arguments to the arguments presented in "Theory For The Hardcore Overclocker.."Thread

[Urbanfox]

The arguments presented and my responses


“that wont do anything becouse the reason things hit the wal on ocing is that data gets lost, not sure how to explain this but i do almost positively know that that wont work”

that’s right, however, my supercooling EVERYTHING the wall is further extended.


“i think there is a limit to how cold a processor can be before it boots up.”

That’s right, but we don’t have to start at -450 F we can gradually work our way up there

“absolute 0. the absense of heat. heat, is the transpher of energy. if you were 1 degree above absolute 0, the transpher of energy would still be way to slow for it to do anything”

true, but were absolute 0 is impossible to achieve, and just 5 degrees above absolute zero electrons are still moving blazing fast enough to work


“When things approach abosloute zero, they theoretically have a mass approaching zero, too. That's gonna make things complicated.”

We not going to and cant reach absolute zero (-463 F) something like gradually building up to -450 F is more likely

“When you cool anything to near absoloute zero, the kinteic energy of the molecules slows down. They lose some vibrational energy (this is the only kinteic energy the molecules have because they are solid). Heat is the average kinetic energy of the molecules.”

True, but just 5 degrees from absolute 0 like I mentioned above, everything is still moving blazing fast


“When transistors are cool, they can switch faster. When they are hot, they'll switch slower. If they are frozen to absoloute zero, the electrons in the conduction band are completely frozen into place, afaik. If the electrons in the conduction band don't have any energy to move.”


Like I pointed above and in the opening thread were not going to (its impossible) achieve absolute 0, but something close too it, but far enough for the electrons to still move

“One last thing. Electrons and their movement can generate heat. In a CPU, it's inevitable. No matter what, you're not going to reach absolute zero. THis is just nuts.”

I never said I was going to, becus its impossible, reach absolute zero (0)


“Several problems with cold temps...the Hrdware physically contracts due to temp..this disrupts the circuitry and can actuall delaminate the layers of silicon... Processors for this reason and a couple others do not tend to run at below -120C”

that’s what im been learning about, but the common idea of super cooling everything, even at its limit will still improve overclocking than just cooling a cpu.

“well just one thing, how would you be able to power something to its full potential anyways? there would be no power supply that could make anything stable if it is running at godly speeds because you would need so much voltage it would be insane.”
Obviosly we cant do that, but the idea im trying to put out is we can push it further than anything close we get now.

“and i just had another interesting convection cooling idea, would be to water seal all of your components and make your computer like a fish take like in the (belive it is) novemeber issue of maximum pc however her computer and fishtank were seperate, if it were possable to keep a whole rig functional but waterproof then that would be I think the ultimate cooling because you wouldnt need any fans what so ever its simple convection, heated water from computer componets would rise, and cooler water would takes its place the heated water would come to the surface release the heat into the air, and then return to the bottom, however it would be preatty hard to water proof your computer to that level....”

nice idea, but the cooling I describe above not only could be fanless but would cool more effectively than a fish tank by more than 1000%++.

“Something to that extent was done, a guy took a non-conductive liquid (Perfectly distilled water works as well ) and submerged an entire computer inside of it, took pipes throught the liquid and shot LN2 through the pipes, cooling the liquid greatly, and a great oc ”

maybe, but did he shatter (more like obliberate) published 3d mark records and other benchmarks? No. that’s whats to be acclomplished.

“Bro, whatever it is you're smoking, I suggest you to stop.”
I think they probably told that to a lot of great inventors, although im not putting myself to the levels they have so-rightly attained lol

“ln2 is liquid nitrogen isnt it? that is preatty bad for cooling...... basiclly because from what i've heard if you keep it sealed, it will blow up, and the fact that i think its on 1ml of liquid nitrogen wil displace something like a liter of air..... yea its just a tad deadly”
liquid nitrogen, but about liquid helium? Better cooling too. And as for blowing up and crap, as many of u have probably already figured out, im going to try to accomplish something like this, and believe me when I say, the guys that will be handling the cooling are professionals, very, very much so.

“sorry to burst your bubble but semiconductive material doesnt like temperatures under 100 kelvin. when the temp drops below 100 kelvin there are not enough electrons in the conduction band for the silicon to conduct. secondly, as we all know (j/k) at near absolute zero metals (conductors) become superconductive (some are high temp superconductors that become super conductive at like 100 kelvin but i wont get into that) and that is bad. a superconductor can pick up stray magnetic or electric flux like a black hole. this excites currents in the metal and i think you can see where im going with this.”

Ahh, but the cooling procedure has yet to be finalized.

“somewhere around -100C, the electronics become a superconductor and won't work. they short themselves out.”
Why do wires short out? Heat. Plain and simple. Too much and the wire literally melts. But we’ll discuss this later.

“im talking in kelvin, your talking in celcius. there is no negative in kelvin, 0 is as low as it goes. and no, -100 celcius is not where things get shorted, its -173 celcius (100 kelvin). also, the electronics will not turn into superconductors anywhere near -100. only special compounds have a superconductivity at that high of a temp. most things become superconductive at about -220 celcius.”
Once again no exactl temp has been finalized, but u ppl r missing the idea.

“that is why liquid nitrogen cooling is still effective. it usually doesnt cool to below -173 celius.”
Liquid nitrogen probably wouldn’t be all that great. And this is coming from ppl who do this for a living, and are very good at it.

