中国能否发展出世界级的半导体产业?
2021-08-01 地平线123 23481
正文翻译


中国能否发展出世界级的半导体产业?

评论翻译
Donald Canton, Positive healthy person in Boston, Massachusetts, USA
upxed Tue
Yes, China absolutely can develop a world-class semiconductor industry[1] because the semiconductor chips, as the source of the rapid development of today's electronic information technology industry, have been widely penetrated and integrated into every field of the national economic and social development, and are an important support for the development of the digital economy and information tech in China. The 28 nanometer process will be a new starting point for 100% domestic chips and the domestic 14 nanometer chips are expected to be mass-produced this year 2021 and next year 2022, respectively according to Dr. Xiaojun Wen, Director, Associate Professor at Institute of ICT Industry in CCID Group.

中国肯定可以发展出世界级的半导体产业,因为现在电子信息技术行业的发展都要仰仗于半导体芯片的发展,半导体芯片已经融入了国民经济和社会发展的方方面面,并对中国数字经济的发展和信息技术的发展起着非常重要的辅助作用。根据中国信息技术研究所主任,副教授文晓军博士的说法,中国28纳米芯片已经全部实现了国产化,14纳米芯片预计在今年2021年到明年2022年间实现大规模的量产。

Despite the technical difficulties the Chinese scientists and engineers have faced, the establishment of a world-class semiconductor industry in China has started to become a reality.

尽管中国科学家和工程师面临着技术上的困难,但在中国建立世界级的半导体产业已经开始成为现实。

The development of 14nm semiconductor industrial processing system in China has overcome many technical problems such as the following: key equipment such as etching machines and thin film deposition have been realized from scratch and applied in large-scale production lines; breakthroughs have been made in the research and development of 14nm process; the results of subsequent packaging integration technology have been fully realized in mass production. Hundreds of key materials, such as polishing agents and sputtering targets, have passed the assessment of large production lines and entered mass sales. These achievements basically cover the entire industrial chain system of the integrated circuit manufacturing system in China[2], and reverse the passive situation of the introduction need from the foreign companies of a complete set of process technologies.

中国的14纳米半导体加工研发体系已经克服了许多的技术难题,例如:关键性设备如蚀刻机和薄膜沉积设备从无到有实现了突破,并在生产线生产中得到大规模的应用;14nm工艺研发取得突破性进展; 后续封装集成技术的成果得到了大规模的应用。数百种关键性材料,如抛光剂溅射靶等,都已经通过大规模生产线的审核要求,可以进行大规模的生产销售。这样的成就在中国集成电路生产体系的各产业链中遍地开花,扭转了要从外国企业那引进成套工艺技术的被动局面。

Behind every example that China can solve is the point where American technology loses its political role. Too many American advantages will collapse. The US blockade of the semiconductor chip industry chain will collapse no later than 2024. Under the positive cycle, it is estimated that by 2024, the localization of most semiconductor chips from raw materials, equipment, design, production and packaging can be solved in China.

中国每解决一个技术难关,美国可以用来打政治牌的技术就少了一个。未来美国的技术优势会纷纷坍塌。美国对半导体芯片产业链的封锁最迟会在2024年瓦解。在这样的良性循环下,预计到2024年,大部分半导体芯片的原材料、设备、设计、生产和包装都可以在中国实现国产化。

The domestic 28nm production line is likely to achieve 100% localization in 2021, thus changing the global semiconductor industry pattern[3]. The rapid development of 14nm and even 28nm process localization means that the Chinese adopt a return strategy and use mature technology to meet general chip needs first. The Chinese do not blindly pursue higher level manufacturing processes, but pay more attention to design and packaging optimization, in exchange for semiconductor applications and the whole industry chain independence capacity.

国内28纳米生产线有望在2021年实现100%国产化,从而改变全球半导体产业格局。14纳米乃至28纳米工艺国产化的快速发展,意味着中国将采取原来的战略,首先利用成熟技术满足通用芯片需求。中国人不一味追求更高水平的制造工艺,而是更注重设计和封装优化,以换取半导体应用和全产业链的自主能力。

In addition, after the mass production of 14nm chips, subsequent expansion of production capacity requires financial support, including photolithography machines, cleaning equipment, polishing equipment and other required equipment that are not only expensive, but also consume a lot of water and electricity during use. It is necessary to do a good job of integration at the supplier level of raw materials and components, and do a good job of customer introduction in advance to ensure that production capacity is fully utilized.

