Superconducting Chip Claim: Keep the Process, Change the Temperature — 28nm Logic at 5nm Performance
Move logic to liquid-helium temperatures and replace transistors with superconducting switches. That's Snowcap's bet: tape out a RISC-V first, then claim 28nm-class logic can deliver 5nm-class performance.
The argument · tap a timestamp to hear it
The ceiling for extreme overclocking is heat dissipation, not the chip
The episode opens with the overclocking industry: everyday chips rely on air cooling, higher-power ones on water cooling, and beyond that you need dry ice or liquid nitrogen — sub-zero media. Extreme overclockers push voltage to chase maximum frequencies, chips get hotter, and they need ever more extreme cooling. The host says he was once an extreme overclocker and held the world No. 2 spot — for only a day and a half. His key reference: such overclocking is only for running a few minutes of benchmarks at 8.5GHz, whereas true superconducting computing isn't about instantaneous records but about logic that runs at near-zero power.
Liquid-helium overclocking costs not the helium but the transfer line
Going from liquid nitrogen to liquid helium, physics starts to interfere. Liquid helium costs about 20 times as much as liquid nitrogen (which is about $1 per liter), but that's not the main expense: the transfer line that moves liquid helium from a large storage tank into a 3-4 kg copper overclocking pot costs $5,000 per use. Intel and Asus have both done liquid-helium overclocking, with records already past 9GHz. The host warns that material behavior changes dramatically at such temperatures — some platforms have a 'cold bug' where the system simply won't start when too cold; only platforms without the cold bug are suitable for liquid-helium play.
The lure of superconductivity isn't speed, it's zero resistance
Overclocking is about squeezing frequency out of existing materials; superconductivity is another narrative: electrons flow through superconducting materials with no resistance, making transport essentially free. Currently, most superconductors require extremely low temperatures, and room-temperature superconductivity remains in a state of repeated research claims and retractions. The key extrapolation here: if the entire chip's switching logic could operate in a superconducting state, per-operation energy would be 'orders of magnitude' better than current laptops. The host deliberately goes big then pulls back — superconductivity is still the holy grail, not yet in hand. This sets the stage for Snowcap's entry.
A former Intel CEO is betting on temperature, not process node
Snowcap Compute has just completed a seed round, and its board includes former Intel CEO Pat Gelsinger. Their bet is on building superconducting chips that run at roughly 4-15K, the liquid-helium temperature range. The structure is no longer billions of transistors but a new type of switch that works at superconducting temperatures — the Josephson junction. For now, Snowcap's goal is to first produce a RISC-V-class CPU at low temperature, making power consumption nearly free; the cost is that the entire machine must stay in the liquid-helium range, and cooling itself consumes energy.
The new variable for superconducting logic is fitting into standard EDA flows
Long-term researchers are not scarce: U.S. defense contractors have made chips integrating tens of millions of Josephson junctions, and IMEC has reached hundreds of millions, with supporting logic and EDA tools. Snowcap's differentiator is compatibility: they say this superconducting logic can fit into standard EDA flows and can be taped out on ordinary CMOS processes, using niobium titanium nitride as the material. The core conclusion the host heard: the energy-efficiency advantage from superconductivity is equivalent to using 28nm logic to achieve performance close to a 5nm chip — note this isn't a smaller process but an equivalent gain from lower power.
Cryo CMOS is only a waypoint; the real hurdle is scale
Snowcap plans to deliver its first chip in 2026, followed by a deployable RISC-V design. The intermediate state can be seen in cryo CMOS: it runs in the liquid-nitrogen range of about 72-77K, and while not superconducting, the low temperature itself lowers power. The real gap to cross is from cryo CMOS into superconductivity; the host's skepticism is 'can you scale up performance at high levels' — being able to run and being able to scale commercially are two different things.
Cold startup isn't rare; the hard part is keeping the whole machine in liquid helium
This section provides a real-world footnote on cold tape-outs: AMD's first Zen 1 A0 sample, which leaked when AMD was most cash-strapped, could only start at sub-zero temperatures on dry ice due to a design flaw, and was later fixed to run at room temperature. So low temperatures in testing and startup aren't unusual; but keeping an entire processor in the liquid-helium range involves infrastructure issues, including helium supply — liquid helium is also the lifeblood of MRI superconducting magnets, and there was a helium shortage a few years back.
Snowcap isn't selling IP; it's selling a sealed appliance
Snowcap positions itself not as an IP seller but as delivering complete systems. The host's judgment: precisely because of the infrastructure barrier of the liquid-helium range, the product will be nearly a sealed appliance, and the software stack will need to be rebuilt. A seed-stage company still has Series A/B/C rounds ahead; but they've already brought in advisors and former employees from Apple, Nvidia, and Google. On this topic, skepticism is warranted, and so is follow-up.
In their own words · checked verbatim
The cost of liquid nitrogen is about, you know, a dollar a litre. And liquid helium is about 20 times that. That's not the main cost with liquid helium. The main cost is actually the tube you need to get it from this massive compressed gas sewer into your massive 3-4 kilogram copper pot. That tube, single use, is five grand.
When you go down to those cold temperatures, the way the materials work is incredibly different. In some chips and in some platforms, you might experience what's called a cold bug. It gets too cold for the system to boot or to work.
There has been a strong effort to make room temperature semiconductors. We're not there yet. Research, if you follow that area, sometimes stalls and sometimes there's a breakthrough. If that happens, one day we may have free energy. Right now, we don't. Right now, superconducting is the holy grail.
But what if you could make a chip that was superconducting? Suddenly, you now have a processor that can do compute for next to no energy. And we're talking orders of magnitude better energy consumption per operation than this laptop, for example.
defense contractors that have designed chips with tens of millions of these Josephson junctions. And then other companies like IMEC doing hundreds of millions of Josephson junctions.
And the headline I got here is that you can essentially build chips with such superior efficiency that you're essentially building five nanometer chips on what is equivalent of 28 nanometer logic.
out comes their first A0 piece of silicon, their first Zen 1 chip. And it would only boot at subzero temperatures using dry ice because of some of the issues that they had in the design of the chip. But they fixed that and it works at room temperature.
Figures
| Liquid nitrogen price | about $1 per liter | 2:01 |
| Liquid helium price relative to liquid nitrogen | about 20 times | 2:01 |
| Liquid-helium overclocking record | past 9 GHz | 3:03 |
| Snowcap operating temperature target | about 4-15 Kelvin (liquid-helium range) | 5:07 |
| U.S. defense contractor Josephson junction integration scale | tens of millions | 7:08 |
| IMEC Josephson junction integration scale | hundreds of millions | 7:08 |
| Snowcap first chip delivery time | 2026 (planned) | 8:12 |
| Cryo CMOS operating temperature range | about 72-77 Kelvin (liquid-nitrogen range) | 8:12 |
| AMD Zen 1 first A0 chip startup condition | could only start at sub-zero temperatures on dry ice | 9:12 |
Glossary
- cold bug
- A phenomenon where a chip, when over-cooled, fails to start or malfunctions.
- Josephson junction
- A switching device made of superconducting materials that operates at extremely low temperatures.
- cryo CMOS
- Low-power CMOS that runs in the liquid-nitrogen range (about 77K), not yet superconducting.
How to listen
Hard-tech investors who want to verify whether superconducting computing is a real path or just a story, and CPU/supercomputer architects studying the chip power wall.
If you're familiar with overclocking and liquid-nitrogen cooling, you can skip the overclocking preamble from 0:00 to 2:01.