The History of the Atomic Bomb

8/13/2019 The History of the Atomic Bomb

http://slidepdf.com/reader/full/the-history-of-the-atomic-bomb 1/56

The History of the Atomic BombDevelopment and History of the Atomic Bomb and The Manhattan Project

"My God, what have we done" ! obert #ewis, the co!pilot of

the $nola Gay, the B!%& that dropped the first atomicbomb'

On August 2, 1939, just before the beginning of WorldWar II, Albert Einstein wrote to then President Franklin

! "oose#elt! Einstein and se#eral other s$ientists told"oose#elt of efforts in %a&i 'er(an) to *urif) uraniu(+

23, whi$h $ould be used to build an ato(i$ bo(b! Itwas shortl) thereafter that the -nited .tates'o#ern(ent began the serious undertaking known thenonl) as /0he anhattan Proje$t!/ .i(*l) *ut, the

anhattan Proje$t was $o((itted to e*editing resear$h that would *rodu$e a#iable ato(i$ bo(b! 0he (ost $o(*li$ated issue to be addressed in (aking of an ato(i$ bo(b was the*rodu$tion of a(*le a(ounts of /enri$hed/ uraniu( to sustain a $hain rea$tion! At

the ti(e, uraniu(+23 was #er) hard to etra$t! In fa$t, the ratio of $on#ersionfro( uraniu( ore to uraniu( (etal is 41! 5o(*ounding this, the one *art of

uraniu( that is finall) refined fro( the ore is o#er 996 uraniu(+237, whi$h is*ra$ti$all) useless for an ato(i$ bo(b! 0o (ake the task e#en (ore diffi$ult, theuseful -+23 and nearl) useless -+237 are isoto*es, nearl) identi$al in their$he(i$al (akeu*! %o ordinar) $he(i$al etra$tion (ethod $ould se*arate the(8onl) (e$hani$al (ethods $ould work! A (assi#e enri$h(ent laborator)*lant was $onstru$ted at Oak "idge, 0ennessee!

:arold 5! -re) and his $olleagues at 5olu(bia -ni#ersit) de#ised an etra$tions)ste( that worked on the *rin$i*le of gaseous diffusion, and Ernest O! ;awren$e

<in#entor of the 5)$lotron= at the -ni#ersit) of 5alifornia in >erkele) i(*le(ented a*ro$ess in#ol#ing (agneti$ se*aration of the two isoto*es! %et, a gas $entrifuge was used to further se*arate the lighter -+23 fro( thehea#ier, non+fissionable -+237! On$e all of these *ro$edures had been $o(*leted,

all that needed to be done was to *ut to the test the entire $on$e*t behind ato(i$fission </s*litting the ato(,/ in la)(an?s ter(s=! O#er the $ourse of si )ears, fro( 1939 to 19@, (ore than 2 billion was s*entduring the histor) of the anhattan Proje$t! 0he for(ulas for refining uraniu( and*utting together a working ato(i$ bo(b were $reated and seen to their logi$al endsb) so(e of the greatest (inds of our ti(e! 5hief a(ong the *eo*le who unleashed

the *ower of the ato( was B! "obert O**enhei(er, who o#ersaw the *roje$t fro($on$e*tion to $o(*letion! Finall), the da) $a(e when all at ;os Ala(os would find out if /0he 'adget/ <$ode+na(ed as su$h during its de#elo*(ent= was going to be the $olossal dud of the

$entur) or *erha*s an end to the war! It all $a(e down to a fateful (orning in(idsu((er, 19@! At 4294@ <ountain War 0i(e= on Bul) 1C, 19@, in a white bla&e that stret$hedfro( the basin of the Be(e& ountains in northern %ew ei$o to the still+darkskies, /0he 'adget/ ushered in the Ato(i$ Age! 0he light of the e*losion thenturned orange as the ato(i$ fireball began shooting u*wards at 3C feet *erse$ond, reddening and *ulsing as it $ooled! 0he $hara$teristi$ (ushroo( $loud ofradioa$ti#e #a*or (ateriali&ed at 3, feet! >eneath the $loud, all that re(ained

of the soil at the blast site were frag(ents of jade green radioa$ti#e glass $reated

b) the heat of the rea$tion!

http://inventors.about.com/library/inventors/bleinstein.htm

http://inventors.about.com/library/inventors/bllawrence.htm


http://inventors.about.com/library/inventors/bllawrence.htm



0he brilliant light fro( the detonation *ier$ed the earl) (orning skies with su$h

intensit) that residents fro( a farawa) neighboring $o((unit) would swear thatthe sun $a(e u* twi$e that da)! E#en (ore astonishing is that a blind girl saw theflash 12 (iles awa)! -*on witnessing the e*losion, its $reators had (ied rea$tions! Isidor "abi feltthat the eDuilibriu( in nature had been u*set ++ as if hu(ankind had be$o(e a

threat to the world it inhabited! B! "obert O**enhei(er, though e$stati$ about thesu$$ess of the *roje$t, Duoted a re(e(bered frag(ent fro( the >haga#ad 'ita! /Ia( be$o(e eath,/ he said, /the destro)er of worlds!/ en >ainbridge, the test

dire$tor, told O**enhei(er, /%ow we?re all sons of bit$hes!/ After #iewing the results se#eral *arti$i*ants signed *etitions against loosing the

(onster the) had $reated, but their *rotests fell on deaf ears! 0he Bornada deluerto of %ew ei$o would not be the last site on *lanet Earth to e*erien$e anato(i$ e*losion! (cientists )ho *nvented the Atomic Bomb +nder the Manhattan Project"obert O**enhei(er, a#id >oh(, ;eo .&ilard, Eugene Wigner, Otto Fris$h, "udolfPeierls, Feli >lo$h, %iels >ohr, E(ilio .egre, Ba(es Fran$k, Enri$o Fer(i, lausFu$hs and Edward 0eller! iew a $o*) of the letter Einstein wrote "oose#elt that

*ro(*ted the anhattan Proje$t!

-o dia % de a.osto de /&0&, lo.o antes o come1o de ($G2-DA G2$AM2-D*A#, o Albert $instein escreve+ a ent3o Presidente 4ran5lin D'oosevelt' 6 $instein e v7rios o+tros cientistas falaram para o ooseveltde esfor1os na -a8i Germany de p+rificar +r9nio!%0: ;+e poderia ser+sado para constr+ir +ma bomba at<mica' $ra brevemente depois disso;+e o Governo de $stados 2nidos come1o+ o empreendimento s=rioconhecido ent3o s> como "6 Projeto de Manhattan'" (implesmente ponha,o Projeto de Manhattan foi cometido a despachar pes;+isa ;+e prod+8iria+ma bomba at<mica vi7vel'

6 ass+nto mais complicado a ser endere1ado fa8endo de +ma bombaat<mica era a prod+13o de amplas ;+antias de +r9nio "enri;+ecido" paras+stentar +ma rea13o de cadeia' -a ocasi3o, +r9nio!%0: era m+ito d+roe?trair' -a realidade, a rela13o de convers3o de min=rio de +r9nio parametal de +r9nio = :@@/' ompondo isto, a +ma parte de +r9nio ;+e =finalmente refinado do min=rio est7 em cima de && +r9nio!%0C ;+e =praticamente intil para +ma bomba at<mica' Para fa8er a tarefa at=mesmo mais difEcil, o 2!%0: til e 2!%0C ;+ase intil s3o isotopes, ;+aseidFntico na ma;+ila.em ;+Emica delesdelas' -enh+m m=todo de e?tra13o;+Emico ordin7rio poderia os separar s> m=todos mec9nicos poderiamtrabalhar'

