Jump to content

Dokar Moore

Daga Wikipedia, Insakulofidiya ta kyauta.
Dokar Moore
rule of thumb (en) Fassara da principle (en) Fassara
Bayanai
Farawa 1975
Suna saboda Gordon Moore (mul) Fassara

Dokar Moore ita ce lura cewa yawan transistors a cikin hadin gwiwar da ke kewaye (IC) ya ninka kusan kowane shekaru biyu. Dokar Moore kallo ce da kuma tsinkaya game da yanayin tarihi. Maimakon dokar kimiyyar lissafi, dangantaka ce ta kwarewa. Sakamakon gogewa ne, wani nau'in lura da ke ƙididdige nasarorin da aka samu daga ƙwarewar da aka koya a cikin samarwa.

An sanya sunan lura ne bayan Gordon Moore, wanda ya kafa Fairchild Semiconductor da Intel kuma tsohon Babban Jami'in na ƙarshe, wanda a cikin 1965 ya lura cewa yawan abubuwan da aka haɗa da kewayon ya ninka sau biyu a kowace shekara, kuma ya tsara wannan ci gaban zai ci gaba da akalla shekaru goma.[lower-alpha 1] A shekara ta 1975, yana sa ran zuwa shekaru goma masu zuwa, ya sake fasalin hasashen zuwa ninka sau biyu a kowace shekara biyu, yawan ci gaban shekara-shekara (CAGR) na 41%. Shaidar Moore ba ta nuna kai tsaye cewa yanayin tarihi zai ci gaba ba; duk da haka, tsinkayarsa ta kasance tun 1975 kuma tun daga lokacin ya zama sananne a matsayin doka.

An yi amfani da tsinkaya ta Moore a cikin Masana'antar semiconductor don jagorantar shirin dogon lokaci da kuma saita manufofi don bincike da ci gaba (R&D). Ci gaba a cikin kayan lantarki na dijital, kamar raguwa a Farashin da aka daidaita da inganci na microprocessors, karuwar damar ƙwaƙwalwar ajiya (RAM da walƙiya), inganta na'urori masu auna sigina, har ma da yawan da girman pixels a cikin kyamarorin dijital, suna da alaƙa da dokar Moore. Wadannan canje-canje masu gudana a cikin kayan lantarki na dijital sun kasance ƙarfin motsawa na canjin fasaha da zamantakewa, yawan aiki, da ci gaban tattalin arziki.

Masana masana'antu ba su kai ga yarjejeniya daidai lokacin da dokar Moore za ta daina amfani da ita ba. Masu gine-ginen Microprocessor sun ba da rahoton cewa ci gaban semiconductor ya ragu a duk masana'antu tun daga shekara ta 2010, dan kadan a ƙasa da saurin da dokar Moore ta annabta. A watan Satumbar 2022, Shugaba na Nvidia Jensen Huang ya yi la'akari da dokar Moore ta mutu, [2] yayin da Shugaba na Intel na lokacin Pat Gelsinger yana da akasin haka. [3]

Tarihi

[gyara sashe | gyara masomin]

A shekara ta 1959, Douglas Engelbart ya yi nazarin ƙaddamar da girman da'irar da aka haɗa (IC), ya buga sakamakon sa a cikin labarin "Microelectronics, da Art of Similitude". [4] Engelbart ya gabatar da bincikensa a taron 1960 International Solid-State Circuits Conference, inda Moore ya kasance a cikin masu sauraro.

A cikin 1965, Gordon Moore, wanda a lokacin yake aiki a matsayin darektan bincike da ci gaba a Fairchild Semiconductor, an nemi ya ba da gudummawa ga fitowar cika shekaru talatin da biyar na mujallar Electronics tare da tsinkaya game da makomar masana'antar kayan aikin semiconductor a cikin shekaru goma masu zuwa. Amsarsa ita ce taƙaitaccen labarin da ake kira "Cramming more components on integrated da'irori". [1] [5] [lower-alpha 2] A cikin editansa, ya yi hasashen cewa a shekara ta 1975 zai yiwu a ƙunshe da abubuwa 65000 a kan kwata-square-inch (~ 1.1.6) semiconductor. 

