Fanless, spinning heatsinks may be in our future!

[Archer0915]

Still won't be practicle on a CPU lenny. I don't care how much the air is agitated and pissed off The concepts are fine but this can NOT (Read NOT) properly deal with the rapid heat changes in a CPU. If the base that were mounted to the CPU were thick it would make a big diffrence but then you have to wic more heat through the air. Sure it will work but the hype is just that. This may break 3Ghz as the article says; that will be 3Ghz on a 3.7Ghz CPU

 

[LennyRhys]

I wouldn't call it hype - this is a matter-of-fact major advance in air-cooling technology. As to whether or not it will be practical on CPUs, let's just wait and see.

As for the 3GHz reference, I have no idea why the CPU chart goes only as far as Pentium 4 - that part definitely needs revising lol.

 

[Bobnova]

Especially due to P4s hitting 3.8ghz factory clock speeds

 

[MattNo5ss]

As for the 3GHz reference, I have no idea why the CPU chart goes only as far as Pentium 4 - that part definitely needs revising lol.

When the research was being done P4 Northwood A was probably at the top. It takes a long time to come up with the research topic and testing procedures/methodologies, get that approved, get/find funding, start the testing, collect data, analyze the data and come up with conclusions, write a research paper/tech report based on the findings, go through the whole peer review process, and finally get it published for public release. It seriously probably just took that long, especially with government research. I know how it can be since I've been working in a government research facility for ~4 years...lol. We still have unpublished, highly useful, head and neck response data that was collected in the 1970s... There's also a Ph.D. here who just got her proposal approved after 3 years of waiting...

 

[Archer0915]

As to whether or not it will be practical on CPUs, let's just wait and see.

Bingo. I never said this tech was not good. If the lower plate that contacted the IHS were thick, as I said, the rapid changes could be handled readily. From what I saw; it is not the case.

As far as industrial applications I think this could be great.

 

[wingman99]

This tech is for the birds, if it's so great we would be using it now so would NASA.

 

[tangletail]

My hole thought process on a spinning heat sink is, how the do you expect to conduct heat, if the device it's self is not panted directly on the CPU or GPU. If the heat sink is conducting by a single pipe or other pipes, then you can expect it create heat instead of removing it.

 

[Dooms101]

This reminds me of something... "if it ain't broken, don't fix it".
I am really not sure why heatsinks would need improvement... they need to be gotten rid of all together and replaced with a more efficient tech. The way I see it, OEM heatsinks can keep a 125watt TDP CPU at reasonable operating temps while using relatively little energy in the process. I can see how a spinning sink even with the debatable air gap would be able to do similar things, but for what benefit? With CPU's becoming more and more efficient and cloud computing taking off, I doubt most consumer PC's will require enough cooling power to justify more than just a passive cooling system within a few years. That's just consumer CPU's though.

 

[Cantebuss]

For anyone wanting more information on this, another similar article is here: http://www.ecogeek.org/component/content/article/3564 along with a paper that was published showing theory and in depth methodology.

EDIT: thought this was newer info but apparently not. Internets let me down again

 

[Theocnoob]

This reminds me of something... "if it ain't broken, don't fix it".
I am really not sure why heatsinks would need improvement... they need to be gotten rid of all together and replaced with a more efficient tech. The way I see it, OEM heatsinks can keep a 125watt TDP CPU at reasonable operating temps while using relatively little energy in the process. I can see how a spinning sink even with the debatable air gap would be able to do similar things, but for what benefit? With CPU's becoming more and more efficient and cloud computing taking off, I doubt most consumer PC's will require enough cooling power to justify more than just a passive cooling system within a few years. That's just consumer CPU's though.

Actually, the TDP of consumer CPUs has been going UP and the size of heatsinks over the past 20 years has been goin gup also.

 

[Archer0915]

Actually, the TDP of consumer CPUs has been going UP and the size of heatsinks over the past 20 years has been goin gup also.

???? My 386 did not even have a heat sink

 

[Bobnova]

Still had a higher TDP than a 286 though.
The change from P4 to c2d was the first TDP drop inx86 CPU history that I know of.

 

[Archer0915]

Still had a higher TDP than a 286 though.
The change from P4 to c2d was the first TDP drop inx86 CPU history that I know of.

That was not my point and you know it

 

[Bobnova]

I'd be really interested to see the history of CPU heatsinks really.
I saw a lot of 286s with nothing, but a few had a little green heatsink.
I saw some 386s with nothing, but many with a dinky little passive thing.
486s said "heatsink required" right on the top of the chip sometimes, the good ones even specified a fan! It's wild compared to now, CPU speed, voltage, and cooling requirements printed on the top of the chip