2m laboratoryplant de enri;+ecimento vol+moso foi constr+Edo em +mede arvalho, Tennessee' Harold ' 2rey e os cole.as dele em 2niversidadede ol+mbia inventaram +m sistema de e?tra13o ;+e trabalho+ noprincEpio de dif+s3o .asosa, e $rnest 6' #awrence Io inventor do iclotronJna 2niversidade de alif>rnia em Ber5eley implemento+ +m processo ;+eenvolve separa13o ma.n=tica do dois isotopes'

#o.o, +ma centrEf+.a de .7s foi +sada para adicional separe o 2!%0: maisclaro do 2!%0C mais pesado, non!capa8 de fiss3o n+clear' 2ma ve8 todosestes procedimentos tinham sido completados, t+do a;+ilo precisados serfeito eram colocar a teste o conceito inteiro atr7s de fendimento at<mico

I"dividindo o 7tomo", na cl7+s+las de lei.oJ'

http://inventors.about.com/library/inventors/blnuclear.htm#oppenheimer

http://inventors.about.com/library/inventors/blnuclear.htm#Szilard

http://inventors.about.com/library/inventors/blfermi.htm

http://inventors.about.com/library/inventors/bl_edward_teller.htm

http://inventors.about.com/library/inventors/bl_einstein_letter.htm

http://inventors.about.com/library/inventors/blnuclear.htm#oppenheimer

http://inventors.about.com/library/inventors/blnuclear.htm#Szilard

http://inventors.about.com/library/inventors/blfermi.htm

http://inventors.about.com/library/inventors/bl_edward_teller.htm

http://inventors.about.com/library/inventors/bl_einstein_letter.htm



$m cima do c+rso de seis anos, de /&0& a /&K:, mais de L% bilh3o estava.asto d+rante a hist>ria do Projeto de Manhattan' 4oram criadas asf>rm+las por refinar +r9nio e re+nir +ma bomba at<mica de f+ncionamentoe c+idadas dos fins l>.icos delesdelas por al.+mas das maiores mentesde nosso tempo' hefe entre as pessoas ;+e soltaram o poder do 7tomoera ' obert 6ppenheimer ;+e vi.io+ o projeto de concep13o para

concl+s3o'

4inalmente, o dia veio ;+ando t+do a #os Alamos descobririam se "6Dispositivo" Ic>di.o!nomeo+ como tal d+rante se+ desenvolvimentoJ iaser a dro.a colossal do s=c+lo o+ talve8 +m fim para a .+erra' T+do desce++ma manh3 fatal em solstEcio de ver3o, /&K:'

Ns :%&K: ITempo de G+erra MontFsJ no dia /O de j+lho de /&K:, em +machama branca ;+e estiro+ da bacia das Montanhas de eme8 em -ovoM=?ico do norte para os c=+s ainda!esc+ros, ;+e "6 Dispositivo" cond+8i+na *dade At<mica' A l+8 da e?plos3o ent3o se ficada laranja como o .lobode fo.o at<mico come1o+ a atirar acima a 0O@ p=s por se.+ndo, en;+anto

avermelhando e p+lsando como esfrio+' A n+vem em forma de co.+melocaracterEstica de vapor radioativo materiali8o+ a 0@,@@@ p=s' $m bai?o dan+vem, t+do a;+ilo permanecido da terra no local de e?plos3o erafra.mentos de jade ;+e copo radioativo verde crio+ pelo calor da rea13o'Part Two -+clear 4ission-+clear 4+sion

elated #in5s -+clear *nnovations ore detailed infor(ation on different nu$lear in#entions and the histor) of nu$lear*h)si$s! 0i(eline of nu$lear te$hnolog) and (ajor figures in radiation histor)! Albert $instein 'er(an+A(eri$an *h)si$ist who de#elo*ed the s*e$ial and general theories ofrelati#it) + the iron) for this idealisti$ (an was that his fa(ous *ostulation of anenerg)+(ass eDuation, whi$h states that a *arti$le of (atter $an be $on#erted intoan enor(ous Duantit) of energ), had its s*e$ta$ular *roof in the $reation of the

ato(i$ and h)drogen bo(bs, the (ost destru$ti#e wea*ons e#er known! Pl+toni+m on the *nternet

*nventors of the Modern omp+terThe 4irst 4reely Pro.rammable omp+ter invented by onrad

http://inventors.about.com/library/weekly/aa050300b.htm

http://inventors.about.com/library/inventors/blnuclear.htm


http://inventors.about.com/gi/dynamic/offsite.htm?site=http://www.nci.org/index.htm

http://inventors.about.com/library/weekly/aa050300b.htm

http://inventors.about.com/library/inventors/blnuclear.htm


http://inventors.about.com/gi/dynamic/offsite.htm?site=http://www.nci.org/index.htm



Q+se

>) ar) >ellis

onrad Q+se <191+199= was a $onstru$tion engineer for the :ens$hel Air$raft5o(*an) in >erlin, 'er(an) at the beginning of WWII! onrad Guse earned the

se(ioffi$ial title of /in#entor of the (odern $o(*uter/ for his series of auto(ati$$al$ulators, whi$h he in#ented to hel* hi( with his length) engineering $al$ulations!

Guse has (odestl) dis(issed the title while *raising (an) of the in#entions of his

$onte(*oraries and su$$essors as being eDuall) if not (ore i(*ortant than hisown!

One of the (ost diffi$ult as*e$ts of doing a large $al$ulation with either a slide ruleor a (e$hani$al adding (a$hine is kee*ing tra$k of all inter(ediate results andusing the(, in their *ro*er *la$e, in later ste*s of the $al$ulation! onrad Guse

wanted to o#er$o(e that diffi$ult)! :e reali&ed that an auto(ati$+$al$ulator de#i$ewould reDuire three basi$ ele(ents4 a $ontrol, a (e(or), and a $al$ulator for the

arith(eti$!

onrad Q+seRs Q/ irca /&0C

In 193C, Guse (ade a (e$hani$al $al$ulator $alled the G1, the first binar)$o(*uter! Guse used it to e*lore se#eral groundbreaking te$hnologies in $al$ulator

de#elo*(ent4 floating+*oint arith(eti$, high+$a*a$it) (e(or) and (odules orrela)s o*erating on the )esno *rin$i*le! Guse?s ideas, not full) i(*le(ented in the

onrad Q+se *nventors of the Modernomp+ter (eries

H Table of ontents H -e?t hapter Bohn Atanasoff 5lifford >err)and theA>5 5o(*uter $-T$

onrad Q+se on the )ebH More eadin. on onradQ+se onrad Guse biogra*h) andte$hni$al details on his$o(*uters!

http://inventors.about.com/mbiopage.htm


http://inventors.about.com/library/blcoindex.htm

http://inventors.about.com/library/weekly/aa050898.htm

http://inventors.about.com/library/inventors/blzuse.htm



http://inventors.about.com/library/blcoindex.htm

http://inventors.about.com/library/weekly/aa050898.htm






G1, su$$eeded (ore with ea$h G *rotot)*e!

