely does harm the chip. People can do what they want with their machines that they have bought - as long as the aware that what they are doing is not helping and is probably harming their systems. I have seen people - even people who know computers well - saying that they have seen their systems run faster after "burning it in" but whatever effect they may or may not be seeing, it's not caused by the CPU running faster.

Patrick Mahoney
Senior Design Engineer
Enterprise Processor Division
Intel Corp.


OMFG
 

daos

Senior member
Jan 2, 2003
940
0
0
well thanks a million patrick. i hope you stay with Intel long enough to see me. ;) i already went and bought the book, and your exactly right, its great. im totally siked! i cant wait to get a better understanding of this technology and how it works.

i feel like a kid on christmas morning, lol

cya around bud and have fun on your sabbatical. God Bless...
 

SuperTool

Lifer
Jan 25, 2000
14,000
2
0
Is there a third edition of that Weste book coming out anytime soon? I am a bit reluctant to invest in a 1994 textbook in 2003.
 

CTho9305

Elite Member
Jul 26, 2000
9,214
1
81
Originally posted by: SuperTool
Is there a third edition of that Weste book coming out anytime soon? I am a bit reluctant to invest in a 1994 textbook in 2003.

this one or this one? Are they different?

edit: wtf, one is by "Niel H E West" and the other by "Niel H E Weste"
 

pm

Elite Member Mobile Devices
Jan 25, 2000
7,419
22
81
The book that I attempted to link - it works on mine when I click it - is the first book. I think they are the same thing except that the latter has a manual on VHDL in it.

As far it being out of date... it is but it doesn't matter. I recommend this book for people who are completely new to any kind of circuitry design and the things that it covers are just as true today as they were in 1994. It explains how CMOS works, shows how to design simple logic circuitry, shows the various stages of manufacturing, and how to turn a schematic into a physical mask for manufacturing. It then extends this by actually working on a couple of examples like the design of a pretty simplistic CPU. Nothing that it discusses is really out of date. CMOS works today essentially like it did in 1994. The manufacturing stages are fundamentally the same. It doesn't talk about phase-shift masking, problems with power leakage, and other more esoteric matters, but it does a good job of covering the fundamentals - and even a few things that I think experienced designers should read such as skew tolerant designs.

It is not a book for those in the industry - there are a couple that I could recommend for this though - but it is very good for someone who has a deep desire to learn about the fundamentals of CMOS circuit design and how to put a chip together. It is not easy to read, but it's not completely unapproachable by someone without a background in EE.