Hacker News Reader

Comment by kken | Hacker News Reader

so this is again one of these "material x was slightly better in a lab experiment" -> "we conclude that it certainly can be used as part of a complex technology" press releases.

I am not too fond of this, but obviously thats the easiest way to explain the significance of research results.

Getting a new material into a semiconductor manufacturing process takes years or even decades. Whatever material will be in TSMCs "1nm" process, it is already known.

View on HN

I remember 15 years ago reading about EUV litography and a lot of people were saying that it looks like a pipe dream because they don't know of any way to produce a EUV light source strong enough, outside nuclear explosions and the like.

So it took about 10 years to get into production. OP isn't claiming it's impossible, just that it's far from being practical. You could definitely say that about EUV 15y ago

It took 25 years to commercialize EUV. And that is from the point where a decision was made to industrialize it, not from the first scientific report about EUV radiation.

edit, see Zeiss page on EUV: https://www.zeiss.com/semiconductor-manufacturing-technology...

Some of people and organizations think in terms of decades and know that "soon" doesn't mean "tomorrow".

The world will not be free of short-term thinking until the last MBA is strangled with the entrails of the last PMP.

Yes, but do you remember all the BS PR releases about new "promising" techniques and materials that would "revolutionize" this or that process, that we read 15 and 20 and 30 years ago, that never materialized at all?

nanoimprint (heard it is making a comeback in Japan), x-ray litography, mass electron beam litography...

It looks like small piece of tin is where the magic begins [1]:

> An EUV system uses a high-energy laser that fires on a microscopic droplet of molten tin and turns it into plasma, emitting EUV light, which then is focused into a beam.

is the spectrum naturally pure, or is there some filtering/selection going on?

> The tricky thing with EUV light is that it’s absorbed by everything, even air. That’s why an EUV system has a large high-vacuum chamber in which the light can travel far enough to land on the wafer. The light is guided by a series of ultra-reflective mirrors

The system uses 1MW of power!

1. https://medium.com/@ASMLcompany/a-backgrounder-on-extreme-ul...

There's filtering, but the spectrum should have (near) discrete wavelengths that it emits at selectively

ahahah, that's not even the craziest thing about EUV. it's that the whole process has two actual laser shots, there is a preshot that heats the tin, forms a plasma that is shaped like a lens. and then comes the stronger, main shot.

https://www.degruyter.com/document/doi/10.1515/aot-2017-0029... see pre-pulse part.

(RIP memristors. Those have similarly been on the cusp of revolutionizing everything for … I think over a decade now?)

Yup, I wrote about them from an HP conference in 2010 and remember being told about them for a roadmap item well before that: http://ewams.net/?date=2010/09/05&view=hptechforum2010

Another one of those "right around the corner" things that is always coming soon.

World is full of technologies that would be cool if you also had a Time Machine. But other tech outpaces is so if it is born behind it never takes off.

Some of it is pretty fun for playing what if with (enforced) pastoralism. Using primitive materials with a modern understanding of them, instead of limiting yourself to historically accurate processes.

I take the slightly more patient approach with this. LCD used very old discoveries, something need to mature through a few industrial cycles before it gets discarded I guess.

The rumor I heard was that HP licensed all the rights to the tech away and development has stalled since no entity can put their hands on enough of the IP to do anything with. : / So development has pretty much stalled since the last time HP showed of The Machine with memristors.

The patents should be nearly expired by now.

A quick patent search shows that we still have a nontrivial amount of time on them.

Patents on fractal compression and related techniques are all expired now. There might still be people who remember how to use them. Do we see any use of it, now?

IMO compression was a lot more important when we had modems and disk sizes measured in megabytes or single digit gigabytes. Once you have "good enough" compression then you solve most issues and in many areas where there used to be compression they've been increasingly removing it. Compression causes latency after all and some of the biggest impediments to processing speed now is latency.

So unlike before, compression has stopped being a primary driving factor in the engineering of many types of devices and types of software.

Raw data compression was once an attractive application, but the method is more general. That said, streaming services still pay a great deal for bandwidth and storage in edge nodes. Cell- and keyframe-oriented video compression are reaching their limit.

We are still interested in feature extraction and scene composition, which both benefit from more compact representation.

To compete with hardware-assisted decoding in display hardware, fractal rendering would probably need to be coded to run in Vulkan.

fractal compression was always a scam.

The renewed interest in them and the PR hoopla is closer to 14 years...

3D XPoint, used in Optane memory, used memristors.

It's funny that it's debatable what's a memristor and what's not: "Intel has stated that 3D XPoint does not use a phase-change or memristor technology, although this is disputed by independent reviewers." (Wikipedia)

In that case it seems the real problem is that the killer app hasn’t been found yet.

Reading the article I wondered if tapered interconnects were the solution. Fat interconnects with thin fins? Probably harder to calculate inductance and capacitance, but these people deal with complexity all day.

Not entirely sure how you’d avoid voids with tapers facing upward, but you can orient them in five out of six directions. And I’ve seen a few examples of the weird games they have to play with light to get the crisp corners they are after so maybe tapers would be easier?

The actual discovery is reported in one of the last paragraphs:

>The breakthrough relates to a new set of materials that can create monolayer—or two-dimensional (2D)—transistors in a chip to scale the overall density by a factor matching the number of layers. The teams at TSMC and MIT have demonstrated low resistance ohmic contacts with a variety of existing semiconductor materials, including molybdenum disulfide (MoS2), tungsten disulfide (WS2), and tungsten diselenide (WSe2).

So it is a better way to make a contact to a semiconductor material that noone knows how to manufacture reliably on a large area wafer.

Reply on news.ycombinator.com