Zuwan wannan guntu ya canza hanyar ci gaban guntu!
A ƙarshen shekarun 1970, na'urori masu sarrafa bit 8 har yanzu su ne fasahar da ta fi ci gaba a wancan lokacin, kuma hanyoyin CMOS sun kasance a cikin matsala a fannin semiconductor. Injiniyoyi a AT&T Bell Labs sun ɗauki mataki mai ƙarfi a nan gaba, inda suka haɗa hanyoyin kera CMOS na zamani mai girman micron 3.5 tare da sabbin tsarin sarrafa bit 32 don ƙoƙarin yin fice a cikin masu fafatawa a aikin guntu, inda suka zarce IBM da Intel.
Duk da cewa ƙirƙirar su, Bellmac-32 microprocessor, ta kasa cimma nasarar kasuwanci na samfuran da suka gabata kamar Intel 4004 (wanda aka fitar a 1971), tasirinsa ya yi yawa. A yau, kwakwalwan kwamfuta a kusan dukkan wayoyin komai da ruwanka, kwamfutocin tafi-da-gidanka, da kwamfutocin hannu sun dogara ne akan ka'idojin semiconductor na ƙarfe-oxide (CMOS) waɗanda Bellmac-32 ya fara.
Shekarun 1980 suna gabatowa, kuma AT&T tana ƙoƙarin canza kanta. Shekaru da dama, kamfanin sadarwa mai suna "Mother Bell" ya mamaye harkokin sadarwa na murya a Amurka, kuma reshensa na Western Electric ya samar da kusan dukkan wayoyin tarho na gama gari a gidaje da ofisoshin Amurka. Gwamnatin tarayya ta Amurka ta yi kira da a wargaza kasuwancin AT&T bisa dalilan kin amincewa da amana, amma AT&T ta ga dama ta shiga fagen kwamfuta.
Ganin cewa kamfanonin kwamfuta sun riga sun kafu sosai a kasuwa, AT&T ta ga ya yi wuya ta cimma hakan; dabarar da ta yi ita ce ta yi nasara, kuma Bellmac-32 ita ce ginshiƙinta.
An karrama iyalan na'urorin Bellmac-32 da lambar yabo ta IEEE Milestone. Za a gudanar da bukukuwan buɗewa a wannan shekarar a harabar Nokia Bell Labs da ke Murray Hill, New Jersey, da kuma a Gidan Tarihi na Tarihin Kwamfuta da ke Mountain View, California.
ƘUNSHIN MUSAMMAN
Maimakon bin ƙa'idar masana'antu ta kwakwalwan kwamfuta masu amfani da bit 8, shugabannin AT&T sun ƙalubalanci injiniyoyin Bell Labs da su ƙirƙiri wani samfuri mai juyi: na'urar sarrafa bayanai ta kasuwanci ta farko da za ta iya canja wurin bayanai bits 32 a cikin zagayowar agogo ɗaya. Wannan ba wai kawai ya buƙaci sabon guntu ba, har ma da sabon tsarin gine-gine—wanda zai iya sarrafa sauya sadarwa da kuma zama ginshiƙin tsarin kwamfuta na gaba.
"Ba wai kawai muna gina guntu mafi sauri ba ne," in ji Michael Condry, wanda ke jagorantar ƙungiyar gine-gine a cibiyar Bell Labs da ke Holmdel, New Jersey. "Muna ƙoƙarin tsara guntu wanda zai iya tallafawa murya da lissafi."
A lokacin, ana ɗaukar fasahar CMOS a matsayin madadin da ke da kyau amma mai haɗari ga ƙirar NMOS da PMOS. Kwamfutocin NMOS sun dogara ne kacokan akan transistors na nau'in N, waɗanda suke da sauri amma suna da buƙatar wutar lantarki, yayin da kwamfutocin PMOS sun dogara ne akan motsin ramuka masu caji mai kyau, wanda yake da jinkiri sosai. CMOS ya yi amfani da ƙirar haɗin gwiwa wanda ke ƙara gudu yayin da yake adana wutar lantarki. Fa'idodin CMOS sun kasance masu ban sha'awa har masana'antar ta fahimci cewa ko da yana buƙatar sau biyu na transistors (NMOS da PMOS ga kowace ƙofa), ya cancanci hakan.
Tare da saurin haɓaka fasahar semiconductor da Moore's Law ta bayyana, farashin ninka yawan transistor ya zama abin sarrafawa kuma daga ƙarshe ba shi da yawa. Duk da haka, lokacin da Bell Labs ta fara wannan caca mai haɗari, fasahar kera CMOS mai girma ba ta da tabbas kuma farashin ya yi yawa.