而且,在14纳米芯片量产之后,后续的产能扩张需要有资金的支持,例如光刻机、清洗设备、抛光设备等,这些设备不仅价格昂贵,而且在使用过程中还要耗费大量的水电资源。因此有必要做好原材料和零部件供应商的整合工作,为了确保产能得到充分利用需要提前做好客户的拓展工作。

The production line of the 14-12nm generation is very critical in current semiconductors. The 14nm process and above can meet the needs of 70% of the current semiconductor manufacturing processes, and the mid-range 5G chips are positioned using the 12nm process. In addition, 14nm can basically meet the needs of the manufacturing process required by domestic desktop CPUs in China.

14-12纳米生产线是目前半导体领域的关键。14纳米以上工艺可以满足目前70%的半导体生产工艺需求,普通的5G芯片可以使用12纳米工艺制造。而且14纳米生产工艺基本上已经可以满足中国台式机CPU的要求了。
原创翻译:龙腾网 http://www.ltaaa.cn 转载请注明出处


According to statistics, in the first half of 2019, the entire semiconductor sales market was about US$200 billion, of which 65% of chips used 14nm process technology, only about 10% of chips used 7nm, and about 25% used 10nm and 12nm. It can be seen that 14nm has become the most widely used process technology with the most market value. It has great development potential in AI chips, high-end processors, and automobiles.

数据统计显示,2019年上半年,整个半导体市场销售额为2000亿美元,其中65%的芯片是14纳米工艺技术,只有10%的芯片使用的是7纳米技术,还有25%的芯片使用的是10纳米和12纳米的技术。由此可见,14纳米是使用最广泛,市场的价值也最高,在人工智能芯片、高端处理器和汽车领域有很大的发展潜力。

The 14nm technology is booming in China now, and it has delivered a good report card, but it cannot be accomplished overnight if they want to control and achieve catch-up perfectly. The industry's leading foreign manufacturers have many years of production experience on 14nm, and the Chinese enterprises have no advantage in cost competition with those manufacturers. Therefore, there is a generation gap between the Chinese and the world's first-rate foundry companies in technology catching up. If they want to catch up as latecomers, the Chinese need to invest more manpower, financial resources and time costs.

现在中国的14纳米技术蓬勃发展,已经取得了不错的成绩,但他们不可能一夜之间在该领域实现完全的追赶。该行业处于领先地位的外国制造商在14纳米生产上已经拥有了多年的经验,中国公司与这些企业相比并没有成本竞争的优势。因此,在技术追赶方面,中国与世界一流的芯片代工企业还存在着巨大的差距。作为后来者,中国如果想要赶上来的话,就需要投入更多的人力,财力和时间成本。

The integrated circuit industry is one of the global frontier industries with ultra-intensive capital, frequent replacements, and extremely sophisticated talents. It has high technical difficulties, high cost of trial and error, and difficulty in troubleshooting. It requires China the determination to develop independently, the perseverance of hard work, the confidence to catch up and surpass success, and it also requires strong and continuous funding, iterative technology, qualified talents, and a market that integrates production and use to gather capital, technology, talent, and the market. The four elements work together to achieve successful catch-up and surpass development.

集成电路产业是全球前沿产业之一,属于资本高度密集型行业,技术更新迭代频繁,极度依赖高精尖人才。这个行业技术难度大,试错成本高,故障排除困难。需要中国有独立研发的决心,艰苦奋斗的精神,成功赶超的信心,同时还要求有源源不断的资金投入,不断的技术进行迭代升级,大量的优秀人才,还要有一个可以获得资金,技术和人才的市场进行生产和消费。

The new high-resolution lithography machine, which the Chinese scientists began building in 2012, can etch circuitry patterns less than 22 nanometers using ultraviolet light. Combined with other techniques, it can be used in the future to create chips of around 10 nanometers. China now has 47 domestic and four international patents on the new technology. It is actually not that the Chinese are unable to make their own lithography machine but that it is very difficult for the Chinese scientists to work around those existing patents to create something brand new from scratch.

中国科学家于2012年开始研制出新型高分辨率光刻机,可以利用紫外光刻蚀小于22纳米的电路图案。结合其他技术,它在未来可以用来制造10纳米左右的芯片。目前,中国在这项新技术上拥有47项国内专利和4项国际专利。事实上,并不是中国人不能制造出自己的光刻机,而是中国科学家很难绕过现有的专利,从零开始创造出全新的东西。

Felix Su
July 14
China will catch up sooner than later at this point in time. It was stupid and idiotic for Trump to ban sales to China.
Because China will become self-sufficient and then what? US sales to China are gigantic. What happens to those US companies that lose their market in China?
28nm this year, 14nm next year.
China is probably thanking the US for short term pain and long term gain.