In 1939, Guse $o(*leted the G2, the first full) fun$tioning ele$tro+(e$hani$al$o(*uter!

onrad Guse $o(*leted the G3 in 19@1, with re$)$led (aterials donated b) fellowuni#ersit) staff and students! 0his was the world?s first ele$troni$, full)

*rogra((able digital $o(*uter based on a binar) floating+*oint nu(ber andswit$hing s)ste(! Guse used old (o#ie fil( to store his *rogra(s and data for the

G3, instead of using *a*er ta*e or *un$hed $ards! Pa*er was in short su**l) in'er(an) during the war!

A$$ording to /0he ;ife and Work of onrad Guse/ <b) :orst Guse=

In 19@1, the G3 $ontained al(ost all of the features of a (odern$o(*uter as defined b) Bohn #on %eu(ann and his $olleagues in

19@C! 0he onl) e$e*tion was the abilit) to store the *rogra( in the(e(or) together with the data! onrad Guse did not i(*le(ent thisfeature in the G3, be$ause his C@+word (e(or) was too s(all to

su**ort this (ode of o*eration! ue to the fa$t that he wanted to$al$ulate thousands of instru$tions in a (eaningful order, he onl)

used the (e(or) to store #alues or nu(bers!

0he blo$k stru$ture of the G3 is #er) si(ilar to a (odern $o(*uter!0he G3 $onsisted of se*arate units, su$h as a *un$h ta*e reader,$ontrol unit, floating+*oint arith(eti$ unit, and in*utout*ut de#i$es!

onrad Guse wrote the first algorith(i$ *rogra((ing language $alled ?PlankalkJl? in

19@C, whi$h he used to *rogra( his $o(*uters! :e wrote the world?s first $hess+*la)ing *rogra( using PlankalkJl!

0he Plan5al5Sl lan.+a.e in$luded arra)s and re$ords and used a st)le of

assign(ent <storing the #alue of an e*ression in a #ariable= in whi$h the new#alue a**ears in the right $olu(n! An arra) is a $olle$tion of identi$all) t)*eddata ite(s distinguished b) their indi$es <or /subs$ri*ts/=, for ea(*le writtenso(ething like AKi,j,kL, where A is the arra) na(e and i, j and k are theindi$es! Arra)s are best when a$$essed in an un*redi$table order! 0his is in$ontrast to lists, whi$h are best when a$$essed seDuentiall)!

Guse was unable to $on#in$e the %a&i go#ern(ent to su**ort his work for a$o(*uter based on ele$troni$ #al#es! 0he 'er(ans thought the) were $lose to

winning the War and felt no need to su**ort further resear$h! 0he G1 through G3 (odels were destro)ed during the war along with GuseA**aratebau, the first $o(*uter $o(*an) that Guse for(ed in 19@! Guse left forGuri$h to finish his work on the G@, s(uggling the G@ fro( 'er(an) in a (ilitar)tru$k, whi$h he hid in stables on route to Guri$h, .wit&erland! :e $o(*leted andinstalled the G@ in the A**lied athe(ati$s i#ision of Guri$h?s Federal

Pol)te$hni$al Institute, in use there until 19! 0he G@ had a (e$hani$al (e(or)with a $a*a$it) of 1,2@ words and se#eral $ard readers! Guse no longer had to use

(o#ie fil( to store *rogra(s8 he $ould now use *un$hed $ards! 0he G@ had*un$hes and #arious fa$ilities to enable fleible *rogra((ing in$luding address

translation and $onditional bran$hing! In 19@9, he (o#ed ba$k to 'er(an) to for(a se$ond $o(*an) $alled Guse ' for the $onstru$tion and (arketing of his

designs! Guse later rebuilt (odels of the G3 in 19C and the G1 in 197@!

http://inventors.about.com/library/inventors/blhollerith.htm

http://inventors.about.com/library/inventors/blhollerith.htm



Konrad Zuse

Born June 22, 1910, Berlin-Wilmersdorf; German inventor of pre-war

electromechanical binar computer desi!nated "1 which was destroed without trace b

wartime bombin!; developed two more machines before the end of the war but was

unable to convince the #a$i !overnment to support his wor%; fled with the remains of

"& to "urich where he developed the "' which was successfull used at ()*+

eveloper of a basic pro!rammin! sstem %nown as .lan%al%/l with which he

desi!ned a chess plain! pro!ram+ Education and Experience: B 192 onrad "use

had enrolled at the )echnical niversit in Berlin-3harlottenbur! and be!an his

wor%in! career as a desi!n en!ineer 45tati%er6 in the aircraft industr 4*enschel

7lu!$eu!wer%e6 and b 19&8 he had completed a de!ree in civil en!ineerin!+ *e

remained in Berlin from the time he finished his de!ree until the end of the war in 19'8,

and it was durin! this time that he constructed his first di!ital computers+ *e later

formed his own compan for the construction and mar%etin! of his desi!ns+ Honors and

Awards: *onorarprofessor, Geor!-u!ust-niversitat, G:ttin!en6, 19; Honorary

Degrees; r+-<n!+(+h+, )++ Berlin-3harlottenbur!, 198; r+rer+nat+h+c+, niversit of

*ambur!, 199; r+rer+nat+h+c, )++ resden, 19=1; r+techn+h+c+, niversitait>e%?avi%, <celand, 19=; r+rer+nat+h+c+, niversit of ortmund, 1991;



r+h+c+sc+techn+, ()* (id!enossische )echnische *ochschule, "urich, 1991; r+-

<n!+(+h+, *ochschule f+ rchite%tur und Ballwesen, Weimar, 1991; ottore ad honorem

in @atematica, niversit of 5iena, <tal, 1992; <nlAndische us$eichnun!en(hrun!enC

Werner-von-5iemens->in!, 5tiftun! Werner-von- 5iemens->in!, 19'; ieselmedaille

in Gold, (D eutscher (rfinder- Derband#urnber!, 199; Grosses Derdienst%reu$

des Derdienstordens der Bundesrepubli% eutschland, 192,, mit 5tern, 19=86;(hrenmit!lied der eutschen %ademie der #aturforscher E(F.FE<#, *alle5aale,

192; achener und @unchener .reis, 3arl-rthur-.astor- 5tiftun!, uratorium der

achener und @unchener Dersicherun!s G, 19=0; (hrenpla%ette der 5tadt Bad

*ersfeld, 19=0; onrad-"use-@edaille, "B"entralverband des eutschen

Bau!ewerbes6 19=& Bernhard-Weiss-.la%ette, D@Derband eutscher @aschinen-

und nla!enbau e+D+usseldorf, 19=1; Baerischer @aimiliansorden, Baerischer

@inisterprasident, 19='; Goldener (hrenrin!, eutsches @useum@unchen, 19=';

