This new transistor from China might end silicon’s reign and turn your next laptop into a speed demon
China has begun mass production of next-generation processors based on molybdenum disulfide instead of traditional silicon semiconductors1. According to Professor Li Hongge's team at Beihang University, these chips merge binary and stochastic logic to achieve better fault tolerance and power efficiency for applications like touch displays and flight systems2.
The breakthrough came through developing a Hybrid Stochastic Number (HSN) system that combines traditional binary with probability-based numbers2. This innovation helps overcome two major challenges in chip technology - the power wall from binary systems' high energy consumption, and the architecture wall that makes new non-silicon chips difficult to integrate with conventional systems2.
- AzerNews - China mass-produces silicon-free chips ↩︎
- SCMP - China starts mass production of world's first non-binary AI chip ↩︎ ↩︎ ↩︎
Beyond 1s and 0s: China starts mass production of world’s first non-binary AI chip
China’s AI chip overcomes traditional computing barriers and will be used in touch displays, flight systems and aircraft navigation.Zhang Tong (South China Morning Post)
like this
mystic-macaroni
in reply to Zerush • • •Fredselfish
in reply to mystic-macaroni • • •Zerush
in reply to mystic-macaroni • • •Wom't make a difference with any current OS, none of those would work with the new arquitecture, way different from the current one. One thing is the new hardware and another is the current lack of any soft or OS for it. Not even DOOM would run in it.
Maybe in one or two years it would make sense to change your PC.
Another alternative to silice are diamonds, there the chips are with the same arquitecture as those from silice, but with the advantage that they support much more heat to the point that they don't even need refrigeration, apart the electric apabilities of diamond is way better as those from silice, that permits a way higher speed and stability. The price isn't much higher as the one from normal chips with sythetic diamonds. They are already in use, even with manufactories in Spain.
Single-Crystal Diamond Wafers
www.df.comlike this
Maeve likes this.
Blue_Morpho
in reply to Zerush • • •That's not a diamond transistor company. They are making diamond wafers to mount to traditional silicon. It's a heatsink.
Here's a good overview of the current state of the art in diamond transistor manufacturing.
youtu.be/NLmd5vL0zmk
- YouTube
youtu.bebufalo1973
in reply to Zerush • • •Grapho
in reply to mystic-macaroni • • •Maeve
in reply to Grapho • • •FEIN
in reply to Grapho • • •irelephant
in reply to Grapho • • •Jankatarch
in reply to irelephant • • •☂️-
in reply to Jankatarch • • •idriss
in reply to mystic-macaroni • • •☆ Yσɠƚԋσʂ ☆
in reply to Zerush • • •like this
Maeve likes this.
4am
in reply to ☆ Yσɠƚԋσʂ ☆ • • •like this
Maeve likes this.
Fluke
in reply to 4am • • •All hanging off a Dutch company that makes arguably the most complicated machine the human race has ever built. (EUV lithography is absolutely astounding, when you have even a passing understanding of the tolerances required to make it work.)
ASML, manufacturer of photolithography machines.
davel
in reply to Fluke • • •like this
Maeve likes this.
HiddenLayer555
in reply to ☆ Yσɠƚԋσʂ ☆ • • •☆ Yσɠƚԋσʂ ☆
in reply to HiddenLayer555 • • •jambudz
in reply to HiddenLayer555 • • •pineapple
in reply to HiddenLayer555 • • •ragas
in reply to ☆ Yσɠƚԋσʂ ☆ • • •Yes gallim arsenid transistors wold be about 10 times faster. But also about 100 times more expensive.
(Numbers pulled out of my ass.)
☆ Yσɠƚԋσʂ ☆
in reply to ragas • • •ragas
in reply to ☆ Yσɠƚԋσʂ ☆ • • •The problrm is that this is already calulated st scale.
Silicon isn't the best material for semiconductors, it never was. What makes silicon special is that it is the cheapest material for semiconductors.
So unless there is some kind of scientific breakthrough with one of the other semiconductor materials, this equation will not change.
☆ Yσɠƚԋσʂ ☆
in reply to ragas • • •ragas
in reply to ☆ Yσɠƚԋσʂ ☆ • • •Exept the fist transistor wasn't even silicon it was germanium.
en.wikipedia.org/wiki/History_…
Silicon is used because it is inherently cheaper.
aspect of history
Contributors to Wikimedia projects (Wikimedia Foundation, Inc.)☆ Yσɠƚԋσʂ ☆
in reply to ragas • • •ragas
in reply to ☆ Yσɠƚԋσʂ ☆ • • •After considering multiple other options for mass production.
