If China’s already at 5nm, they’re getting close. Smallest in use now is 3nm, I think.
This poses the question, how long until China can achieve its goals without Taiwan and TSMC?
I imagine they will be able to produce bleeding edge chips within a year or two tops. What’s even more interesting though is that China is now producing chips on different substrates like gallium that don’t have any equivalents in the west. https://www.scmp.com/news/china/science/article/3323448/luo-weiwei-former-nasa-scientist-who-became-chinas-semiconductor-trump-card
Idk about 2 years, but they can definitely do it by 2030.
It took TSMC a while to get to that point if I’m not mistaken, so even with all the financial might of the CCP, I still doubt it will happen THAT quickly.
Sure, it’s hard to predict exact timing on these sort of things, so another 5 years is a plausible timeline. It does tend to be easier to do things the second time around. When you’re doing something from scratch then you don’t know what the right approach will be, you’re going to have false starts, and so on. When the general idea has been proven, it’s much easier to replicate because now you know what the general direction to pursue is.
deleted by creator
The only reason I can see for china to invade china 2 is if usa places nukes there
deleted by creator
So you said the same as me, but framed like a cia stenographer would. Why tho? You gain something from convincing people that china bad?
deleted by creator
I always saw it as TSMC potentially being a key reason for China to invade Taiwan so they get their hands on semiconductors, with this being LESS likely now, instead of more, because China no longer needs TSMC and their research.
deleted by creator
Because it is. They’ve also made it very clear that they intend to follow a path of peaceful unification and the only reason that hasn’t happened already is because of US meddling and military threats.
deleted by creator
It seems you’re mostly right (except there are rumours that a Taiwan takeover is part of China’s long term plan for economic dominance).
I’m worried at how many China shills there are around here.
deleted by creator
Don’t you have that backwards? Without TSMC’s outstanding technology, the island’s value decreases, both for China and for the USA. Conventional wisdom is that reduced tensions also reduces the risk of war.
deleted by creator
The challenge more is that even if they can produce the chips, the machines they are using they can’t get parts for. EUV is still out of reach. Probably another 5-8 years.
It’s kind of funny how you just confidently pull a number out of your ass here. On what basis do you think China is 5-8 years away from making their own EUV machines exactly?
The reality is that China is already starting to experiment with their own EUV approaches. A lot of the complexity in ASML machines comes from the fact that they need to be compact enough to ship around the world. China is looking at the approach of using an accelerator instead.
I’ve been following this story for 5 years. There’s lots of hopey articles like this over that time. SMEE was supposed to release a 28-nm DUV machine last year in 2023, then 2024 summer, then 2024 December. To date no release. Most of these innovations by thr Chinese are academic discoveries that will take years further to productionize. Generally if the Chinese say they will do something new-technology-wise next year, it will happen 2-5 years after when they say it will.
I find that generally we tend to see tech progress in China outpace predictions. For example, nobody expected China to start producing 7 and 5nm chips by now, nobody thought Huawei clusters would be able to compete with Nvidia, etc. Of course, doing new things is always tricky, and 2-5 years certainly sounds plausible to me.
They’re producing these 7 nm and 5 nm chips on ASML machines they bought or smuggled into the country. The problem is they can’t fix them on warranty if they break and they can’t figure out how to reverse engineer them yet. I’m hopefully but I’m realistic.
They’re not doing that, they’re using an alternative approach to EUV