3othenius-@edaille, E(F.FE<#, eutsche %ademie der #aturforscher*alle -

5aale, 19=8 ; (rnst->euter-.la%ette, 5enat Berlin, 19=8; D(-(hrenrin!, Derband

eutscher (le%trotechni%er e+D+usseldorf, 19=; .hilip-@orris-(hrenpreis, .hilip

@orris Gmb*B(<, 19=; Wilhelm-Eeuschner-@edaille, *essischer@inisterprasident, 19=; (hrenmit!liedC eutsche %ademie der #aturforscher

E(opoll<#, *alle5aale, 192; %ademischer Derein @otiv, 19=2; Derein des

5chleswi!-*olsteinisches @useums fur >echen- und 5chreibtechni% e+D+, lt$enhol$,

heuteC @<3F@, 19=&; Gesellschaft fur <nformati% e+D+G< 19= Derein islandischer

<n!enieure, >e%?avi%<sland, 19=8; eutsches @useum, @unchen, 1990; Dereini!un!

der 7reunde und 7:rderer der <n!enieur- schule an der 7achhochschule 5chmal%alden

e+D+, 1992; (hrenburaerrechtC (hrenbur!errecht der 5tadt *unfeld, 198;

#amensHebunHC onrad-"use-5trasse, in Bad *ersfeld*essen, 192; onrad-"use-

5chule, in */nfeld*essen, 19=; onrad-"use-@edaille, "B "entralverband des

eutschen Bau!ewerbes und G<Gesellschaft f+ lnformati% e+D+, 19=1; onrad-"use-

"entrum f/r <nformationstechni% Berlin"<B, in Berlin, 19='; onrad-"use-"ertifi%at,

7reundes%reis der Berufl+ 5chulen e+D+Bad *ersfeld, 19=8; "use->aum, Berufliche

5chulenBad *ersfeld, 19=8; onrad-"use-Gesellschaft, Grundun! am + 9+ == in

*/nfeld, 19==; onrad-"use-*aus, 7a+ .5 .ro!ramm I 5oftware Gmb*,

>otenbur!Wumme, 19=9; onrad-"use-.ro!ramm, 7:rderun! v+ Gastdo$enten

auslAndischer *ochschullehrer - eutscher %a-demischer ustauschdienst,

Bonn, 1991; onrad-"use-"immer, 5chel$tor-Gmnasium, (sslin!en, 1991;

"usestrasse, in 0-*oerswerda, 1991+

During 1936 to 1938 Konrad Zuse developed and built the first binary digitalcomputer in the world (Zl! " copy of this computer is on display in the #useum

for $ransport and $echnology (%#useum fur &er'ehr und $echni'% (since 1989in erlin!$he first fully functional program)controlled electromechanical digital computerin the world (the Z3 was completed by Zuse in 19*1+ but was destroyed in19** during the war! ecause of its historical importance+ a copy was made in196, and put on display in the -erman #useum (%Deutsches #useum% in#unich!.e/t came the more sophisticated Z*+ which was the only Zuse Z)machine tosurvive the war! $he Z* was almost complete when+ due to continued air raids+it was moved from erlin to -ottingen where it was installed in the laboratory ofthe "erodynamische &ersuchanstalt (D&02/perimental "erodynamics

nstitute! t was only there for a few wee's before -ottingen was in danger ofbeing captured and the machine was once again moved to a small village



%4interstein% in the "llgauavaria! 5inally it was ta'en to wit7erland where itwas installed in the 2$4 (5ederal olytechnical nstitute%2idgenossisch$echnische 4ochschule% in Zurich in 19,! t was used in the nstitute of "pplied #athematics at the 2$4 until 19!

My first computer and first thoughts about data processing1K

started in 193*+ wor'ing independently and without 'nowledge of otherdevelopments going on around me! n fact+ hadn:t even heard of ;harlesabbage when embar'ed on my wor'! "t that time+ the computing industrywas limited to mechanical calculators using the decimal system! unched carddevices were slightly further developed and able to deal with relatively comple/operations for statistical and accounting purposes! 4owever+ these machineswere almost entirely designed for commercial application! $his meant thatmathematicians and engineers had to develop computers on their own+ wor'ingindependently from one another! was no e/ception! "t the beginning of the 3,s+ while studying civil engineering in erlin+ decidedto develop and build bigger calculating machines+ more suitable for engineeringpurposes! approached the problem from various angles<5irstly+ from a logical and mathematical point of view<$his involved1! program control+=! the binary system of numbers+3! and floating point arithmetic!$oday+ these concepts are ta'en for granted+ but at the time this was newground for the computing industry!

econdly+ from the design angle<1! allowing fully automatic arithmetical calculation+=! a high)capacity memory+3! and modules or relays operating on the yesno principle!#y research was initially aimed at pure number calculation+ but soon led on(19336 to new ideas about %computing% in general! ersonally+ believe thatwas the birth of modern computer science! recogni7ed that computing couldbe seen as a general means of dealing with data and that all data could berepresented through bit patterns+ generally spea'ing!$hat led to my basic hypothesis that<%data processing starts with the bit%

"t that time+ of course+ didn:t tal' of %bits%+ but of %yesno status%! >n the basisof this hypothesis defined %computing% as%the formation of new data from input according to a given set of rules%$his basic theory meant that all computing operations could be carried out byrelays operating according to the dual status principle ?ust mentioned! $he mostsuitable devices available at the time were telephone relays!.ow a lin' with mathematical logic had been forged! "s an engineer had noidea of the e/istence of such a discipline! developed a system of %conditionalpropositions% for relays ) something that corresponded appro/imately to what is'nown as oolean algebra today! #y former mathematics teacher showed methat this sort of calculation was identical with the propositional calculus of

mathematical logic!



5rom the engineering point of view+ the gap between this and puremathematical logic was bridged in order to simplify the design and programmingof computing machines! "t roughly the same time in 2ngland+ themathematician and logician "lan $uring was in the process of solving thisproblem from a different angle! 4e used a very simple computer as a model in

order to place theoretical logic on a more formal basis! $uring:s wor' was ofma?or importance for the theory of computer science! 4owever+ his ideas hadlittle influence on the practical development of computing machines!$he theories needed to be put into practice! 5irst of all high)capacity memorieshad to be designed! "t that time+ (193+ memory consisted of single registersoperating a system of numbered wheels using the decimal system! $ypicalproblems were the input and retrieval of information+ as well as the choice ofcounters! ;apacity was fairly restricted+ although some punch card machineswere able to deal with up to =, counters! $hese machines generally functionedon the basis that a number could be %added on%!ut a new problem had to be overcome< pure memory was needed without the

adding)on facility+ but with high capacity and a special selection facility+ as wellas an elegant way of communicating with the periphery! thought it was a goodidea to base such a memory device on binary numbers from the outset! #y ideawas to divide the machine up into cells which would be able to hold data for acomplete number+ in other words+ the operational sign+ e/ponent and mantissa(where a floating point was being used+ as well as additional specifications!@sing the yes)no principle a %word% ) as we would call it today ) could be formedfrom a series of bits! $he memory elements only needed to store yes)no values!>ne device that could deal with this type of operation was the electro)magneticrelay+ which can adopt two positions+ %open% or %closed%! 4owever+ at the time felt that the problem could be better solved mechanically! played around withall sorts of levers+ pins+ steel plates+ and so on+ until finally reached what was avery useful solution+ for those days! #y device consisted mainly of pins andsteel plates+ and in principle could be e/tended to 1+,,, words! " propermachine using telephone relays would have needed *,+,,, relays and filled awhole room!$he basic principle was that a small pin could be positioned right or left of asteel lug+ thus memori7ing the value , or 1! nput and retrieval were alsoeffected via a steel)plate construction+ and the individual parts could be stac'edon top of one another in a system of layers! $he address system also usedbinary code!A=B $hese machines had the advantage of being made almost

entirely of steel+ which made them suitable for mass production!ndividual memory elements could be easily arranged in matri/ form+ which wasvery useful as far as constructing computers was concerned! .ot only was anumber memory now available+ but it could also be used to store general datadrawn from practically any source! 0ogic studies conducted at the same timehad already shown that general calculations with any sort of data structure werepossible+ and that this data could be made up entirely of bit combinations! $hatis why had already called the storage system a %combination memory% in thepatent application!$his was something new on the abbage designs! t was clear that programscould be stored provided they were composed of bit combinations ) one reason

why programmable memory had already been patented by 1936!



