If that's true, how would one rank pumping supercritical liquid nitrogen at high rates through a heatsink? Super-grandmaster?

Seems like the heat flow would be substantially impeded by any boiling of the LN2.

Or, for that matter, simply using a chilled copper ingot as the heat sink? There must be some threshold at which the limiting problem is getting the heat out of the die, not getting the heat out of the chip's package.

Exposing the die directly and removing heat spreaders is increasingly common in "grandmaster" overclocking circles.

Quantum computer research might be super-grandmaster

https://www.qats.com/cms/2019/08/02/quantum-computing-coolin...

They also screwed a cool little tower on top of the CPU and are wearing a Doom helmet. Don't under represent their mastery.

With no gloves, to boot!

Of course, the real grandmaster move is to use liquid helium - its boiling point is about 70C colder than nitrogen :)

Carrier freezout actually makes this non-workable -- there's a limit to how cold you can make CMOS devices before they stop functioning. To say nothing of the specific heat of liquid helium, which is miniscule compared to LN2

Using a blowtorch as a heat-source is common in extreme overclocking, to heat up VRMs and memory modules up enough to boot the machine, and keep it working.

liquid helium might get into the device itself and cause other problems. (e.g. iphones stop working in the presence of helium)

Yeah, my favorite part of the article was the warning to please use proper safety precautions with liquid N, right under the no-gloves picture...

Due to the Leidenfrost effect, incidental contact of bare skin and LN2 is not usually a problem (it boils off locally and creates a gaseous N2 barrier). Other things superchilled by it will not be as forgiving, however.

http://cookingissues.com/primers/liquid-nitrogen-primer/#sec...

Then again, pouring a liquid out of a cup without touching the liquid isn't exactly high level coordination.

How does even N2 not crack and warp the packages and boards? If I plunged my TV into liquid nitrogen I would expect it to not work - in a violent way.

You insulate the motherboard. The aim is to allow only die contact the cup your N2 is in.

There are still problems like cpu/board/components not working due to low temperature or condensation. Generally they are solved by better insulation and trying different components until you find ones that work.

Overclocking is 25% cooling and 75% picking parts that can operate at low temperatures and high frequencies.

Don't forget the crazy binning for ram sticks, cpu dies, and graphics cards. There's a reason all the top guys have major sponsorships, it takes a lot of money to find the top 1% of parts.

TIL processors can actually hit a stable 6.6 GHz if you just pump liquid nitrogen through them…

Today we didn't learn actually. They don't mention it being stable:

“It's a lot easier to control a benchmark which is always the same”, explains Rywak. "A game makes the whole process less predictable in terms of hardware load which could lead to stability issues while working on frozen hardware.”