“Gang! The things we dream up for Max OC! If you remember back many years ago, the Cray Supercomputers were liquid nitrogen cooled. Don't know how they got around expasion/contraction problems, though. Have put serious thought before into the fishtank idea cooling the fluid to around zero degrees.. Brings up some problems of its own. First,for longevity of computer, liquid medium would have to be non conductive, non corrosive,non reactive, and with a lower freeze point than water. Fluid would need to have higher specific gravity than water, or some way thought out to deflect and drain condensation from surface level of fluid. Have thought about certain very thin aviation oil like chemicals used for corrosion protection of canon plug electrical connectors as a medium. Next problem I see is that fluid will eventually wick into connectors (IDE Ribbons,PCI & AGP & RAMM slots, etc.) from natural seeping and vibration, so all connectors would need to be sealed. I'm not real hip on permanantly gluing my vid card and RAMM to the mothreboard. I've seen the way hydraulic fluid can seep even under sealant on sheet metal, so I don't think silicon would work. If you don't have a problem with permanant components with no option to upgrade or trade out, give it a shot.Lol!!!”

oh a big one. 1) why is everyone thinking liquid lol! We can cool to to -450 F with gas, remember that, and no where have I talked about the experiment ever having any liquid. 2) permanent components lol? This is the mother of god overclocking experiments, which burning everything is of no concern, its getting the benchmark scores that r.

“you could have a closed tank full of radiator fluid. it wouldnt matter if the liquid got into the connectors as long as they were connected right because the copper would always be touching the terminals. i have submerged motherboards in distilled water for days at a time and didnt have any problems with it. just dont plan on your components coming out looking clean, submersion is a dirty game.”

Argh lol no liquid!

‘Ummm....yeah, look here”
nice, but what about these excerts from the article “This is the clock rate we achieved with the P4. At this clock rate, however, benchmark tests were no longer possible.” Wtf? 5ghz+ nice but no stable for even 1 benchmark? Who cares. 2) ermm look at the cpu-z pic, that looks like a special cpu from intel, not one u pick up at newegg. 3)have they ran out of rent money whats up with the balcony? Wheres there uber lab? Just a little comic relief.



also while looking at the toms hardware thread look here http://www.tomshardware.com/cpu/20031230/5ghz-12.html at the cpu-z pic. wonder what "special" p4 chip from intel that is...

[babalouj]

yo, liquid helium is 3 kelvin (-270 degrees celcius). thats pretty freakin close to -273.15 (absolute zero).

Why do wires short out? Heat. Plain and simple. Too much and the wire literally melts. But we’ll discuss this later.

oh an as temperature decreases resistivity decreases which makes electronics short out too. the lack of resistance is a short, im sure joep will back me up on this.

[babalouj]

check this out. http://www.sciencefriday.com/pages/2...r2_011400.html scientists have come within .9999999999% of absolute zero, thats like 0.0000000001 kelvin. if thats not close enough for you then i dont know what it

[tennvols_69]

Originally posted by babalouj
yo, liquid helium is 3 kelvin (-270 degrees celcius). thats pretty freakin close to -273.15 (absolute zero). oh an as temperature decreases resistivity decreases which makes electronics short out too. the lack of resistance is a short, im sure joep will back me up on this.

i thought an idiot dropping a screw in a case with power applied caused shorts? lol ill have to agree with you babalouj

[Urbanfox]

Originally posted by babalouj
check this out. http://www.sciencefriday.com/pages/2...r2_011400.html scientists have come within .9999999999% of absolute zero, thats like 0.0000000001 kelvin. if thats not close enough for you then i dont know what it

actually we want to stay about 5 degrees away from absolute zero. and remember ppl, im talking about cooling everything, as in putting the whole cpu (minus hdd etc) in a controlled chamber with GAS NOT LIQUID (how many times must i say this for someone to stop saying "how will u seal the videocard from the liquid blah blah blah"). im betting cooling everything and not just the cpu and chipset for toms hardware's 5ghz+ overclock will reduce higher results than what we have now at stable levels.

oh yes, and toms hardware screw u. 5ghz my *** u cheated. special board from asus REALLY special cpu from intel u cheating bastards. big deal world record u got special samples from the **** company who's product ur trying to oc!!! try picking up a reg mb and cpu.

ah huh. enough ranting.

[babalouj]

yeah toms really cheated on that one!!! hand picked chip huh?? i bet they made it just for them to show the world that intel overclocks better. cheaters!!

[Urbanfox]

Originally posted by babalouj
yeah toms really cheated on that one!!! hand picked chip huh?? i bet they made it just for them to show the world that intel overclocks better. cheaters!!

lol, but we still love intel....and how do i get my hands on one of em....

[babalouj]

call toms and beg

[Urbanfox]

Originally posted by babalouj
call toms and beg

sorry, im born in the USA. if i want it bad enough ill take it from the those nazis not beg.


edit: and i mean that as in free world v.s. nazism not the usa doing whatever it plz's such as lore as invading iraq for its oil. and anyone who says that can f off. seriously. but have a happy new year.

[ShortBtwnHdset]

Hey Urban Fox! Replying to the comment some of us have had to your thread. Some of those commments were'nt directed at you. Threads can sometimes take on a life of thier own. Many of the discussions seem to be aimed at a more practical/doable range of project. Meanderings in our head, as it were. The reason liquid cooling is more practical is the nature of heat transfer. Heat flows best from solid to like solid, next best is solid to liquid. Solid to gas is the least efficient, as any one running HSF's for cooling probably realizes. Besides, I thought most gases became liquid or solid as they approached Kelvin. Boundary layers also become a problem. Gas or liquid along a surface has a certain resistance to flow, and those molecules closest to the surface flow the least. This directly counters your cooling effect, and the boundary layer is thicker with gas. You will probably still need to find some way to move the gas over those components with highest heat output. Aircraft use vortex generators to disrupt the flow of air over a wing, and break up this boundary layer for greater lift (lower stall speed) and control of aircraft in lower airspeeds.