Wannan bai tsorata Bell Labs ba. Kamfanin ya yi amfani da ƙwarewar harabarsa a Holmdel, Murray Hill, da Naperville, Illinois, kuma ya tattara "tawagar mafarki" ta injiniyoyin semiconductor. Tawagar ta haɗa da Condrey, Steve Conn, wani tauraro mai tasowa a fannin ƙirar guntu, Victor Huang, wani mai tsara ƙananan na'urori masu sarrafawa, da kuma ma'aikata da dama daga AT&T Bell Labs. Sun fara ƙwarewa a sabon tsarin CMOS a shekarar 1978 kuma suka gina ƙaramin na'ura mai sarrafawa mai girman bit 32 daga farko.
Fara da tsarin zane
Condrey tsohon IEEE Fellow ne kuma daga baya ya yi aiki a matsayin Babban Jami'in Fasaha na Intel. Ƙungiyar gine-gine da ya jagoranta ta himmatu wajen gina tsarin da ke tallafawa tsarin aiki na Unix da harshen C. A lokacin, harsunan Unix da C har yanzu suna cikin ƙuruciyarsu, amma an ƙaddara su mamaye su. Domin su karya ƙa'idar ƙwaƙwalwar kilobytes (KB) mai matuƙar muhimmanci a lokacin, sun gabatar da wani tsari mai rikitarwa wanda ke buƙatar ƙananan matakan aiwatarwa kuma zai iya kammala ayyuka a cikin zagayowar agogo ɗaya.
Injiniyoyin sun kuma ƙera kwakwalwan kwamfuta waɗanda ke tallafawa bas ɗin VersaModule Eurocard (VME) mai layi ɗaya, wanda ke ba da damar rarraba kwamfuta kuma yana ba da damar sarrafa bayanai a layi ɗaya. Kwamfutocin da suka dace da VME kuma suna ba da damar amfani da su don sarrafa lokaci-lokaci.
Ƙungiyar ta rubuta nata sigar Unix kuma ta ba ta damar aiki a ainihin lokaci don tabbatar da dacewa da sarrafa kansa na masana'antu da makamantansu. Injiniyoyin Bell Labs sun kuma ƙirƙiro domino logic, wanda ya ƙara saurin sarrafawa ta hanyar rage jinkiri a cikin ƙofofin dabaru masu rikitarwa.
An ƙirƙiro ƙarin dabarun gwaji da tabbatarwa tare da tsarin Bellmac-32, wani aiki mai rikitarwa na tantancewa da gwaji mai yawa wanda Jen-Hsun Huang ke jagoranta wanda ya cimma sifili ko kusan sifili a cikin kera guntu mai rikitarwa. Wannan shine karo na farko a duniyar gwajin da'ira mai girma (VLSI). Injiniyoyin Bell Labs sun ƙirƙiro wani tsari mai tsari, sun sake duba aikin abokan aikinsu akai-akai, kuma daga ƙarshe sun cimma haɗin gwiwa mara matsala a tsakanin iyalai da yawa na guntu, wanda ya ƙare da cikakken tsarin kwamfuta.
Na gaba shine ɓangaren da ya fi ƙalubale: ainihin kera guntu.
"A lokacin, tsarin aiki, gwaji, da fasahar kera kayayyaki masu yawan amfani sun yi karanci sosai," in ji Kang, wanda daga baya ya zama shugaban Kwalejin Kimiyya da Fasaha ta Koriya (KAIST) kuma memba na IEEE. Ya lura cewa rashin kayan aikin CAD don tabbatar da cikakken guntu ya tilasta wa ƙungiyar buga manyan zane-zanen Calcomp. Waɗannan zane-zanen sun nuna yadda ya kamata a shirya transistor, wayoyi, da haɗin gwiwa a cikin guntu don samar da sakamakon da ake so. Ƙungiyar ta haɗa su a ƙasa da tef, ta samar da babban murabba'i mai zane sama da mita 6 a gefe. Kang da abokan aikinsa sun zana kowace da'ira da fensir mai launi, suna neman haɗin gwiwa da ya karye kuma haɗin gwiwa da ya yi karo ko kuma wanda ba a sarrafa shi yadda ya kamata ba.
