PDA

View Full Version : Core Duo successfully over-clocked to 3.8GHz


MacBytes
May 9, 2006, 08:29 AM
http://www.macbytes.com/images/bytessig.gif (http://www.macbytes.com)

Category: Processors
Link: Core Duo successfully over-clocked to 3.8GHz (http://www.macbytes.com/link.php?sid=20060509092948)
Description:: The T2600 is the current top of the line 2.16GHz chip Intel ships. Yet one user has gotten the chip to run as high as 3.21 GHz using air-cooling, and up to 3.8GHz using phase change cooling.
.... theoretically, the current stock of Intel chips could likely run much faster than they currently do, providing they were put inside something BIG enough to support such a cooling systemů something like, say, a PowerMac.

Posted on MacBytes.com (http://www.macbytes.com)
Approved by Mudbug

combatcolin
May 9, 2006, 08:50 AM
Blimey.

adamfilip
May 9, 2006, 08:54 AM
gimme some benchmarks

i really dont care what speed the processor runs at
as long as it can perform.

id take a Core Treo 24mhz cpu over a 4ghz Core Duo if it performed faster

sunfast
May 9, 2006, 09:01 AM
That's quite fast. But, yeah, I'd like to see some benchmarks too.

Platform
May 9, 2006, 09:16 AM
gimme some benchmarks

i really dont care what speed the processor runs at
as long as it can perform.

id take a Core Treo 24mhz cpu over a 4ghz Core Duo if it performed faster

Yup give me some of those Treos :cool:

SPUY767
May 9, 2006, 09:51 AM
Here's the rub. . . It's not all about clock speed. You have to find a comfortable medium betwen clock speed and temperature. For instance, a 3.6 GHz chip running at a right chilly 34C will outperform a 4.0 GHz chip running at 60C+ easily. You see, the higher the temperature, the higher the resistance a material presents. The higher the resistance, the more trouble electrons have flowing through said material, and the harder Silicon transistors have to work to get the job done. Often higher temperatures, such as those present in laptops can lead to errors in the data paths that the chips have to correct for and this further slows processing down. That is generally why an equally clocked desktop will handily outpace a laptop, other factors such as HDD speed also play a role. So if this guy has a 2.16 GHz chip running at 4.0 and he's pushing 70C he can keep it, I'll take a 3.0 and 40-45C any day.

SPUY767
May 9, 2006, 09:54 AM
Yup give me some of those Treos :cool:

He's actually referring to the chip containing one TRILLION 24Mhz cores!

Mord
May 9, 2006, 09:58 AM
gimme some benchmarks

i really dont care what speed the processor runs at
as long as it can perform.

id take a Core Treo 24mhz cpu over a 4ghz Core Duo if it performed faster


on the same cpu core speed increases with clock speed in a linear fassion until the bottleneck hits something elce like the memory speed. which with the core duo would happen at 4.32GHz

Macnoviz
May 9, 2006, 10:28 AM
Magnificent, in a few months, the Powermac will be the fastest computer able to run Windows. I can't wait to let Oblivion go wild on such a system.

ChrisA
May 9, 2006, 10:29 AM
....You see, the higher the temperature, the higher the resistance a material presents. The higher the resistance, the more trouble electrons have flowing through said material, and the harder Silicon transistors have to work to get the job done.

Hey that's good. The same logic applies to disk drives too, you know. If the
platters spin to fast the force causes the recorded bits to slide to the outside
of the disk and eventually collect on the insides of trhe dive case.

One more tip: I can make my PC run faster by staightening the cables as best I can.
You see, the electrons have to slow down to make those turns, straight cables equals faster electrons equals fater computer.

adamfilip
May 9, 2006, 10:44 AM
He's actually referring to the chip containing one TRILLION 24Mhz cores!

That would be sweet!
of course running single threaded apps would be dreadfully slow!

Mord
May 9, 2006, 11:35 AM
anyone who thinks heat has an effect on cpu performance at a given clock speed needs their head examined, if anything happened in a way that it shouldent that error will probably snowball and freeze the pc, either it is stable at that speed or it does not, simple as.

SPUY767
May 9, 2006, 11:35 AM
Hey that's good. The same logic applies to disk drives too, you know. If the
platters spin to fast the force causes the recorded bits to slide to the outside
of the disk and eventually collect on the insides of trhe dive case.