在这种情况下,中国迟早都会赶上来。特朗普禁止对华销售是很愚蠢的事情
中国一旦可以自给自足之后,那么美国要怎么办。美国芯片对华销售额是极其庞大的。失去了中国市场的美国公司要怎么办?今年是28纳米,明年就是14纳米。
从短期来看中国会承受痛苦,但长期来看,这对中国是有好处的,中国应该感谢美国

Henry Young
July 1
Although I am nota technical man, from what I read in the press, I have every confidence in China, my motherland, to catch up and become self-sufficient.

虽然我不是技术人员,但从新闻报道来看,我对中国在半导体领域的追赶和自给自足是有信心的。

Tianren Tan
June 29
Once China dominates the low end of chips, it will move up the value chain to 7nm, then 5nm …. We seen it many times already in other industries.

一旦中国主导了低端芯片,那么他们就会向7纳米,5纳米这些价值链上游攀爬,在其他行业我们已经见识过很多次这种情况了

Gareth Cheong
July 14
When you placed barriers in front of people, they will learn to overcome it. When you raised them, they will jump higher to clear it. Maybe not today but they will surely improve. Such is the nature of humankind. It is not just confined to the Chinese but they are the subject of technology barricade now and so they will no doubt improve on this IC thingy. We can expect that the US will not want to lose its place and will so do with what it can to stay ahead. If the US is light years more advanced then it should not need to impose extra-competitive measures to trip up China. So it is just true that China is fast closing the gap.

当你给别人设置障碍的时候,别人就会学着如何去克服这个障碍。当你把门槛升高以后,他们也会相应跳的更高。或许现在还做不到,但他们将来一定会有所改善。因为这就是人的本性。中国现在受困于技术上的障碍,所以他们肯定会想办法提高集成电路技术。可以预见的是,美国不希望失去自己的地位,无论如何都要保持领先。如果美国的技术领先中国几光年,那么它就不需要给中国使绊子了。中国现在的确是在快速的缩小和美国的差距

Aries Budhyanto
July 3
Actually, 15 nano technology is good enough, only slightly bigger. Smaller doesn’t mean better performance. I just read an article that 90% gadgets using 16 nano technology only 10% gadgets go to 6 nano.

实际上,15纳米已经足够好了,只是稍微有点大而已,芯片更小并不意味着性能就更好,我在一篇文章中看到过,90%的电子产品使用的是16纳米技术,只有10%使用的是6纳米的

Gully Foyled
July 15
Remember when blocking Huawei from using Android was going to put them out of business?
Less than two years later they have their own OS which is being deployed to hundreds of millions of devices and is starting to be adopted by other Chinese IT companies.
Independent tests show it to be faster and more energy-efficient than Android.

还记得禁止华为使用安卓操作系统,意图使其破产的事情吗?
两年不到他们就有了自己的操作系统,这些操作系统被安装上了数亿部华为的手机上,中国其他的信息技术企业也开始采用这套操作系统。
独立测试表明,它比安卓更快、更节能。

John Bould
June 29
China’s push to become more or less self-sufficient in semiconductor technology was happening even before the decoupling started in earnest.

中国在推动半导体技术领域独立自主方面取得了或多或少的成就,这甚至在脱钩开始之前就已经开始在做了

Henry Hui
July 17
The presence of demand due to a hugh local market will always allow these development efforts to succeed.
The void created by overseas supply ban is actually an accelerator for the development of local production capability and technological advancement.

由于庞大的本土市场拥有巨大的需求,这些研发努力总会成功的。
海外禁令创造的市场空白实际上成为了中国国内生产能力和技术发展的加速器
原创翻译:龙腾网 http://www.ltaaa.cn 转载请注明出处


Erik Lister
July 11
Lol. TSMC is light years ahead of anything China backed institutions will ever be able to produce. Samsung as well. Luckily, ASML never produced their machinery in China, so they weren’t able to steal that tech through “technology sharing” agreements. Better not anger Taiwan too much or China will be SoL.