n the course of pursuing the basic principles of mechanical memory developed a mechanical relay technology! $his applied to both programmingand calculating parts! "t the time it was not clear whether all operations couldbe run according to the yes)no principle+ or even whether that was a good idea!$hat was something that was only discovered later after much hard wor'!

nitially developed various adding machines for binary numbers which usedelements providing up to three or four positions! $his was done using bothelectro)magnetic and mechanical relays! 5inally found a solution which wor'edon the yes)no principle alone! y this time the similarities between essentiallyvery different technologies were becoming increasingly obvious! was facedwith the choice of using either telephone relays or mechanical technology formy computing machine! "s mechanical memory had proved successful ( wasable to build a wor'ing model in si/ wee's+ and because of the frighteningnumber of relays needed in the alternative system+ (around 1+,,,+ decided infavor of the mechanical version+ at first!nventors are often faced with that sort of decision! $oday+ 'now that opting for

relays immediately would have been better! 4owever+ wor'ing on a completelyprivate basis+ with) ?ust the help of some friends+ started to construct amechanical model of the computer! "t first thought it would be possible toproduce it Cuic'ly! n fact it too' two years to set up a half)way functioningmachine which could present to the e/perts! @nfortunately the survivingphotos are not very good and the machine itself proved somewhat unreliable nfact+ with the help of switching algebra+ it proved easy to convert mechanicalrelay circuits for use in electro)magnetic relay technology! "t this point would li'e to mention my friend 4elmut chreyer who was wor'ingon the development of electronic relays at that time! 4elmut was a high)freCuency engineer+ and on completing his studies (around 1936 startedwor'ing at rof! tEblein:s nstitute at the $echnical @niversity in erlin);harlottenburg! 4elmut+ who was a close personal friend of mine+ suddenly hadthe bright idea of using vacuum tubes! "t first thought it was one of his studentpran's ) he was always full of fun and given to fooling around! ut after thin'ingabout it we decided that his idea was definitely worth a try! $han's to switchingalgebra+ we had already married together mechanics and electro)magnetics )two basically different types of technology! Fhy+ then+ not with tubesG $heycould switch a million times faster than elements burdened with mechanical andinductive inertia!$he possibilities were staggering! ut first basic circuits for the ma?or logical

operations such as con?unction+ dis?unction and negation had to be discovered!$ubes could not simply be connected in line li'e relay contacts! Fe agreed that4elmut should develop the circuits for these elementary operations first+ while dealt with the logical part of the circuitry! >ur aim was to set up elementarycircuits so that relay technology could be transferred to the tube system on aone)to)one basis! $his meant the tube machine would not have to beredesigned from scratch! chreyer solved this problem fairly Cuic'ly!$his left the way open for further development! Fe cautiously told some friendsabout the possibilities! $he reaction was anything from e/tremely s'eptical tospontaneously enthusiastic! nterestingly enough+ most criticism came fromchreyer:s colleagues+ who wor'ed with tubes virtually all the time! $hey were

doubtful that an apparatus with =+,,, tubes would wor' reliably! $his criticalattitude was the result of their own e/perience with large transmitters which



contained several hundred tubes! "part from that+ conditions were not e/actlypropitious for the development of a fully tube)operated machine! $he Far hadbro'en out in the interim+ ma'ing the procurement of staff and material verydifficult! .othing could be done by private initiative! Fe therefore proposed theconstruction of a =+,,,)tube computer for special use in anti)aircraft defense to

the military authorities! "lthough the reaction was initially sympathetic towardsthe pro?ect we were as'ed simply+ %4ow much time do you thin' you need foritG%! Fe replied+ %"round two years! $he response to this was+ %"nd ?ust howlong do you thin' it:ll ta'e us to win the warG%! $he outcome was considerableobstruction and delay in the development of a -erman electronic computingmachine! chreyer was by now fully engaged in other pro?ects! y the end ofthe Far he had constructed a small e/perimental machine for 1, binary digitsand around 1,, tubes! ut this machine was also lost in the general confusion ?ust after the Far! "fter the Far was finally over+ news of the @niversity of ennsylvania 2.";machine went all round the world ) %18+,,, tubes%! Fe could only sha'e our

heads! Fhat on earth were all the tubes forG chreyer and parted companyafter the Far! "t that time it was prohibited to develop electronic eCuipment inFest -ermany! "s chreyer saw no means of continuing his very interestingresearch he emigrated to ra7il to ta'e up a university chair!chreyer died in 198!$he 2nglish development 'nown as ;>0>@ was unheard of outside thecircle of those wor'ing on it! t was only much later that the wraps came off thisvery interesting pro?ect! n 198, chreyer and had the opportunity to spea' tothe ;>0>@ people in 2ngland! Fe compared our circuits and it turned outthat there were considerable similarities! $he 2nglish had also been wor'ing onlogical operations and other similar design principles!y the end of 1938 it seemed clear that electro)magnetic relays offered the bestchance of producing a reliable operating computer Cuic'ly! efore redesignedthe Z1 to operate completely with relays made a test with a small pilotmachine+ the Z=! used the mechanical memory of the Z1 with a low storagecapacity (16 words+ as well as the card punch and reader to build a simplecomputer with =,, relays operating with 16 bits and on the basis of fi/ed) pointarithmetic!Houng transmitter specialists+ including chreyer and other friends of mine+helped me design the circuits and choose the appropriate components! utalthough their advice was a great help+ to a certain e/tent e/pressly set out to

e/plore new ground! $he most important thing seemed to be to 'eep thefreCuency absolutely even+ so that one cycle eCualed one addition+ itselfcomprising several steps! 5reCuency was set using rotating dis's or rollerswhich were covered in alternating strips of conducting and non)conductingmaterial+ contact being made via carbon brushes $his principle had manyadvantages! $ests could be run on the machine at any speed! "notheradvantage was that spar' e/tinction too' place at the brushes and not at therelay contacts when circuits were being shutdown! Despite well)meaning advicefrom some friends+ did not ma'e use of certain well)'nown telephonecommunication tric's such as delayed)response relays! "ll in all+ was able to gather enough e/perience with the Z= in order to convert

the complete Z1 design for relay operation! Fhat emerged was the Z3+ which isnow considered to have been the first properly functioning computer in the



world! n order to ma'e fast progress the memory was also given a 6*)wordcapacity+ ma'ing use of relays!$he Z3:s basic specifications were<