One more tip: I can make my PC run faster by staightening the cables as best I can.
You see, the electrons have to slow down to make those turns, straight cables equals faster electrons equals fater computer.

Do you actually know anything about electricity? Because I have an engineering degree from GA Tech. Warmer materials conduct electricity more poorly than cooler materials. Have you ever heard of a thing called a super conductor? Superconductance is the transmission of electrons as electricity through a material with and infinitely small loss i.e. none. Last time I checked, the warmest superconducting material worked at 125 degrees kelvin. That's a right chilly -148C, or for the lay person who can't do a simple C to F conversion, -234F. Last time i checked, computers ran, oh about 200C or 392f too warm to conduct electricity without loss. Therefore, loss of electrons = slower and more costly transfer of information. Therefore, a conducting material running at nearly boiling is much less efficient than a material running at 30C.

Next time you want to treat someone like an idiot, I suggest you know what the hell you're talking about.

SPUY767
May 9, 2006, 11:40 AM
anyone who thinks heat has an effect on cpu performance at a given clock speed needs their head examined, if anything happened in a way that it shouldent that error will probably snowball and freeze the pc, either it is stable at that speed or it does not, simple as.


Have you ever overclocked a GPU? I ask because it's the easiest way to see a processor make errors because of heat. You see, at a certain point, the processor becomes unable to cope with the errors generated from lost data. This is when the GPU goes total guano and starts putting gibberish on the screen. any type of CPU can only cope with so much heat before the errors generated become to great to deal with. All modern chips have internal thermometers that tell them to abandon ship when things get too warm. Where do you guys think the heat in a computer comes from, Nesting sea birds? Heat in a computer and in a chip more specifically, is the by product of lost electrons which are converted from electrical energy to heat energy. The generation of this heat becomes exponential at a certain point because higher heat cosses less efficient transfer of energy, and thus more heat is generated as a byproduct.

dukebound85
May 9, 2006, 11:42 AM
anyone who thinks heat has an effect on cpu performance at a given clock speed needs their head examined, if anything happened in a way that it shouldent that error will probably snowball and freeze the pc, either it is stable at that speed or it does not, simple as.

Dude, know what you're talking about before stating complete bull. I'm amazed at how people who don't know the first thing about electronics act like they know it all. Heat does have an effect on performance and for those who don't think so are just ignorant like none other.

SPUY767
May 9, 2006, 11:45 AM
Dude, know what you're talking about before stating complete bull. I'm amazed at how people who don't know the first thing about electronics act like they know it all. Heat does have an effect on performance and for those who don't think so are just ignorant like none other.


::Golf Clap::

Mord
May 9, 2006, 11:53 AM
seeing as i've overclocked more computers than most have used, have been offered electrical engineering scholarships learnt to solder before i could ride a bike and am at the top of my electronics class (allbeit with only two others) i think i know a little bit more than you. even if you do infact have an engineering degree, which has nothing to do with electronics or physics.


the clock speed effects the speed, the heat effects the maximum clock speed attainable, gpu's are different as the information does not go back and forth much , only now do they have a bidirectional bus, so they are very tolerant of errors, but add another 20MHz and bam the screen go's black

the clock speed and the clock speed only governs the speed of a given processor the "MHz" is the number of times per second a cpu can complete a cycle, how faster the electrons go has no effect on performance it just allows higher clock speeds to be available.


someone who actually knows something back me up here, or if you don't go do some research before contradicting me.

amendum:

as for the "speed of electricity" the speed of electricity is c, the speed of light, the speed of the electrons is much much slower and has very little effect on anything, think of the electrons like a circular conga line, when one as the back starts the move the one at the front moves too, and the speed that the forward push go's at is the speed of light, however once the signal to go has gone all the way around the electrons are all going at a slow speed, however the "attack" of the electrons is always C.

i dont have the time right now but i'll do a full explanation of how voltage effects the stability of a cpu with diagrams.