Matsakaicin farashin mafi ƙarancin kayan aiki ya karu da kusan kashi biyu a kowace shekara. Tabbas a cikin gajeren lokaci ana iya sa ran wannan adadin zai ci gaba, idan ba zai karu ba. A cikin dogon lokaci, yawan karuwa ya fi tabbas, kodayake babu wani dalili na gaskata cewa ba zai kasance kusan ba har zuwa akalla shekaru 10.[1]

Wannan jadawalin ya nuna yadda dokar Moore ta fi dacewa tun daga shekara ta 1965 da kuma yadda ci gaba kafin 1965 ya kasance mai jinkiri.

Moore ya gabatar da alaƙar layi tsakanin rikitarwa na'urar (mafi girman kewayon a rage farashi) da lokaci.[8][9] A cikin wata hira ta 2015, Moore ya lura game da labarin 1965: "... Na yi kawai karin bayani yana cewa zai ci gaba da ninka sau biyu a kowace shekara na shekaru 10 masu zuwa. " Wani masanin tarihin doka ya ambaci Dokar Stigler ta eponymy, don gabatar da gaskiyar cewa sau biyu na abubuwan da aka sani ga mutane da yawa da ke aiki a fagen. [9] 

A shekara ta 1974, Robert H. Dennard a IBM ya gane fasahar sikelin MOSFET mai sauri kuma ya tsara abin da aka sani da sikelin Dennard, wanda ya bayyana cewa yayin da transistors na MOS suka zama karami, ƙarfin ƙarfin su ya kasance daidai da yadda amfani da wutar lantarki ya kasance daidai tare da yankin.[10] Shaida daga masana'antar semiconductor ta nuna cewa wannan bambancin dangantaka tsakanin ƙarfin wutar lantarki da yawan yankin ya rushe a tsakiyar 2000s.[11]

A taron IEEE International Electron Devices na 1975, Moore ya sake fasalin tsinkayar sa, [12] yana hasashen rikitarwa na semiconductor zai ci gaba da ninka sau biyu a kowace shekara har zuwa kusan 1980, bayan haka zai ragu zuwa ninki biyu kusan kowace shekara biyu. Ya bayyana dalilai da yawa da ke ba da gudummawa ga wannan halayyar: [8][9]

  • Zuwan fasahar ƙarfe-oxide-semiconductor (MOS)
  • Yawan karuwa a cikin girman mutuwa, tare da raguwa a cikin ƙarancin ƙwayoyin, tare da sakamakon cewa masana'antun semiconductor zasu iya aiki tare da manyan yankuna ba tare da rasa raguwar amfanin gona ba.
  • Ƙananan girman
  • Abin da Moore ya kira "ƙwarewar kewayawa da na'urar"

Ba da daɗewa ba bayan 1975, farfesa na Caltech Carver Mead ya shahara da kalmar Moore's law. Dokar Moore ta ƙarshe ta zama an yarda da ita a matsayin burin masana'antar semiconductor, kuma masana'antun semiconductors masu gasa sun ambaci ta yayin da suke ƙoƙari su kara ƙarfin sarrafawa. Moore ya kalli dokar da ta yi amfani da ita a matsayin abin mamaki da kuma kyakkyawan fata: "Dokar Moore ta saba wa Dokar Murphy. Duk abin da ke ci gaba da kyau. " Har ma an ga lura da ita a cikin Annabci mai cika kansa. [13]