• a binar number sstem

• floatin! point arithmetic

• 22-bit word len!th, with 1 bit for the si!n, eponential bits and a 1'-bitmantissa

• 2,'00 relas, 00 in the calculatin! and pro!ram section and 1,=00 in the

memor+

$he calculations possible were addition+ subtraction+ multiplication+ and division+ta'ing the sCuare root+ as well as some ancillary functions! ;onstruction of themachine was interrupted in 1939 when was called up for military service! twas typical of the attitude prevalent in -ermany at the time that should be laterreleased from active service+ not to develop computers+ but as an aircraftengineer! 4owever+ in my spare time+ and with the help of friends+ was able to

complete the machine! y 19*1 it was wor'ing and was able to show it to theaircraft construction authorities! $he -erman "ircraft Iesearch nstitute inerlin)"dlershof showed greatest interest! rofessor $eichmann+ who had beenwor'ing on the problem of wing flutter+ was particularly attracted! @nli'e aircraftstress+ wing flutter results in critical instability due to vibration of the wings+sometimes in con?unction with the tail unit! ;omple/ calculations were neededin order to overcome this design problem! $he most difficult part was calculatingthe so)called %KJssner determinants% based on comple/ numbers and un'nownCuantities in the main diagonal! achieved a brea'through using my eCuipmentfor this calculation! @nfortunately the "ircraft Iesearch nstitute had not beengiven a high enough priority for me to be released from military service! >nlyrofessor 4erbert Fagner+ who was wor'ing on the development of remote)controlled flying bombs+ and for whom wor'ed as a stress analyst+ was in thisenviable position! 4owever+ Fagner was very understanding+ and helped asmuch as possible by allowing me to use some of my wor' time on the pro?ect!y then had already set up my own small engineering business+ the %Zuse)ngenieur)Jro% in erlin! $he Z3 was later destroyed after bombing raids!ecause of its historic importance we rebuilt it =, years later a replica nowstands in the Deutsches #useum in #unich! "round 19*= it was decided to build a more powerful+ improved Z*! Fe thoughtthat we would be able to have it ready within one+ to one)and)a)half years! t

was to have a mechanical memory with a capacity of 1+,=* words+ several cardreaders and punches+ and various facilities to enable fle/ible programming(address translation+ conditional branching!;onstruction of the machine started well but it was not long before the Farimposed its delays! n the end+ construction was not completed until the close ofthe Far! rocurement of staff and materials became increasingly difficult+ andaround 19*3 the erlin blit7 began+ with heavy bomber raids nearly every day!everal times we had to move location with the machine! During the last fewwee's of the Far we found refuge in -Lttingen! $he Z* was the only model wewere able to save+ and this in the face of considerable difficulties! >n the =8 "pril 19* we were able to demonstrate the Z* to rofessors randtl+ et7 and

KJssner! ut the Festern and 2astern fronts were drawing closer daily andnobody could say whether -Lttingen would be bombed or not+ or whether the



Z* was safe there! $he #inistry of "viation ordered us to ta'e the machine tothe underground wor's in the 4ar7! t was there that we first learnt of the terribleconditions under which the so)called reprisal weapons ) the &1 and &= ) werebeing built! Fe refused to leave the machine there+ and+ with the help ofFernher von raun:s staff+ we managed to get hold of a truc' to transport it

elsewhere! "nd so the Z* odyssey continued! Fe then moved south+ ending upin a small "lpine village called 4interstein in the "llgEu+ where we were finallyable to find a good place to store the machine!

"round 19,+ after a number of modifications+ the machine was set up in the$echnical @niversity of wit7erland in Zurich where it remained for severalyears+ the only wor'ing computer in 2urope ! $oday it is a historic model andcan be seen in the Deutsches #useum in #unich! @nfortunately it:s no longer infull wor'ing order! "nother off)shoot of computer research ought to be

mentioned here+ too! y that mean process control! "t the 4enschel "ircraft5actory rofessor 4erbert Fagner+ for whom wor'ed as a stress analyst+ wasinvolved in developing remote) controlled bombs! $o this end+ the tailplane andwings ) which were constructed of metal with a relatively low degree of precision) were sub?ected to detailed measurements using gauges at some 8, differentpoints! $he necessary ad?ustments were then calculated to allow formanufacturing inaccuracies! $his reCuired a rather comple/ calculation! nitially constructed a special)purpose computer for a fi/ed seCuence of operationsusing around ,, relays! $his machine replaced a do7en calculators andwor'ed very reliably for two years+ two shifts a day! $he procedure reCuired amechanic to read off the gauges! $he values were then recorded and operatorsentered the figures into the computer! $his led me to build an improved modelwhich could read the gauges automatically and transfer data directly into thecomputer! $he heart of the machine was a device that today would be called ananalogdigital transformer! erhaps this was the first process control system inthe world! $he machine itself had its own history! t completed trial tests on aproduction line in udetenland+ but never reached full operational use as theFar forced the whole factory to be re)located! $he e/act fate of the machine isun'nown ) it:s possible the factory fell into the hands of the Iussians+ who musthave been the only ones at the time to own a fully)operational computer! $he Z3had already been destroyed+ the Z* was not completed+ and the first @

machines+ #ar' and 2.";+ were not operational at that time! 4owever+ it isunli'ely that the Iussians would have 'nown what to do with the machine evenif they had found it in an undamaged condition! "longside my practical wor'with various computer models started to consider certain theoretical aspects!$he brea'through to a new computing age went hand)in)hand with newscientific ideas and the development of new components! $oday we tal' abouthardware and software! $hese e/pressions were really only introduced muchlater by the "mericans although the terms have now become established! t wasapparent that a special branch of %computer science% was needed! ut shortageof time meant that could only scratch the surface in this field! nitially wor'edon my own+ but towards the end of the Far 4err 0ohmeyer+ an outstanding

mathematician+ was assigned to assist me! 0ohmeyer was a product of 4einrichchol7:s school in #Jnster the latter himself a famous logician! $he lin' with



mathematical logic had already been established! "s a civil engineer wasattracted by the prospect of drawing on predicate and relational calculus ande/ploring the possibilities they offered as a basis for computing! $a'e theframewor's used in building construction for e/ample ) were they not similar tothe graphs used in relational calculusG @sing pair lists+ it was relatively easy to

digitali7e the structure of a framewor' with the aid of relational calculus+ in otherwords+ to brea' it down into its component data! $his could then be entered intothe combination memory+ which had been invented by this time+ and serve as abasis for combination calculations! $his ought to mean that not only purelynumerical calculations could be dealt with+ but construction design itself! @p tillnow only the human mind had been capable of this! $he same idea applied toframewor's and other types of building design! became e/tremely pre)occupied with this new aspect of computing! even went as far as learning toplay chess in order to try to formulate the rules of the game in terms of logicalcalculus! ;hess offered a mass of data structures within a limited space! "symbolic language (the e/pression %algorithmic language% was un'nown to me

at the time which could describe chess problems seemed to me to be suitablefor all computer machine problems! lan'al'Jl was later (19* devised withthis principle in mind!