Germanium transistors are still mass produced to this day, but only for the niche products where silicon doesn't cut it.
The semiconductor industry is still constantly looking for other materials to use. Graphene is a big contender.
You act like the industry can switch to a bunch of materials and have better products but they are just too lazy to do it.
But actually more likely is that through its physics and availability silicon is just the best material for the job.
Of course unless some scientific breakthough comes along but it is not here yet.
Looking into history is distorted here because you only see what succeeded.
☆ Yσɠƚԋσʂ ☆
in reply to ragas • • •What I keep explaining to you here is that silicon is not inevitable, and that it's obviously possible to make other substrates work and bring costs down. I've also explained to you why it makes no business sense for companies already invested in silicon to do that. The reason China has a big incentive is because they don't currently have the ability to make top end chips. So, they can do moonshot projects at state level, and if one of them succeeds then they can leapfrog a whole generation of tech that way.
You just keep repeating that silicon is the best material for the job without substantiating that in any way. Your whole argument is tautological, amounting to saying that silicon is widely used and therefore it's the best fit.
ragas
in reply to ☆ Yσɠƚԋσʂ ☆ • • •Without substantiating? I linked a Wikipedia article as a source, which explains quite a lot of the reasoning for choosing silicon.
The only thing that you reiterate here is economics of scale and you haven't provided any source that substantiates that there are other materials where the economics of scale might lead to a better and/or cheaper product.
☆ Yσɠƚԋσʂ ☆
in reply to ragas • • •ragas
in reply to ☆ Yσɠƚԋσʂ ☆ • • •☆ Yσɠƚԋσʂ ☆
in reply to ragas • • •ragas
in reply to ☆ Yσɠƚԋσʂ ☆ • • •☆ Yσɠƚԋσʂ ☆
in reply to ragas • • •ragas
in reply to ☆ Yσɠƚԋσʂ ☆ • • •locuester
in reply to ragas • • •ragas
in reply to locuester • • •☆ Yσɠƚԋσʂ ☆
in reply to ragas • • •Oh right, the famous laws of physics that apparently decree silicon must forever be the cheapest material. Let me check my physics textbook real quick. Yep, still says nothing about global supply chains and sixty years of trillion-dollar investment being a fundamental force of nature.
Silicon is cheap because we made it cheap. We built the entire modern world around it. We constructed factories so complex and expensive they become national infrastructure projects. We perfected processes over many decades. That's not physics, that's just industrial inertia on a planetary scale.
To claim nothing else could ever compete requires ignoring how technological progress actually works. Remember when aluminum was a precious metal for royalty? Then we figured out how to produce it at scale and now we make soda cans out of it. Solar panels, lithium batteries, and fiber optics were all once exotic and prohibitively expensive until they weren't.
As you yourself pointed out, germanium was literally the first transistor material. We moved to silicon because its oxide was more co
... Show more...Oh right, the famous laws of physics that apparently decree silicon must forever be the cheapest material. Let me check my physics textbook real quick. Yep, still says nothing about global supply chains and sixty years of trillion-dollar investment being a fundamental force of nature.
Silicon is cheap because we made it cheap. We built the entire modern world around it. We constructed factories so complex and expensive they become national infrastructure projects. We perfected processes over many decades. That's not physics, that's just industrial inertia on a planetary scale.
To claim nothing else could ever compete requires ignoring how technological progress actually works. Remember when aluminum was a precious metal for royalty? Then we figured out how to produce it at scale and now we make soda cans out of it. Solar panels, lithium batteries, and fiber optics were all once exotic and prohibitively expensive until they weren't.
As you yourself pointed out, germanium was literally the first transistor material. We moved to silicon because its oxide was more convenient for the fabrication tricks we were developing at the time, not because of some cosmic price tag. If we had poured the same obsessive investment into germanium or gallium arsenide, we'd be having this same smug conversation about them instead.
Similarly, graphene isn't too expensive because physics. It's too expensive because we're still learning how to make it in bulk with high quality. Give it a fraction of the focus and funding that silicon has enjoyed and watch the cost curve do the same dramatic dive. The inherent cost argument always melts away when the manufacturing muscle shows up.
The only real law at play here is the law of economies of scale. Silicon doesn't have a magical property that makes it uniquely cheap. It just has a sixty-year head start in the world's most aggressive scaling campaign. If and when we decide to get serious about another material, your physical laws will look a lot more like a temporary price tag.
ragas
in reply to ☆ Yσɠƚԋσʂ ☆ • • •Still no source but ok.