Sau da yawa ana ambaton lokacin ninka sau biyu a matsayin watanni 18 saboda tsinkaya daban-daban daga abokin aikin Moore, mai kula da Intel David House . A shekara ta 1975, House ya lura cewa dokar Moore ta sake fasalin ta ninka ƙididdigar transistor sau biyu a kowace shekara 2 ta nuna cewa aikin kwakwalwar kwamfuta zai ninka sau biyu a kowane watanni 18, ba tare da karuwar amfani da wutar lantarki ba.[14] Ta hanyar lissafi, dokar Moore ta yi hasashen cewa ƙididdigar transistor za ta ninka sau biyu a kowace shekara 2 saboda raguwar girman transistor da sauran ci gaba. A sakamakon raguwar girma, Dennard scaling ya yi hasashen cewa amfani da wutar lantarki a kowane yanki zai kasance daidai. Haɗakar da waɗannan tasirin, David House ya ƙaddara cewa aikin kwakwalwar kwamfuta zai ninka kusan sau biyu a kowane watanni 18. Har ila yau, saboda sikelin Dennard, wannan karuwar aikin ba zai kasance tare da karuwar iko ba, watau, ingancin makamashi na kwakwalwan kwamfuta na silicon kusan ninka sau biyu a kowane watanni 18. Dennard scaling ya ƙare a cikin 2000s.[11] Koomey daga baya ya nuna cewa irin wannan ingantaccen inganci ya riga ya wuce kwakwalwan silicon da dokar Moore, don fasahar kamar bututun iska.

Large early portable computer next to a modern smartphone
Kwamfuta mai ɗaukar hoto na Osborne Executive na 1982, tare da 4 MHz 8-bit Zilog Z80 CPU, da kuma Apple iPhone na 2007 tare da 412 MHz 32-bit ARM11 CPU; Zartarwa yana da sau 100 na nauyi, kusan sau 500 na ƙarar, kusan sau 10 na farashin daidaitawa, da kuma 1/100th na sa'a na wayar salula.  

Masu gine-ginen Microprocessor sun ba da rahoton cewa tun daga shekara ta 2010, ci gaban semiconductor ya jinkirta masana'antu a duk faɗin ƙasa da saurin da dokar Moore ta annabta.[1] Brian Krzanich, tsohon Shugaba na Intel, ya ambaci sake dubawa na Moore na 197 a matsayin misali ga raguwar yanzu, wanda ya haifar da ƙalubalen fasaha kuma "wani ɓangare ne na tarihin dokar Moore". [2] [3][4] Yawan ci gaba a cikin girman jiki da aka sani da sikelin Dennard ya ƙare a tsakiyar 2000s. A sakamakon haka, yawancin masana'antar semiconductor sun mayar da hankali ga bukatun manyan aikace-aikacen kwamfuta maimakon sikelin semiconductors.[5][undefined][1] Duk da haka, tun daga shekarar 2019, manyan masana'antun semiconductor TSMC da Samsung Electronics sun yi iƙirarin bin doka ta Moore tare da 10, 7, da 5 nm nodes a cikin samar da taro. [7][8][undefined][undefined][9][undefined][10][11][6]

Dokar Moore ta biyu

[gyara sashe | gyara masomin]

Yayinda farashin wutar lantarki ga mabukaci ya fadi, farashin ga masu samarwa don cika dokar Moore ya bi wani abu daban: R & D, masana'antu, da farashin gwaji sun karu a hankali tare da kowane sabon ƙarni na kwakwalwa. Kudin kayan aikin, musamman matsanancin lithography (EUVL), wanda aka yi amfani da shi don ƙera kwakwalwa sau biyu a kowace shekara 4. Hawan farashin masana'antu muhimmiyar la'akari ce don kiyaye dokar Moore. Wannan ya haifar da tsara Dokar Moore ta biyu, wanda kuma ake kira dokar Rock (mai suna bayan Arthur Rock), wanda shine cewa farashin babban birnin masana'antar kirkirar semiconductor yana ƙaruwa sosai a tsawon lokaci.[15][16]

Dubi kuma

[gyara sashe | gyara masomin]
  • Accelerating change - Karin canjin fasaha ta hanyar tarihi
  • Beyond CMOS - Fasahar da za a iya amfani da ita a nan gaba
  • Dokokin ƙididdigar ƙididdiga - Ƙididdigar abubuwan da ke faruwa a cikin ƙididdigari
  • Ephemeralization - Ka'idar ci gaban fasaha
  • Eroom's law - Bincike game da gano sabbin magunguna
  • Huang's law - Binciken kimiyyar kwamfuta
  • Koomey's law - Hanyar kayan aikin lantarki
  • Limits of computation
  • List of eponymous laws - Adires da maganganu da aka sanya wa sunan mutum
  • List of laws § Technology
  • Microprocessor chronology - Tsarin lokaci na microprocessors
  • Neural scaling law - Dokar kididdiga a cikin ilmantarwa ta inji
  • Power law - Dangantakar aiki tsakanin adadi biyu
  • Wirth's law - Adage akan aikin kwamfuta
  • Rent's rule - Bincike a cikin ƙirar kebul na kwamfuta