哈哈。台积电生产的产品要比中国大陆支持的机构生产的产品领先几个光年。三星同样也领先中国。幸运的是,阿斯麦从来没有在中国生产自己的设备,所以中国无法通过技术共享协议来XX技术。最好不要激怒台湾地区,否则中国大陆会,,,

Siew Wong
July 12
If you are from the US, then you should be ashamed for your country to coerce the Dutch ASML not to sell their products to China. The Netherlands clearly does not belong to the US. The US is nothing but the world's biggest bully and most evil nation.

如果你是美国人,那么你应该为自己国家逼迫荷兰阿斯麦不卖自己的产品给中国而羞愧。荷兰显然不属于美国,美国就是世界上最大的恶霸,最邪恶的国家。

Jason
July 15
You are right. So US has nothing to worry about. Keep thinking that way.

说的没错,美国没有什么好担心的,继续这样想就好

Tony Hurtuno
July 16
C’mon be a sport. Your comments are based on what you heard and read in the news media or what you fantasize. Chinese can do a lot more than you think. Remember they are CCP and not American bureaucratic system. Once they set their plan, they follow it through. Ths US has already been left behind you care to believe it or not.. I and many of us own 5G “capable" phone made in China or Korea and yet we have to dial it down to the old 4G LTE. Why? Because there is not enough network to support it causing a slow down in the reception.

你的评论是受到了新闻媒体的影响或基于你自己本人的幻想。中国人能做到的事情比你想象的多得多。要知道中国的政府和美国的官僚体系是不一样的。一旦他们制定了计划,他们就会坚持到底。不管你信不信,美国已经被甩在了后面。我和很多人都有中国或韩国制造的5G手机,但我们不得不使用旧的4G LTE网络,这是为什么?因为这里没有足够的网络支持,导致接收速度变慢。

Jim Cy Sia
July 11
Yes china alrwady prodyced all 28nm.chips required domestically and some 14nm. By 2022 7nm will also be produced. China has 270000 companies s producing all types of chips

中国国内已经能够生产28纳米芯片,部分的14纳米芯片也能生产出来。2022年,中国将能生产出7纳米芯片。中国生产各类芯片的企业高达27万家。

Matthew Lam
July 14
I truely believe China will become a leader in the semiconductor industry in the very near future….. China has proved itself in its capability, determination and preseverance in making great achievements . Examples include 5G, BeiDou Navigation Satellite System, Supercomputer, Space Station … and the list goes on and on.

我相信中国在不久的将来会成为半导体领域的领先者。中国在不断取得伟大成就的过程中证明了自己的能力和决心。例如5G技术,北斗导航,超级计算机,空间站等。

Richard Kwok, lives in Hong Kong
Answered April 29, 2018
In certain part of the value chain we might see some world class players, or even champions. The money and prospect are likely attract global talents, as semi engineers are generally under paid vs their software cousins, so apart from the global champions, China is probably the only new player going to throw tons of money around. You will probably see overseas Chinese engineers including Taiwanese who will be the first ones to move.

在价值链的某些部分,我们可能会看到一些世界级的企业,甚至是冠军企业。薪水和前景可能会对全球的人才产生吸引力,因为半导体工程师的薪水没有软件工程师那么高,所以除了全球性的冠军企业以外,中国或许是唯一愿意投入大量资金的国家了。你可能会看到第一批前往中国效力的工程师都是海外华人,其中包括台湾地区。

No, China will not be able to create a complete local industry, the semi industry was developed over many decades of hard work and trillions of investment. The IP barrier is high, you can’t rework a lot of things. So I suspect that ZTE incident is a big wake up call for China, the proud internet and telecom champions are so vulnerable that they can be choked by US any second.

但是中国无法迟早一个完整的本地产业,半导体行业的发展经历了数十年的时间,投入了数万亿美元的投资。其中知识产权的壁垒非常的高,很多东西是无法重做的。作为中国自豪的互联网和电信冠军企业,中兴竟然会如此的脆弱,随时都有可能被美国扼杀,我认为中兴事件给中国敲响了警钟。

Semiconductor industry has a huge sub category, and itself is a globalisation marvel.
On top of value chain we have the American Lam Research, Applied Materials and Dutch ASML who provides the state of the art equipments. China has…. not much here.
Wafers we have the Japanese Shinetsu and Sumco, and Taiwanese Global Wafers. China has… not much here.

