If calculation of european_guy is correct, 1000t of photoresist is enough for 200,000 wafer per month of YMTC.
1000t / 0.38 t * 1000 wafer = 2,631,578 wafer
Chinese semiconductor industry
- Thread starter Hendrik_2000
- Start date
- Status
1000t / 0.38 t * 1000 wafer = 2,631,578 wafer
According to many Korean news articles, Samsung just hired a former TSMC high level officer....
He was assigned as a vice president of Samsung.
Samsung's goal is to catch up with the advanced packaging technologies from TSMC.
Chinese government should not lose Taiwanese talents to Korea....
He was assigned as a vice president of Samsung.
Samsung's goal is to catch up with the advanced packaging technologies from TSMC.
Chinese government should not lose Taiwanese talents to Korea....
Bro we have Liang Mong Song and I believed among former TSMC engineer he is the best.
And another thing, recruiting more talent without the necessary tools to exploit it is a waste of resource. So the focus for now borrowing the famous quote from our professor @tokenanalyst is "tools, tools and more tools"
Last edited:
Actually I have calculated total photoresist (Arf + Krf + i-line): 380Kg total photoresist for 1000 wafers
This is the key info: they say "In total, there are 38 lithography exposures at 32nm/28nm, 15 of which are immersion exposures"
The big problem is of course the Arf photoresist. KrF suspension (while knowing that YMTC has anyhow an alternative) could be a "warning" salvo.
Chinese govt. has diplomatic channels with Japan to ask them what is all this suspension about. We at the moment don't have any clue on why they came out with this thing, nor we know if the intention of Japanese government is to escalate it to Arf or to blackmail with KrF for some hidden negotiation with China....
Anyhow quite a bold move by Japan. China will not forget it even in 20 years.
How hard is it to switch to Arf photoresist production?
EUV news
2023.3.7 Good news: The ultra-high-precision laser interferometer for vacuum was successfully tested in a user unit.
The displacement resolution is 5pm, and the standard deviation of displacement measurement reaches 30pm.
2023.3.7 喜讯：真空用超高精度激光干涉仪在某用户单位成功初试
位移分辨力5pm，位移测量标准差达到30pm。
真空用超高精度激光干涉仪，集成了超高精度激光稳频技术、皮米级高精度激光干涉镜组、皮米分辨力信号处理等极限技术。标志本着研究室已成功登上皮米测量的第一个坚实台阶，为本年度中德皮米激光干涉仪比对奠定了良好基础。
2023.3.7 Good news: The ultra-high-precision laser interferometer for vacuum was successfully tested in a user unit.
The displacement resolution is 5pm, and the standard deviation of displacement measurement reaches 30pm.
2023.3.7 喜讯：真空用超高精度激光干涉仪在某用户单位成功初试
位移分辨力5pm，位移测量标准差达到30pm。
真空用超高精度激光干涉仪，集成了超高精度激光稳频技术、皮米级高精度激光干涉镜组、皮米分辨力信号处理等极限技术。标志本着研究室已成功登上皮米测量的第一个坚实台阶，为本年度中德皮米激光干涉仪比对奠定了良好基础。
Why speculate about Japan's motives?
If the news is true, then its pretty clear to me that this development is part of the secret US-Netherlands-Japan agreement to restrict China's IC development
They can't, that's why they are not doing it. China has the 2nd largest fab capacity after South Korea, cut that from the market and the semiconductor shortage of 2021 will be a walk in the park. It is not as issue with the most advanced nodes because they are only manufactured in two countries and the capacity does not need to be as high as mature nodes since they are extremely expensive and only needed in select few products. That's not even mentioning the fact that enforcing such a thing would be a pain the ass since China is also the largest importer of semiconductors because it is the largest electronics exporter. The only way to ensure that sanction works would be to prohibit all electronics coming from China and that would be a disaster for the West - consumer electronics, hardware parts, cars, etc. would jump X-fold in price and utterly decimate their already ailing economies via collapse of their middle class.
escobar
Brigadier
China AI groups targeted by US sanctions have found ways to obtain restricted technology by using cloud providers and rental arrangements with third parties, as well as purchasing the chips through subsidiary companies. Rental system is a good, albeit more expensive, alternative.
Way too optimistic. I am OK if we will see China running a EUV prototype by 2027. From there commercialization will take a few more years. For parity with ASML, 2040 is more realistic.
- Status