$his led to my first confrontation with what is 'nown today as %artificialintelligence%! .aturally reali7ed my computer would never be able to run thatsort of calculation! ut combination memory and the general circuitry were astep in the right direction! #any developments were predictable+ of course+others were still in the realm of fantasy! remember mentioning to friends bac'in 1938 that the world chess champion would be beaten by a computer in ,years time! $oday we 'now computers are not far from this goal!

ut even in those days Cuite a lot was achieved at the drawing board! $oday wecall this computer architecture! $he latter type of machine is 'nown as the Mohn)von).eumann computer+ after its namesa'e who first produced it 1, years latertogether with -oldstine and ur's! Fe now 'now this was a very elegantsolution!

$he Cuestion is why did not use this concept in 1939 if already 'new about it!Fell+ at that time it would have been senseless to try to build that sort ofmachine+ as the necessary facilities were simply not available! 5or e/ample+

storage capacity was not big enough to cope ) an efficient program memoryneeds to be able to store several thousand words! peed was also too low! t:strue that floating point arithmetic can be performed by simply following a seriesof single instructions (as is the case today! ut that means giving 1, to =,times as many instructions! "s long as the electronical prereCuisites were notavailable it was a waste of time! $wo things were needed first ) high storagecapacity ) around 8+,,, words+ as in the first magnetic)drum memories ) andelectronic speed! $owards the end of the *,s this seemed possible+ but as-ermans we were not able to participate in this development at the time!

$he possibility of a computer being able to deal with numeric calculations and

logic organi7ation was so e/citing that gave serious thought to a %logicalcomputing machine%! $his led to the %program compiling machine% pro?ect! For'



was to be split between numerical and logic computers! $hat included suchareas as<

N construction of e/tensive programs made up of subprograms according tospecific parameters

N address translation necessary for these programs

N capability of dealing with engineering structures (e!g! framewor's using pairlists from which a numerical program could be developed!

$his pro?ect was commissioned towards the end of the Far by the #inistry of "viation! 4owever+ the wor' soon proved to be too broadly)based and the ideasnever left the drawing board! ince then+ the concept of dividing computersaccording to numerical or logical operation has failed to find favor anywhere!$he dominance of electronics made this unnecessary! $he high speeds

obtainable meant that such operations could be carried out by a single machine!4owever+ one aspect became clear to me in view of all this research between1936 and 19*6! ome means was necessary by which the relationshipsinvolved in calculation operations could be precisely formulated! #y answer was%lan'al'Jl% ) today it would be termed an %algorithmic% language! 4owever+ inthose days+ 'nown mathematical and logical forms were not advanced enough!$here were also several other differences today:s languages<

1! lan'al'Jl was not conceived as a means of programming the Z* or othercomputers available at the time! ts true purpose was to assist in establishingconsistent laws of circuitry+ e!g! for floating point arithmetic+ as well as inplanning the seCuence of instructions a computer would follow ) what we wouldterm %hardware% and %software% today!

=! t was meant to cover the whole spectrum of general calculating!

y contrast+ the program languages which were developed around ten yearslater were relatively one)sided! $hey were designed specifically for e/istingcomputing machines+ in other words+ for the first really fle/ible electroniccomputers! n the first instance+ these languages were concerned withconditional branching+ address translation and suchli'e! $here was hardly any

demand for logic=1 operations+ such as the application of predicate andrelational calculus for engineering constructions+ chess programs and so on!$hat also applied to the brea'down of data into yes)no combinations! n otherwords+ mathematicians did not consider my principle of %data processingstarting with the bit% to be of any fundamental importance! lan'al'Jl:swea'ness was that it went into too much depth with regard to difficultcalculations which seemed better left to the future! $he importance of my chessprogram+ as an e/ample of applied logic+ was simply ignored! n addition to this+in the early fifties+ was completely absorbed in building up my business at atime when program languages started to become more relevant! $his meant could not participate in the debate on "lgol and so on! lan'al'Jl was later

published out of historical interest (in 2nglish+ too+ although it is nowunfortunately out of print!



"s has already been mentioned+ 19* was a hard time for the -ermans! >ur Z*had been transported with incredible difficulty to the small "lpine village of4interstein in the "llgEu! #y group and my erlin firm had been dissolved! "ll ofus who had managed to get out of erlin were happy ?ust to have survived theinferno there! For' on lan'al'Jl now continued in wonderful countryside+

undisturbed by bomber attac's+ telephone calls+ visitors and so on! Fithin abouta year we were able to set up a %revamped% Z* in full wor'ing order in what hadonce been a stable! @nli'e research going on in the @"+ where every possiblefacility was available+ our means were very limited!

$here was no body or organi7ation able to support our wor' at that time in-ermany! .evertheless word got out abroad that some sort of machine wasoperating in outh avaria! #@" instructed the -erman firm 4ollerith-mb4 to see what this was all about! "ll sorts of promising discussions ensuedabout possible applications for computers in various areas! ut as everythingwas decided on the other side of the "tlantic at the time+ no contract was

signed! nterest was only shown in the industrial property rights! t wasn:t evenpossible to secure a promise that would be able to continue wor' ondevelopment! "t that time computers were simply not that important!

4owever+ we did have more success with Iemington Iand+ who commissionedus to continue development+ initially in a special pro?ect dealing with mechanicalrelay technology! eople didn:t trust electronics fully at the time and wanted tohave a second option! Fe ourselves were convinced that the future lay withelectro)magnetics and electronics! $his meant wor'ing against our ownconvictions! .evertheless the machine was interesting as it became the firstpipeline computer in the world ) a punched card machine with a series ofmechanical adding devices using mechanical circuits incorporated between thecard reader and the card punch! #ultiplication was carried out by a series ofadding operations! $his resulted in a reasonable degree of accuracy despite therelatively slow mechanics! ut the development of electronic alternatives wasprogressing at that it was soon not worthwhile continuing with mechanicaldesigns!

4owever+ we were able to produce an interim solution based on our relays priorto the introduction of pure electronic machines+ prototypes of which were beingprepared by Iemington in the @"! $he production run consisted of around

forty additional units for punched card machines+ most of which were e/portedto wit7erland! "bout the same time we had been able to set up the Z* as ascientific machine in the $echnical @niversity in Zurich! oth contracts helpedus reestablish the Z@2 K- company+ which founded near ad 4ersfeld in19*9 with two other partners!

$he -erman mar'et slowly began to develop and we started to receive ordersfrom -erman companies! >ne of our first clients was 0eit7+ which than's to theworld)famous %0eica% had the necessary financial means to purchase acomputer for optics calculations! 0ater other optics firms followed suit+ so that bythe mid),s we had a virtual monopoly in the field of scientific computers in the

optics industry in central 2urope!



y that time everyone was tal'ing about electronic machines+ but their reliabilitystill left a lot to be desired! $he first were built by scientific institutes for theirown special purposes+ as the commercially available machines were not goodenough! Fe were able to bridge this gap with our relay computers! >ne thing inour favor was land consolidation+ which was in full swing in -ermany at the

time! ;omputers were needed to calculate how fields and land were to bereallocated! $o this end we developed the Z11 which was later further refinedand used elsewhere in urban and agricultural surveying+ optics+ etc! $hismachine corresponded to what the users wanted ) namely to always 'now howthe calculation was progressing+ and to have everything under complete control!n the course of time full automation led to a ma7e of branches andimpenetrable procedures+ and software became an increasingly urgent problem!Fe also e/perienced and influenced this+ not always happy+ development withour machines!