SPUY767
May 9, 2006, 11:58 AM
seeing as i've overclocked more computers than most have used, have been offered electrical engineering scholarships learnt to solder before i could ride a bike and am at the top of my electronics class (allbeit with only two others)


the clock speed effects the speed, the heat effects the maximum clock speed attainable, gpu's are different as the information does not go back and forth much , only now do they have a bidirectional bus, so they are very tolerant of errors, but add another 20MHz and bam the screen go's black

the clock speed and the clock speed only governs the speed of a given processor the "MHz" is the number of times per second a cpu can complete a cycle, how faster the electrons go has no effect on performance it just allows higher clock speeds to be available.


someone who actually knows something back me up here, or if you don't go do some research before contradicting me.

Yes, I guess you turned away the droves of doctors just groveling at your feet to give you electronics degrees. No one cares, nor does it matter any more, this post is dead, and I petition the thread to be closed or at the very least, all of these posts erased. This land must be cleansed and born anew.

And about the soldering bit. . . Did you know that Chuck norris lost his virginity before his dad did? Beat that.

TheMasin9
May 9, 2006, 12:21 PM
some ppl need to take a chill pill... dont get angry at ignorant people, help them to understand thier misconceptions...

EGT
May 9, 2006, 12:24 PM
And about the soldering bit. . . Did you know that Chuck norris lost his virginity before his dad did? Beat that.

:confused:

What's this stuff all about, people saying silly things about Chuck Norris? Please, explain it to me.

Mord
May 9, 2006, 12:37 PM
Yes, I guess you turned away the droves of doctors just groveling at your feet to give you electronics degrees. No one cares, nor does it matter any more, this post is dead, and I petition the thread to be closed or at the very least, all of these posts erased. This land must be cleansed and born anew.

And about the soldering bit. . . Did you know that Chuck norris lost his virginity before his dad did? Beat that.


just accept that your wrong, preferably gracefully, heck it would be better if you never posted in this thread before rather than having a hissy fit because i know more about electronics and physics than you.

Edit: oh and you need to improve your sarcasm detection skills.


oh and chuck norris is an old, kind of washed up karate dude, who is almost as cool as Mr T, he's the butt of many a "Chuck Norris once roundhouse kicked someone so hard that his foot broke the speed of light, went back in time, and killed Amelia Earhart while she was flying over the Pacific Ocean." joke.

miniConvert
May 9, 2006, 12:41 PM
I think this news is a testament to how good the new Intel chips are. Just a few more months and everything will have transitioned over to the new way of doing things and both Windows and OS X will be very glad of the change.

I wonder if we'll finally break 4GHz now?

jayb2000
May 9, 2006, 12:48 PM
:confused:

What's this stuff all about, people saying silly things about Chuck Norris? Please, explain it to me.
www.chucknorrisfacts.com

jared_kipe
May 9, 2006, 12:48 PM
just accept that your wrong, preferably gracefully, heck it would be better if you never posted in this thread before rather than having a hissy fit because i know more about electronics and physics than you.
I know more than you *piss piss*

Electricity doesn't move at the speed of light. Massless force carriers move at the speed of light. Massive charge carriers do not, which means electrons and quarks.

Lets say you turn on an electric field that is made to accelerate an electron distance d down your wire. That electron will not accelerate until it feels the force, thus it will take t=d/c to start moving. Causality means that action at a distance requires time.

SPUY767
May 9, 2006, 12:59 PM
just accept that your wrong, preferably gracefully, heck it would be better if you never posted in this thread before rather than having a hissy fit because i know more about electronics and physics than you.

Edit: oh and you need to improve your sarcasm detection skills.


oh and chuck norris is an old, kind of washed up karate dude, who is almost as cool as Mr T, he's the butt of many a "Chuck Norris once roundhouse kicked someone so hard that his foot broke the speed of light, went back in time, and killed Amelia Earhart while she was flying over the Pacific Ocean." joke.

Prolly the SAE units confusung the brits.

on a side note, I think that this has spurred many quite good intellectual statements. Very good stuff coming out of some of our members.

And don't claim that that bit about the scholarships was sarcasm, cause if it was, you're not at all funny. And if it was the bit about the soldering, that would be hyperbole, not sarcasm.

God I love a good pissing match!

Mord
May 9, 2006, 01:07 PM
no none of that was sarcasm, i was referring to a post you quoted way back up.


and jared, effectively it's close to the speed of light and in this respect such accuracy has no real effect.