半导体行业由很多部分组成,它本身就是全球化的奇迹
在价值链的顶端,是美国的泛林集团,应材(全球最大的半导体设备和服务供应商)和提供最先进设备的荷兰阿斯麦,中国在半导体价值链顶端的企业并不多
在晶圆片方面,代表企业有日本的信越化学工业和株式会社,台湾地区的环球晶圆,,中国的代表企业也没有多少
原创翻译:龙腾网 http://www.ltaaa.cn 转载请注明出处


Then from wafers into chipsets we have Intel and Samsung. But biggest fab or chip manufacturer is TSMC, the Taiwanese industrial crown jewel. China has SMIC who is about 2–3 gens behind. One of the choke point is ASML whose major shareholders are: Samsung, Intel and TSMC. So SMIC will never get the latest equipment even it has the knowhow and the engineers.

从晶圆到芯片,代表企业有英特尔和三星。但最大的晶圆厂或芯片制造商是台积电,这是台湾地区工业王冠上的宝石。中国的中芯国际落后两到三代。其中一个瓶颈就是阿斯麦,它的主要股东是三星,英特尔,和台积电。所以即便中芯国际掌握了技术并且拥有最好的工程师,但它还是无法获得最新的设备。

So as you can see from above the semi industry is dominated by Americans and American military allies. After decades of consolidation the entry barrier became extremely high for new comers, most sub-sectors are dominated by very few companies who enjoy monopoly or oligopolies in their “little” own multi-billion dollar markets. So understandably China the poor kid who’s late to the party is trying to move up the value chain, so the people can have a higher income (this is pretty much what Made in China 2025 is about), but it’s impossible for China to create a complete new ecosystem.

半导体是由美国及美国军事盟友主导的一个行业。经过几十年的整合,对于新玩家来说,行业的进入壁垒已经变得非常的高了,大部分的子行业由极少数的几家公司主导,这些公司在这个价值数十亿美元的市场上占据了垄断或寡头的地位。中国作为后来者试图向价值链上游移动,这是可以理解的,因为这样人们才能获得更高的工资(这也是中国制造2025的一大目的),但中国要想创建一个全新的生态系统是不可能的事情。

However recent event shows that globalisation can be switched off by few people that causes major disturbance in the semiconductors supply chain, and such disturbance works both ways. So China is likely to invest heavily in few critical sectors of the value chain to create the semiconductors version of Mutral Assured Destruction.

然而,最近一些事情表明全球化有可能被少数几个人切断,这会对半导体供应链造成巨大的干扰,而这种干扰是双向的。因此中国可能会投资少数关键性的半导体领域,打造半导体版的相互毁灭。

So this is what’s happening: Buyers of China tech products: shit, is the Americans gonna cut China off so I’m not gonna receive my order in time? Americans: shit, should I make this expansion plan now my gov might block my sell? Chinese: shit, Americans can cut us off anytime, better invest hard to replace at least some of them. Koreans/Japanese/Taiwanese: shit, Americans can chock us anytime if we sell to China, what should we do? Electric Engineer students: yeah, maybe I don’t need to switch to software engineering anymore might look for a job in China.

所以现在的情况是这样的
中国买家:糟了,难道美国人要切断与中国的联系,这样的话我不就无法及时收到货了吗?
美国人:糟了,现在美国政府要禁止我卖货给中国,那我还应该实施扩充产能的计划吗?
中国人:糟了美国人可以随时不卖货给我们,最好是大力投资,至少要进行部分替代。
韩国,日本和台湾地区:如果我们继续卖货给中国,美国人会随时掐死我们,我们要怎么办?
电子工程师学生:或许我不需要改学软件工程了,在中国或许可以找到一份好工作。

Richard Kwok
May 1, 2018
The thing with semi is it’s a commodity, you have to be good and cheap, I’m pretty sure China can make Intel/Qualcomm chips with some effort just like the super computer after the Intel ban, but they will not be economical. I think as silicon is reaching its physical limit, and quantum computing on the horizon, there might be some kind of catch up opportunity, but in the near future it’s just not possible to replace the current eco system.

半导体是一件商品,必须做到物美价廉,我相信经过一番努力,中国可以做出向英特尔和高通那样的芯片,正如在被英特尔禁止后,中国做出了超级计算机一样,但这样的产品还是不够经济。我认为硅已经快要达到自己的物理极限了,量子计算机即将到来,这或许会是中国赶超的一个契机,但在可预见的将来,要想取代当前的生态系统还是不可能的事情。

Zee Li
I believe YMTC can catch up with Samsung before 2030

我相信2030年之前长江存储科技可以赶上三星

很赞 5
收藏