|
不是由于有意压抑,只是由于鉴别不清,也会妨碍新生事物的成长。
为我国2070年开始的诺贝尔科学奖“井喷”清除障碍、铺平道路!
客观规律是客观的;独立于人而客观存在。
[感慨] 芯片历史:真理与人数无关;与同行评议无关
一、陶卓彬老师观点摘录
陶卓彬老师2020-06-07在《中国芯能否登顶?从芯片产业发展史找答案》里写到:
第三场战役是革新晶体管架构。当制程继续缩小到45纳米以下时,原来的晶体管架构存在的问题被放大了。…… 但是MG出现了两种技术方案,分别叫Gate-first与Gate-last,IBM、英飞凌、NEC、格罗方德、 三星、意法半导体以及东芝等采用Gate-first方案,而英特尔与台积电采用Gate-last。结果证明真理掌握在少数派手里。因为正确的判断和严格的工艺,台积电良率大幅提升。
制程继续向22纳米以下进发时,又有事了,还是老问题:尺寸缩小后漏电变严重了。为了解决这一问题分为两派,一派是采用FinFET(Fin Field-Effect Transistor,鳍式场效晶体管),原理简单说是将S极和D极由水平改为垂直,沟道被栅极三面环绕,增厚绝缘层增加接触面积,减少漏电现象的发生。阵营成员包括台积电、英特尔、三星。而且FinFET的发明人胡正明在2001年被台积电聘为第一任CTO。另一派是采用SOI(Silicon-On-Insulator,绝缘层上硅),该技术是在顶层硅和衬底之间增加一层氧化绝缘体,减少向底层的漏电,成员是继承IBM遗志的格罗方德。结果证明真理不是掌握在少数派手里,是掌握在台积电手里。FinFET胜,2015年台积电顺利实现16nm制程。
求陶老师上述观点的原始出处。感谢!
二、感慨
“结果证明真理掌握在少数派手里。”和“结果证明真理不是掌握在少数派手里”,证明:
(1)客观规律是客观的;独立于人而客观存在。
(2)同行评议越严,人类科技发展越慢。
遇上胡克的“同行评议”,就连著名的牛顿都输了!
“同行评议”,天下无敌!!
Peer review: Troubled from the start?
Peer review: Destroys true sciences from the start?
难怪“中国工程院院长、上海市市长、中国工程院院士、美国国家工程院外籍院士、俄罗斯工程科学院外籍院士、瑞典皇家工程院外籍院士”徐匡迪院士,大声喊出“颠覆性技术是被专家‘投’没的。”
三、建议放缓集成电路与通讯、互联网的发展
Moore’s law, prediction made by American engineer Gordon Moore in 1965 that the number of transistors per silicon chip doubles every year. Moore's law. Gordon E. Moore observed that the number of transistors on a computer chip was doubling about every 18–24 months.
https://www.britannica.com/technology/Moores-law
Moore's law | computer science | Britannica
戈登·摩尔(Gordon Moore)在1965年4月的《电子》杂志(Electronics)提出的“摩尔定律”,其核心内容为:集成电路上可以容纳的晶体管数目在大约每经过18-24个月便会增加一倍。
Moore's Law - The number of transistors and resistors on a chip doubles every 24 months
https://www.azquotes.com/quote/570104
https://www.azquotes.com/picture-quotes/quote-moore-s-law-the-number-of-transistors-and-resistors-on-a-chip-doubles-every-24-months-gordon-moore-57-1-0104.jpg
https://www.zdnet.com/article/ai-is-changing-the-entire-nature-of-compute/
后期调整过。
https://www.zdnet.com/article/ai-is-changing-the-entire-nature-of-compute/
后期调整过。
感谢上面所有图片的原作者和有关人员!
Zenas advice(忠告):
现代集成电路、数字通讯、互联网等的性能,对于民用已经足够了。没有必要再去花大力气发展了。集成电路芯片性能提高一倍的时间,最好超过 370个月!或许490个月更好!!
这就是说:支配集成电路发展50多年的摩尔定律(Moore's law),应该被全人类理性地主动地停用了。
呼吁全人类,从今天开始,主动改用 Zenas advice 吧!
参考资料:
[1] Moore's law | computer science | Britannica
https://www.britannica.com/technology/Moores-law
Moore’s law, prediction made by American engineer Gordon Moore in 1965 that the number of transistors per silicon chip doubles every year. For a special issue of the journal Electronics, Moore was asked to predict developments over the next decade. Observing that the total number of components in these circuits had roughly doubled each year, he blithely extrapolated this annual doubling to the next decade, estimating that microcircuits of 1975 would contain an astounding 65,000 components per chip.
