Chapter 319 Bottlenecks in the CPU Industry
Chenchen did not answer Xingchen's guess because his goal was not clear.
Because under the current political landscape of the earth, even if you want to do it yourself, there are many things that cannot be completed.
Chen Chen had no experience in thinking about how to advance the next stage of civilization on a planet without a unified government.
So for the time being, we can only take one step at a time.
After Chen Chen and Xingchen discussed and decided, Xingchen spent about 3 hours integrating all other functional modules such as the processor and baseband, and conducted multiple simulation tests to ensure that there were no problems at all.
Then he sent all the information and a complete set of 5G communication standard documents to Chen Jiansheng, asking him to find time, hand over the processor to Intel for production, and then took the 5G communication standard to prepare to attend this year's communications industry conference.
Generally speaking, a chip foundry starts to obtain a chip production order from a customer.
It takes about another month to produce all silicon wafers for the customer within 60 days.
In other words, it takes at least 80 to 90 days from start of construction to leaving the factory.
The time is a bit long, mainly because the mature lithography machine 193i used has been used for many years and has very low accuracy. To be used to produce 14nm chips, it requires too many complex technologies to increase the accuracy.
Traditional craftsmanship, achieving 10nm is basically the limit. If you want to continue to explore, you must change to 7nm and 5nm, you must change to new technology.
This new technology, namely Intel, Samsung, and TSMC, is all waiting for EUV extreme ultraviolet lithography technology. It uses projection lithography technology with extreme ultraviolet light with a wavelength of 13.5nm as its working wavelength
Because of higher accuracy, process flow and complexity can be greatly reduced.
Therefore, if EUV lithography technology can be mature and put into mass production, the production time of chips will be greatly shortened and the product yield will be greatly improved.
According to current estimates, factories using EUV technology can leave the factory for as fast as more than 20 days, and the time is only one-quarter of that of traditional lithography technology.
However, EVU technology is not mature at present, and there are still many key problems that have not been solved.
Intel, Samsung and TSMC Semiconductor have invested huge amounts of money to ASML, the world's top lithography machine manufacturer, to develop EUV technology in order to maintain the sustainable development of the semiconductor industry.
EUV technology is the hope of continuing Moore's Law and allowing the 7nm and 5nm process to be successfully mass-produced.
Moreover, EUV technology is also a rare opportunity to overtake in Donghua's domestic semiconductor production technology and surpass internationally.
In traditional craftsmanship, Donghua is no longer able to surpass the West. Even if it exceeds the situation, it will soon be outdated and can only be used to produce low-level chips.
Donghua began to conduct research on EUV-related aspects in the 1990s.
In 2002, the first set of EUV extreme ultraviolet lithography principle device in China was developed.
In 2008, under the leadership of the Chinese Academy of Sciences, many universities and research institutions have participated in the research.
In 2017, the research on key technologies of EUV extreme ultraviolet lithography has reached the final stage of research and project acceptance is about to be carried out.
In fact, at this time, it was the latest in Donghua and International's lithography machine technology, and it was almost within the same generation.
And the key is that whenever domestic technical research is done, we must do a complete set of technologies.
Including various materials and equipment used in the production process, there will be overall technical research. Because if you don’t do it yourself, no one will sell you other equipment and processes.
ASML can purchase the best materials in different countries.
The prospects seem to be very good, but like ASML, it is still impossible to mass produce. Domestic EUV is more difficult to mass produce than foreign countries because the supporting system cannot keep up and overall research and development is required.
Against this background, Intel gave Xingchen Semiconductor's 7nm wafer factory drawings and technology, which are still semi-finished products, and there are multiple sets of immature backup solutions.
Regarding the realization of 7nm, is it necessary to use EUV technology? Can EUV technology catch up with the mass production of 7nm? Intel itself has no idea.
Therefore, if we follow Intel's drawings, the factory construction will eventually move to the experimental stage, and traditional processes and EUVs must be tested instead of direct mass production, and Intel itself is the same.
However, after the modification of the stars, this drawing has been perfect, mature, and confirmed.
Xingchen solved the problem of EUV difficult labor.
The equipment required for factory construction already includes EUV equipment, and both domestic and ASML are available.
The main reason is that the production capacity of this thing is too low. ASML ships only 12 units per year, and the domestic production capacity is even lower. In order to increase the production and scale of the factory in the future, buy one more if you can buy one more.
The undertaker of the domestic EUV project is also one of the shareholders of Xingchen Semiconductor, and one of the only technical shareholders besides Xingchen Technology.
While building the factory, Xingchen Technology will help them to mature the equipment that is still in the laboratory to mass production.
Moreover, during the entire construction process, Intel people could not interfere at all. Unlike the previous technical cooperation with other countries, students here would chase the experts there and study hard.
This construction process will even be kept confidential in some places and isolated Intel's people.
Chen Chen knew that if nothing unexpected happened, Xingchen Semiconductor's factory would be completed in one breath at a very fast speed.
And after completion, it can be mass-produced directly, and then the dog eyes of Intel, Samsung and TSMC are blinded.
But now it is obviously no longer available, and Intel can only use traditional lithography to produce its own processor.
Therefore, the processor order of Xingchen Technology will take as soon as three months to deliver.
However, Chen Chen is not in a hurry. The processor takes time, and other accessories also take time.
So overall, the second generation of Star Mobile should be launched in the fourth quarter and before winter.
In the following time, Chen Chen's life became easier again.
Every day, I accompany my sister, take a few maids, drink tea, watch animations, play games, go to hot springs, go shopping, or go to Xingchen Technology to serve as a person, or visit Xingchen Semiconductor under construction and see the scenery.
I went to chat with Qingkong at night to cultivate my feelings, and also carried out subsequent game market planning with Qingkong.
Every day, Chen Chen would take some time to chat with Lin Qian and Sikong Yue and communicate with each other.
A few days ago, Lin Qian went out for a trip with a few female teachers and posted photos of various scenic spots in the space.
Lin Qian also told Chen Chen that she was approached by a high school student in the scenic area. Lin Qian smiled and said for a long time, while Chen Chen looked helpless.
Chapter completed!