Mord
May 9, 2006, 01:31 PM
this is the ideal situation, however when the voltage of the cpu is only just enough to switch the transistors the lag as the voltage climbs up to the required level can be considerable, however if you raise the voltage that slopey bit increases in gradient untill it is near vertical and near instant, in which case the speed of the cpu is bound by the longest chain of transistors divided by the pipeline (how many clock pulses are going at a time, one can start before the previous one has finished, this has some negative effects though a pipeline of about 14 is about best) and the speed of light.

http://img300.imageshack.us/img300/6851/overclock0lv.jpg

BOOMBA
May 9, 2006, 02:53 PM
They just posted some benchmarks.

anyone know whata good 3dMarks score is?

http://macenstein.com/default/archives/303

:confused:

This allegedly set a new world record of over 20,000 ?
http://tyrou.net/screens/20797WR.png

jared_kipe
May 9, 2006, 03:13 PM
no none of that was sarcasm, i was referring to a post you quoted way back up.


and jared, effectively it's close to the speed of light and in this respect such accuracy has no real effect.

No effect? Either something is moving at the speed of light, or it never will. The "effect" gets bigger and bigger the closer to the speed of light you get.

Secondly thats not a gradient. The gradient operator turns a scalar field into a vector field. Like E=-gradV or G=-grad phi.

You don't actually understand what that graph represents, I'll give you a hint, it is a potential.

Mord
May 9, 2006, 03:25 PM
from our perspective it has no effect, to us 99% of the speed of light is just 99% of the speed of light as we do not experience the time dilation, heck nothing is moveing that fast, it's just a knock on effect thats not so much moveing.

anyway, effectively what i'm saying is right, can you just accept that so people dont think i'm wrong about everything.

jared_kipe
May 9, 2006, 03:30 PM
No you said the slope of that line depends on voltage, it doesn't you're wrong. It depends on temperature. The potential I was talking about is the electrical chemical potential. This is a statistical mechanics problem, and thus the curve "straitens" out when kT goes to zero, or rather when T goes to zero. The the potential is just the chemical potential of the electrons, and it behaves like an ideal fermi gas. Way beyond the scope of your "physics and electrical" knowledge.

Stridder44
May 9, 2006, 03:30 PM
Dude, know what you're talking about before stating complete bull. I'm amazed at how people who don't know the first thing about electronics act like they know it all. Heat does have an effect on performance and for those who don't think so are just ignorant like none other.

http://pirateship.gotdns.com/albums/webpics/applause1.gif

Mord
May 9, 2006, 03:44 PM
No you said the slope of that line depends on voltage, it doesn't you're wrong. It depends on temperature. The potential I was talking about is the electrical chemical potential. This is a statistical mechanics problem, and thus the curve "straitens" out when kT goes to zero, or rather when T goes to zero. The the potential is just the chemical potential of the electrons, and it behaves like an ideal fermi gas. Way beyond the scope of your "physics and electrical" knowledge.


never said it did not, the temperature has a large effect too.

but to get back to the original point this has no effect on cpu speed, only the maximum stable speed attainable, the speed is what you set it at, weather it works or not is decided by voltage and temperature.

Fiveos22
May 9, 2006, 03:57 PM
:confused:

What's this stuff all about, people saying silly things about Chuck Norris? Please, explain it to me.

www.chucknorrisfacts.com


I think a more accurate explanation would be to direct you attention over to 4Q.cc where they began collecting user submitted "facts" about Chuck Norris and other action legends. It was most likely due to this fact generator that Chuck Norris became a full blown (read mainstream) internet meme (http://en.wikipedia.org/wiki/Internet_memes).

The facts used to be posted so that anyone could vote on them thus creating a rating score, but now you have to sign in to rate the facts.

Check it out. (http://4q.cc/index.php?pid=top100&person=chuck)





/the vin diesel facts (http://4q.cc/index.php?pid=top100&person=vin)are much better

Fiveos22
May 9, 2006, 04:03 PM
frying Says:
May 9th, 2006 at 10:20 am

has somebody figured out eventual multiplier resistors on the macbook pro motherboard


Who has that much disposable income to just take apart a brand new laptop and futz with the jumpers?

imacintel
May 9, 2006, 04:26 PM
Yes, but now i think of this this is being only slightly overclocked. You see, with both cores adding up to 3.66 Ghz, its not a huge diffy.