[2] Gordon Moore | Biography & Facts | Britannica
https://www.britannica.com/biography/Gordon-Moore
Gordon Moore, in full Gordon E. Moore, (born January 3, 1929, San Francisco, California, U.S.), American engineer and cofounder, with Robert Noyce, of Intel Corporation. Moore studied chemistry at the University of California, Berkeley (B.S., 1950), and in 1954 he received a Ph.D. in chemistry and physics from the California Institute of Technology (Caltech), Pasadena.
[3] 英特尔公司,Gordon Moore,Raising the bar for silicon technology and innovation.
https://www.intel.cn/content/www/cn/zh/history/museum-gordon-moore-law.html
[4] 网易,2016-03-29,预言不准了 摩尔定律瓶颈突破还有戏吗
https://www.163.com/mobile/article/BJAA35GM0011665S.html
[5] Computer History Museum, 1965: "MOORE'S LAW" PREDICTS THE FUTURE OF INTEGRATED CIRCUITS
https://www.computerhistory.org/siliconengine/moores-law-predicts-the-future-of-integrated-circuits/
"The Future of Integrated Electronics" attempted to predict "the development of integrated electronics for perhaps the next ten years." Extrapolating the trend to 1975 he projected that the number of components per chip would reach 65,000; a doubling every 12 months. Edited for publication as a magazine article, "Cramming more components onto integrated circuits" was published in Electronics on April 19, 1965.
[6] Gordon E. Moore. Lithography and the future of Moore's Law [J]. IEEE Solid-State Circuits Society Newsletter, 2006, 11(3): 37-42.
https://ieeexplore.ieee.org/document/4785861
[7] Moore, Gordon. "The Future of Integrated Electronics." Fairchild Semiconductor internal publication (1964).
https://www.chiphistory.org/20-moore-s-law-original-draft-1965
[8] ZDNET, 2019-06-30, AI is changing the entire nature of compute
https://www.zdnet.com/article/ai-is-changing-the-entire-nature-of-compute/
[9] 袁岚峰,2020-06-11,了解历史,才知道中国芯片如何胜出
https://weibo.com/ttarticle/p/show?id=2309404514772266647591
https://zhuanlan.zhihu.com/p/147525190
[10] 跟陶叔学编程,2020-06-07,中国芯能否登顶?从芯片产业发展史找答案
https://weibo.com/ttarticle/p/show?id=2309404513090149417162
https://zhuanlan.zhihu.com/p/146503741
[11] Michael Park, Erin Leahey, Russell Funk, arXiv.org > cs > arXiv:2106.11184, 2021-06-26, Dynamics of Disruption in Science and Technology
https://arxiv.org/abs/2106.11184
[12] 中华全国工商业联合会,2019-04-30,徐匡迪院士:**颠覆性技术是被专家“投”没的
http://www.acfic.org.cn/fgdt1/zjgd/201904/t20190430_120889.html
相关链接:
[1] 2021-08-10,[求证] ASML 腾飞的技术原因是什么?【immersion system】
http://blog.sciencenet.cn/blog-107667-1299147.html
[2] 2019-07-01,[请教] 量子集成电路、量子芯片 Quantum Chip 今后30年内的实用前景?
http://blog.sciencenet.cn/blog-107667-1187623.html
[3] 2019-07-14,有关集成电路 Integrated Circuit 的网页
http://blog.sciencenet.cn/blog-107667-1189467.html
[4] 2019-07-17,[求助] 集成电路 Integrated Circuit 当前最新技术资料?
http://blog.sciencenet.cn/blog-107667-1189948.html
[5] 2021-08-10,[求证] ASML 腾飞的技术原因是什么?【immersion system】
http://blog.sciencenet.cn/blog-107667-1299147.html
[6] 2020-08-01,破除论文“SCI至上”:[历史] 艾萨克·牛顿斗不过“同行评议”!
http://blog.sciencenet.cn/blog-107667-1244556.html
[7] 2016-03-31,评审越严,人类科技发展越慢?
http://blog.sciencenet.cn/blog-107667-966229.html
[8] 2021-07-24,Zenas 公理:2021年《Dynamics of Disruption in Science and Technology》里的图示
http://blog.sciencenet.cn/blog-107667-1296808.html
感谢您的指教!
感谢您指正以上任何错误!
感谢您提供更多的相关资料!
Archiver|手机版|科学网 ( 京ICP备07017567号-12 )
GMT+8, 2024-11-7 22:25
Powered by ScienceNet.cn
Copyright © 2007- 中国科学报社