I never said that.
True. However that doesn't mean that, at the current point of technology available to us, scaling a different material in the same way woul get us chepaper or better computing.
I never claimed that.
... Show more...However this is untrue. There are regular atempts to use Gallium in silicon processing and gallium transistors are in fact already mass produced for power handling applications. So not even the scaling argument holds true about Gallium. The issue is just that gallium transistors are still inherently more
Still no source but ok.
I never said that.
True. However that doesn't mean that, at the current point of technology available to us, scaling a different material in the same way woul get us chepaper or better computing.
I never claimed that.
However this is untrue. There are regular atempts to use Gallium in silicon processing and gallium transistors are in fact already mass produced for power handling applications. So not even the scaling argument holds true about Gallium. The issue is just that gallium transistors are still inherently more costly to produce.
softhandtech.com/is-gan-better…
We would need a technological breakthrough to make Gallium viable against silicon. But with current technology it is just worse than silicon from a price/performance standpoint.
True except as I was saying and you are saying here too, we would need some kind of technological breakthrough to make graphene viable.
This is on a Level of development where they hope to have first viable products for some edge cases in the next 10 to 15 years.
semiengineering.com/the-race-t…
blacksemi.com/2025/02/06/black…
So yes in this case we could say invest all into graphene and nothing else. Which will mean that all other semiconductor innovation stops so that maybe in 15 years we have cool brand new graphene computing, or maybe not.
As explained above, in reality there is just no other option available that makes any sense. If you have any other option that will work please tell me and only me so that I can start founding my startup.
Because the big player sure as hell know that silicon shrinking is not working any more and researching for alternatives.
inf.news/en/science/0e165f2238…
Except that is actually the case. Silicon is a widely available material that is easy to work with. And through that beats many other materials immediately.
That doesn't mean that it will stay that way forever. But it is disingenuous to say that just switching to something else will be better.
Oh I don't disagree with you here. The question is just how big will the price tag be. Because with what we currently can foresee all other price tags are still pretty enormous.
GaN vs. Silicon: The Next Generation of Semiconductor Technology - SoftHandTech
Travis Sharrow (SoftHandTech)☆ Yσɠƚԋσʂ ☆
in reply to ragas • • •ragas
in reply to ☆ Yσɠƚԋσʂ ☆ • • •☆ Yσɠƚԋσʂ ☆
in reply to ragas • • •Eugene V. Debs' Ghost
in reply to ragas • • •☆ Yσɠƚԋσʂ ☆
in reply to Eugene V. Debs' Ghost • • •Zerush
in reply to ragas • • •Riskable
in reply to Zerush • • •Poach
in reply to Riskable • • •hedge_lord
in reply to Zerush • • •pineapple
in reply to Zerush • • •Um actually even smooth water pipes have a lot of tubulance since all materials aren't perfectly smooth so the edges of the pipe have a lot of turbulance which dramatically slows down the water from the theoretical maximum.
ShinkanTrain
in reply to pineapple • • •pineapple
in reply to ShinkanTrain • • •Agent641
in reply to ShinkanTrain • • •Etterra
in reply to pineapple • • •pineapple
in reply to Zerush • • •B0rax
in reply to pineapple • • •tomenzgg
in reply to B0rax • • •Meldrik
in reply to Zerush • • •orc_princess
in reply to Meldrik • • •frightful_hobgoblin
in reply to Zerush • • •en.wikipedia.org/wiki/Abundanc…
silicon – 28.2% of earth's crust; 7,200,000 tonnes extracted per year
bismuth – 0.00000085% of earth's crust; 10,200 tonnes extracted per year
Wikimedia list article
Contributors to Wikimedia projects (Wikimedia Foundation, Inc.)buddascrayon
in reply to frightful_hobgoblin • • •Zerush
in reply to frightful_hobgoblin • • •gandalf_der_12te
in reply to frightful_hobgoblin • • •doesn't it say "molybdenum disulfide" in the text?
molybdenum - 1.2 ppm (0.00012%); 227,000 tonnes extracted per year
frightful_hobgoblin
in reply to gandalf_der_12te • • •Hirom
in reply to frightful_hobgoblin • • •CommanderCloon
in reply to frightful_hobgoblin • • •gandalf_der_12te
in reply to Zerush • • •certified_expert
in reply to gandalf_der_12te • • •gandalf_der_12te
in reply to certified_expert • • •ArcaneSlime
in reply to Zerush • • •Alsjemenou
in reply to Zerush • • •Zerush
in reply to Alsjemenou • • •