Mord
May 9, 2006, 04:33 PM
you cant just add the clock speeds of an SMP system together, it does not work like that, besides even if you could this core duo would be 7.6GHz.

jared_kipe
May 9, 2006, 09:26 PM
never said it did not, the temperature has a large effect too.

but to get back to the original point this has no effect on cpu speed, only the maximum stable speed attainable, the speed is what you set it at, weather it works or not is decided by voltage and temperature.
Well of course heat doesn't affect how fast the processor is going. Unless heat leads to an error that slows down the processor.

To anyone who is saying to the contrary, would a car going 70mph at 300degK be going 70mph at 310degK?

Heb1228
May 9, 2006, 09:34 PM
oh and chuck norris is an old, kind of washed up karate dude, who is almost as cool as Mr T,
Nobody as cool as Mr. T. And don't you foo's forget it.

And Hectors right, temperature has nothign to do with processor performance. It only limits how fast the cock speed can be set to in the first place. But a 2GHz processor does the exact same amount of work hot as it does cold.

SPUY767
May 10, 2006, 08:56 AM
you cant just add the clock speeds of an SMP system together, it does not work like that, besides even if you could this core duo would be 7.6GHz.

Not to mention that the last G5 PM would have been 10 Ghz.

SPUY767
May 10, 2006, 09:09 AM
from our perspective it has no effect, to us 99% of the speed of light is just 99% of the speed of light as we do not experience the time dilation, heck nothing is moveing that fast, it's just a knock on effect thats not so much moveing.

anyway, effectively what i'm saying is right, can you just accept that so people dont think i'm wrong about everything.

The only person here claiming that anyone is wrong is you claiming that I am wrong. Now I'm not going to go on a four-page tangent about band gaps and functional properties of semiconductors, just know, that semiconductors work best in a slim range of temperatures, when that range is exceeded, errors are produced. Now, assuming that this gentleman, the one who did the overclocking, disabled the chip's on-board heat protection, which I'm guessing he had to cause Intel would not allow a 2 G chip to be OC'd to 4G, that's just begging for lots of sad-in-the-pants fan boys to come trampling down your door when they burn up their CPU's, then the chip would probably reach temperatures where it would have normally shut itself down. Not seeing any screenshot information showing temperatures, we can only speculate, but a 75% increase in speed certainly would produce a great deal more heat. This extra heat might push the Duo beyond the windows of the semiconductor's functionality. This, in turn, would cause the chip to garble some of its data in passing through, and in turn, it would be able to process less than it would at a lower clock speed generating less heat.

bousozoku
May 10, 2006, 09:31 AM
...
anyway, effectively what i'm saying is right, can you just accept that so people dont think i'm wrong about everything.

Of course, you're always right--in your own mind. Why should anyone accept anything that you say, when you can't accept anything else, even when you're very, very wrong? Of course, since you write in such a sloppy way that we have to guess at what you really mean, you can be wrong, right, or both at the same time.

To say that heat has no effect is a huge generalisation when you know that in practical matters, there is an effect.

jared_kipe
May 10, 2006, 10:41 AM
The only person here claiming that anyone is wrong is you claiming that I am wrong. Now I'm not going to go on a four-page tangent about band gaps and functional properties of semiconductors, just know, that semiconductors work best in a slim range of temperatures, when that range is exceeded, errors are produced. Now, assuming that this gentleman, the one who did the overclocking, disabled the chip's on-board heat protection, which I'm guessing he had to cause Intel would not allow a 2 G chip to be OC'd to 4G, that's just begging for lots of sad-in-the-pants fan boys to come trampling down your door when they burn up their CPU's, then the chip would probably reach temperatures where it would have normally shut itself down. Not seeing any screenshot information showing temperatures, we can only speculate, but a 75% increase in speed certainly would produce a great deal more heat. This extra heat might push the Duo beyond the windows of the semiconductor's functionality. This, in turn, would cause the chip to garble some of its data in passing through, and in turn, it would be able to process less than it would at a lower clock speed generating less heat.
Thats why they use phase cooling. To keep the temperature down. And whats this about a narrow band of temperatures? I don't consider 0-340deg K to be a "narrow band". Plus WTF is heat protection? You mean a thermometer that turns off the cpu when it gets hot right? It doesn't detect "heat" there is no such physical thing.