Manazarta

[gyara sashe | gyara masomin]
  1. ↑ 1.0 1.1 1.2 Moore, Gordon E. (1965-04-19). "Cramming more components onto integrated circuits" (PDF). intel.com. Electronics Magazine. Archived (PDF) from the original on 2019-03-27. Retrieved April 1, 2020.
  2. ↑ Witkowski, Wallace (2022-09-22). "'Moore's Law's dead,' Nvidia CEO Jensen Huang says in justifying gaming-card price hike" (in Turanci). MarketWatch. Retrieved 2022-09-23.
  3. ↑ Machkovech, Sam (2022-09-27). "Intel: 'Moore's law is not dead' as Arc A770 GPU is priced at $329" (in Turanci). Ars Technica. Retrieved 2022-09-28.
  4. ↑ Empty citation (help)
  5. ↑ Markoff, John (September 27, 2015). "Smaller, Faster, Cheaper, Over: The Future of Computer Chips". The New York Times. Retrieved September 28, 2015.
  6. ↑ Kanellos, Michael (2005-04-11). "Intel offers $10,000 for Moore's Law magazine". ZDNET News.com. Retrieved 2013-06-21.
  7. ↑ "Moore's Law original issue found". BBC News Online. 2005-04-22. Retrieved 2012-08-26.
  8. ↑ 8.0 8.1 Schaller, Bob (September 26, 1996). "The Origin, Nature, and Implications of 'MOORE'S LAW'". Microsoft. Retrieved September 10, 2014.
  9. ↑ 9.0 9.1 9.2 Tuomi, I. (2002). "The Lives and Death of Moore's Law". First Monday. 7 (11). doi:10.5210/fm.v7i11.1000.
  10. ↑ McMenamin, Adrian (April 15, 2013). "The end of Dennard scaling". Retrieved January 23, 2014.
  11. ↑ 11.0 11.1 John L. Hennessy; David A. Patterson (June 4, 2018). "A New Golden Age for Computer Architecture: Domain-Specific Hardware/Software Co-Design, Enhanced Security, Open Instruction Sets, and Agile Chip Development". International Symposium on Computer Architecture – ISCA 2018. In the later 1990s and 2000s, architectural innovation decreased, so performance came primarily from higher clock rates and larger caches. The ending of Dennard Scaling and Moore's Law also slowed this path; single core performance improved only 3% last year!
  12. ↑ Moore, Gordon (1975). "IEEE Technical Digest 1975" (PDF). Intel Corp. Archived (PDF) from the original on 2022-10-09. Retrieved April 7, 2015. ... the rate of increase of complexity can be expected to change slope in the next few years as shown in Figure 5. The new slope might approximate a doubling every two years, rather than every year, by the end of the decade.
  13. ↑ Markoff, John (September 27, 2015). "Smaller, Faster, Cheaper, Over: The Future of Computer Chips". The New York Times. Retrieved September 28, 2015.
  14. ↑ "Moore's Law to roll on for another decade". Retrieved 2011-11-27. Moore also affirmed he never said transistor count would double every 18 months, as is commonly said. Initially, he said transistors on a chip would double every year. He then recalibrated it to every two years in 1975. David House, an Intel executive at the time, noted that the changes would cause computer performance to double every 18 months.
  15. ↑ Markoff, John (August 31, 2009). "After the Transistor, a Leap into the Microcosm". The New York Times. Retrieved 2009-08-31.
  16. ↑ Markoff, John (September 27, 2015). "Smaller, Faster, Cheaper, Over: The Future of Computer Chips". The New York Times. Retrieved September 28, 2015.


Cite error: <ref> tags exist for a group named "lower-alpha", but no corresponding <references group="lower-alpha"/> tag was found

Daga "https://ha.wikipedia.org/w/index.php?title=Dokar_Moore&oldid=760048"