Da zarar an kammala ƙirar zahiri, ƙungiyar ta fuskanci wani ƙalubale: ƙera guntun. An samar da guntun a masana'antar Western Electric da ke Allentown, Pennsylvania, amma Kang ya tuna cewa yawan amfanin ƙasa (kashi na guntun wafer ɗin da ya cika ƙa'idodin aiki da inganci) ya yi ƙasa sosai.
Domin magance wannan matsala, Kang da abokan aikinsa sun tuka mota zuwa masana'antar daga New Jersey kowace rana, suna naɗe hannayensu suka yi duk abin da ya wajaba, ciki har da share benaye da daidaita kayan aikin gwaji, don gina abokantaka da kuma shawo kan kowa cewa mafi sarkakiya samfurin da masana'antar ta taɓa ƙoƙarin samarwa za a iya yin sa a can.
"Tsarin gina ƙungiyar ya tafi cikin sauƙi," in ji Kang. "Bayan 'yan watanni, Western Electric ta sami damar samar da guntu masu inganci a adadi da ya wuce buƙata."
An fitar da sigar farko ta Bellmac-32 a shekarar 1980, amma ta kasa cika tsammaninta. Mitar aikinta da aka tsara shi ne 2 MHz kawai, ba 4 MHz ba. Injiniyoyin sun gano cewa kayan aikin gwajin Takeda Riken na zamani da suke amfani da su a lokacin suna da lahani, tare da tasirin layin watsawa tsakanin na'urar bincike da kan gwajin yana haifar da ma'auni marasa daidai. Sun yi aiki tare da ƙungiyar Takeda Riken don ƙirƙirar teburin gyara don gyara kurakuran aunawa.
Kwamfutocin Bellmac na ƙarni na biyu suna da saurin agogo fiye da 6.2 MHz, wani lokacin har zuwa 9 MHz. Ana ɗaukar wannan a matsayin mai sauri sosai a lokacin. Mai sarrafa Intel 8088 mai bit 16 da IBM ya fitar a cikin kwamfutarsa ta farko a 1981 yana da saurin agogo na 4.77 MHz kawai.
Me yasa Bellmac-32 bai yi ba't zama babban abu
Duk da alƙawarin da ta yi, fasahar Bellmac-32 ba ta samu karbuwa sosai a kasuwa ba. A cewar Condrey, AT&T ta fara duba kamfanin kera kayan aiki NCR a ƙarshen shekarun 1980 kuma daga baya ta koma ga siyan kayayyaki, wanda ke nufin kamfanin ya zaɓi tallafawa layukan samfuran guntu daban-daban. A lokacin, tasirin Bellmac-32 ya fara ƙaruwa.
Condry ya ce, "Kafin Bellmac-32, NMOS ta mamaye kasuwa." "Amma CMOS ta canza yanayin domin ta tabbatar da cewa hanya ce mafi inganci ta aiwatar da ita a cikin fasaha."
A tsawon lokaci, wannan fahimta ta sake fasalin masana'antar semiconductor. CMOS zai zama tushen na'urorin sarrafa bayanai na zamani, wanda zai ƙarfafa juyin juya halin dijital a cikin na'urori kamar kwamfutocin tebur da wayoyin komai da ruwanka.
Gwajin Bell Labs mai ƙarfin hali—ta amfani da tsarin kera kayayyaki wanda ba a gwada shi ba kuma ya mamaye tsararrun tsarin gine-ginen guntu—ya kasance wani muhimmin ci gaba a tarihin fasaha.
Kamar yadda Farfesa Kang ya faɗa: “Mun kasance a sahun gaba a cikin abin da zai yiwu. Ba wai kawai muna bin wata hanya da muke bi ba ne, muna kuma nuna sabuwar hanya.” Farfesa Huang, wanda daga baya ya zama mataimakin darakta na Cibiyar Microelectronics ta Singapore kuma shi ma ɗan ƙungiyar IEEE ne, ya ƙara da cewa: “Wannan ya haɗa da ba kawai tsarin guntu da ƙira ba, har ma da babban tabbatar da guntu - ta amfani da CAD amma ba tare da kayan aikin kwaikwayo na dijital na yau ko ma allon burodi ba (hanya ce ta yau da kullun ta duba ƙirar da'irar tsarin lantarki ta amfani da guntu kafin a haɗa sassan da'irar tare har abada).”
Condry, Kang da Huang sun yi waiwaye a lokacin da kyau kuma sun nuna sha'awarsu ga ƙwarewa da sadaukarwar ma'aikatan AT&T da yawa waɗanda ƙoƙarinsu ya sa aka sami damar samar da guntuwar Bellmac-32.
Lokacin Saƙo: Mayu-19-2025