So I'm not sure what youre saying. You're saying temperature doesn't affect speed, but that nobody knows that temperature affects stability so they will all ruin their CPUs by attempting a 4GHz overclock and they're to stupid to think about the consequences? Right so many "enthusiasts" don't understand simple things like that. :rolleyes:

Mord
May 10, 2006, 11:38 AM
This extra heat might push the Duo beyond the windows of the semiconductor's functionality. This, in turn, would cause the chip to garble some of its data in passing through, and in turn, it would be able to process less than it would at a lower clock speed generating less heat.

no it would crash, not slow down, crash, either it works stable or it does not.

do you have any comprehension of how a digital system works? in an analog system sure this has an effect but this is digital, either it produces errors and crashes the process or it runs fine, either the 1 triggers the transistor or it falls below the trigger voltage. and the transistor does not trigger and an error cascades through the system probably bringing it down, the speed is not altered, only the outcome.

heat and voltage are the two most important variables in overclocking but on thier own they do no effect speed at all, they only effect stability.

jesus so some research instead of just reiterating your idiotic thoughts.

jared back me up here.

Of course, you're always right--in your own mind. Why should anyone accept anything that you say, when you can't accept anything else, even when you're very, very wrong? Of course, since you write in such a sloppy way that we have to guess at what you really mean, you can be wrong, right, or both at the same time.

To say that heat has no effect is a huge generalisation when you know that in practical matters, there is an effect.


Prove me wrong in any respect, please do, otherwise don't post it just makes you look immature.

I said it has no effect on PERFORMANCE, learn to read and stop being so attacking when anyone disagrees with me just because I'm not afraid of confrontation.

EDIT: jared is arguing with me about wether if an error happens and is detected (EEC) then the cache is reloaded thus slowing it down, theoretically this could have an effect however all the transistors have roughly the same trigger voltage, once a few start making errors more and more start and the process crashes, it can crash in odd ways, like a scrambled screen or a simple error message but it'll crash anyway, this is in the region of instability which either you hit or you dont if a cpu is stable and completes the benchmark no matter what the heat is it's done it and the lag due to small very unlikely recoverable errors having to be reloaded will be less than .00000000001% otherwise the errors would be too numerous and crash the system completely.

jared_kipe
May 10, 2006, 11:55 AM
I will say this much. There are things that could go wrong in a CPU that is temperture related that would slow a processor. Such as an error in the cache that the CPU just recaches. But when the errors would compound and happen so quickly at the speeds we are talking about, that even the most rare error would most likely crash the computer's operating system, or damage the CPU.

As for talking about how temperature affects components, what about HDD? The smaller the magnetic bit gets the higher the probability that kT is going to be sufficient energy to flip bits at random. Corrupting your data. Scary to think of considering how these bits are becoming smaller all the time.

Mord
May 10, 2006, 12:06 PM
the errors themselves dont cause cpu failure, it's either the heat physically damageing the cpu or the voltage has been put too high so you get tunneling and electromigration.

jared_kipe
May 10, 2006, 03:07 PM
the errors themselves dont cause cpu failure, it's either the heat physically damageing the cpu or the voltage has been put too high so you get tunneling and electromigration.
I'll say it again, Heat is a bad word. There is no such thing as heat, heat doesn't damage things temperature, or specifically the average energy of the medium would damage the cpu.
The tunneling is what would make the errors. Tunneling wouldn't directly damage the CPU.
Electomigration takes time, but is sped up by having a high temperature as the relatively stable nuclei are more free to wanter around.

Mord
May 10, 2006, 03:17 PM
errors are caused by a number of things,

First off tunneling and electromigration, they are caused by too high voltage, secondly temperature causes resistance and the transistors not getting a proper trigger voltage this is combated with a higher voltage and finally physical heat damage where the cpu melts down and incurs permanent physical damage.

only temperature tunneling and migration cause permanent damage, if the temperature if kept low and the clock speed is put up high enough that transistors are not triggered no damage occurs, same with if you overclock and the heat is not enough the damage it physically but is enough to causes resistance enough to stop transistors triggering.

SPUY767
May 10, 2006, 04:08 PM
Thats why they use phase cooling. To keep the temperature down. And whats this about a narrow band of temperatures? I don't consider 0-340deg K to be a "narrow band". Plus WTF is heat protection? You mean a thermometer that turns off the cpu when it gets hot right? It doesn't detect "heat" there is no such physical thing.


First I'm trying to put this as lay as possible, and people know what heat is. Second, semiconductors typically function as insulators at lower temperatures. So i would say that the window of optimum functionality would be somewhere in the 280-350K range. Listen, I am done arguing about this. Someone please lock the thread.

Mord
May 10, 2006, 04:15 PM
translation: i was wrong and i'm going to nitpick to avoid the fact that i was wrong and thats all i'm going to say.


^flamebait

jared_kipe
May 10, 2006, 04:28 PM
First I'm trying to put this as lay as possible, and people know what heat is. Second, semiconductors typically function as insulators at lower temperatures. So i would say that the window of optimum functionality would be somewhere in the 280-350K range. Listen, I am done arguing about this. Someone please lock the thread.
Oh yes, because you are done, that is that. :rolleyes:

What is heat? Since you're so well versed. I don't belive semiconductors are insulators at low temperature. Why would you say they work between 280-350K?? because what about http://www.tomshardware.com/2003/12/30/5_ghz_project/page13.html

They recorded a temperture of about 83K (which for those keeping track is above the boiling point of nitrogen). You could argue that the processor itself is hotter, but I would say not by much, you couldn't set up a big temperature gradient over a small distance of copper and silica. More importantly, it appears they added the liquid nitrogen before they turned on the system, thus the CPU would have been in thermal equilibrium with the liquid nitrogen (77K) when they flipped the switch. Disproving your 280-350K theory.

So stop making things up.

SPUY767
May 10, 2006, 06:26 PM
translation: i was wrong and i'm going to nitpick to avoid the fact that i was wrong and thats all i'm going to say.


^flamebait


No, I have run out of time. I have a job, and am taking doctorate classes, so I have no time for this.

jared_kipe
May 10, 2006, 09:07 PM
No, I have run out of time. I have a job, and am taking doctorate classes, so I have no time for this.
Oh wow I just read the begining of the thread.

Your proof of how a conductor near the boiling point would be worse of a conductor than when it is cold and thus super-conducting. That was nothing short of inspired. :rolleyes:

How do you explain a diamond? It has really high resistivity when it is cold, but then has almost no resistivity at around 470K?

How do you explain how P doped Silicon has a stable resistivity from about 40K on up to at least 1000K?

There are a wide array of substances where their resistivity decreases as they heat up. Saying "oh some things become superconducting when cold means that high temperature they are more resistive" is a blanket statement, same with talking about phase changes having to do with conductance. Mercury is conductive even at room temperature. Fluids can be conductors too.

SPUY767
May 11, 2006, 05:59 AM
Oh wow I just read the begining of the thread.

Your proof of how a conductor near the boiling point would be worse of a conductor than when it is cold and thus super-conducting. That was nothing short of inspired. :rolleyes:

How do you explain a diamond? It has really high resistivity when it is cold, but then has almost no resistivity at around 470K?

How do you explain how P doped Silicon has a stable resistivity from about 40K on up to at least 1000K?

There are a wide array of substances where their resistivity decreases as they heat up. Saying "oh some things become superconducting when cold means that high temperature they are more resistive" is a blanket statement, same with talking about phase changes having to do with conductance. Mercury is conductive even at room temperature. Fluids can be conductors too.


Random statements much? Piss is a conductor too? Should we make computer chips out of it?

jared_kipe
May 11, 2006, 09:42 AM
Random statements much? Piss is a conductor too? Should we make computer chips out of it?
You must mean salt is a conductor, actually NaCl is one of those conductors that gets more conductive as it heats up, thanks for playing.

Mord
May 11, 2006, 10:48 AM
i think he's referring to the water part, and infact IBM pioneered some process to use water to focus lasers to make better chips, bit random i know.

anyway, your both off topic.

jared_kipe
May 11, 2006, 11:20 AM
i think he's referring to the water part, and infact IBM pioneered some process to use water to focus lasers to make better chips, bit random i know.

anyway, your both off topic.
Oh hector, water without ions is an insulator.