"fter some years+ electronic components achieved a degree of reliability that

warranted their production in large numbers! nitially we developed a tubemodel+ the Z==+ moving later to the transistori7ed versions Z=3+ Z31 and Z=!$hese were based on analytical code+ which meant they were e/tremely fle/ibleas far as programming was concerned! $hese were the first types of machinewhere general calculation using any desired data structure and program storagecould be carried out! $he machines were very popular with scientists andengineers+ as they were e/cellent to play around with and all sorts of models forthe most diverse problems could be tried out! adly+ in most cases ourmachines could only serve to whet people:s appetites! Fhile our clients werevery short of means during the early years of the electronic computer+ researchdepartments were later much better funded+ allowing them to purchase muchbigger+ more e/pensive eCuipment! @nfortunately my firm hardly profited fromthis as we were oriented towards small and medium)si7ed companies! .or didwe have enough capital to ta'e part in the development of larger machines! >urZ11+ Z== and Z=3 are now sought after only as collector:s pieces for museumsand so on!

>ne development which received much impetus from specialists wor'ing in thesurveying field was the automatic drafting board! $he aim was the automaticrepresentation of various maps which had been calculated beforehand bycomputer! t is interesting to note that the surveyors were loo'ing for a very high

degree of accuracy and initially only wanted to plot polygon vertices! 5rom thisemerged the -raphomat+ the first computer)controlled automatic drawing board!#any others were interested+ too! $hese were our first tentative steps in thedirection of ;"D! 0ittle is 'nown about this side of our wor'! 4ere+ too+ we foundout that it is not always a good idea to be the early bird! "s early as 196* wasinvolved in negotiations with a ma?or 2uropean carpet manufacturer to build acomputeri7ed control system for a large weaving machine proposed that westart at the design stage of the carpet pattern! $he intention was not to ma'ethe artisan redundant+ but simply to give him a new tool! ut this suggestion metwith complete opposition from all parties concerned+ and the contract failed tocome off!



;ompetition in the computer sector became increasingly tough! .ot only werethe costs of hardware constantly rising+ software development costs were alsogrowing! #y company with its thousand)odd employees faced growing capitalshortages+ ma'ing it necessary to bring in new shareholders! $his led+ step bystep+ to the company being completely ta'en over by iemens!

$oday+ this leaves me free to devote more time to purely scientific problems+and still wor' on a free)lance basis for iemens "- #unich! am currentlyinvolved in computer architecture+ and am particularly interested in the paralleloperation of machines! ac' in the ,s designed a machine for themeteorological office which today would be termed a %cellular computer%! 4ere+too+ however+ was guilty of trying to run before could wal'!

erhaps can round off by mentioning a few of the ideas have e/amined onpaper without ever being able to turn them into reality!

%$he ;omputing @niverse% is based on the idea that the whole cosmos is a 'indof cellular computer+ something that some physicists are seriously consideringtoday! $his theory has yet to be confirmed due to the lac' of e/perimentalevidence! " paper on the sub?ect under the same title has also appeared in2nglishA3B! am sure the idea will gain considerable significance in the futureand might help theoretical physics to solve a number of problems!

"nother idea of mine was %$he elf)Ieproducing ystem%! approached thisconcept differently to Mohn von .eumann+ who dealt with it using puremathematics in the conte/t of cellular computers! "s an engineer was moreinterested in setting up the conditions necessary for actual construction! nessence+ the idea envisages a tool factory which is capable of reproducing itsown essential component parts! $his idea has met with complete opposition!eople have been reluctant to consider such a radical solution for all sorts ofreasons! $oday traditional means of production are being automated step bystep! Fe have yet to build the factory of the future! ut one day these far)sighted developments will become reality+ leading to a complete revolution inthe production process throughout the economy!

Quotations

>f one thing am sure ) computer development has still a long way to go!Houng people have got plenty of wor' ahead of them yet

Bibliography

iographicalauer+ 5riedrich 0! %etween Zuse and Iutishauser ) $he 2arly Development ofDigital ;omputing in ;entral 2urope%+ in #etropolis+ .!+ M! 4owlett+ and -ian);arlo Iota! 198,! " 4istory of ;omputing in the $wentieth ;entury+ "cademicress+ nc!+ .ew Hor'! pp! ,)=*!;eru77i+ aul 2! 1981! %$he 2arly ;omputers of Konrad Zuse+ 193 to 19*%+ Ann. Hist. Comp., &ol! 3+ .o! 3+ pp! =*1)=6=!



;7auderna+ Karl)4ein7! 19O9! %Konrad Zuse+ der Feg 7u seinem ;omputer Z3(Konrad Zuse+ the ath to 4is Z3 ;omputer% Ieport 1=,+ Gesellschaft fürMathematik und Datenverarbeitung, I! >ldenbourg+ #unich+ 1, pp!Iitchie+ David! 1986! $he ;omputer ioneers+ imon P huster+ nc!+ .ewHor'+ ;hapter 3!

chwart7+ 4! I! 1981! %$he 2arly Hears of ;omputing in wit7erland%+ Ann.Hist. Comp., &ol! 3+ .o! 1+ pp! 1=1)13=!later+ Iobert! 198O! ortraits in ilicon+ #$ ress+ ;ambridge #"+ ;hapter !peiser+ "! ! 198,! %$he Ielay ;omputer Z*%+ Ann. Hist. Comp., &ol! =+ .o! 3+pp! =*=)=*!Zemane'+ 4ein7! 1983! %Zuse+ Konrad% in Ialston+ "nthony+ and 2dwin D!Ieilly+ Mr! 1983! 2ncyclopedia of ;omputer cience and 2ngineering+ &an.ostrand Ieinhold ;o!+ .ew Hor'!Zuse+ Konrad! 19O,! Der ;omputer ) #ein 0ebenswer'+ &erlag #odernendustrie+ #unich!Zuse+ Konrad! 198,! %ome Iemar's on the 4istory of ;omputing in -ermany%+

in #etropolis+ .!+ M! 4owlett+ and -ian);arlo Iota! 198,! " 4istory of;omputing in the $wentieth ;entury+ "cademic ress+ nc!+ .ew Hor'! pp! 611)6=O!Zuse+ Konrad! 198,! %nstallation of the -erman ;omputer Z* in Zurich in19,%+ Ann. Hist. Comp., &ol! =+ .o! 3+ pp! =39)=*1!

Footnotes

1K 3omputer esi!n - .ast, .resent, 7uture, )al% !iven b .rof+ onrad "use, in

Eund5weden, 2nd Fctober 19=

2K 7urther details can be found in patent no+ 90 9'=, l+ '2 m, !roup 18 @a 19&

entitled @echanical >ela and patent no+ 92' 10, l+ '2 m, !roup 18 @a 19&@echanical >ela @emor

&K <nternational Journal of )heoretical .hsics, 21st edition, nos -, from p+ 8=9,

June 19=2

;opyright M! "! .! 0ee+ eptember 199*!

The Z1

$he Z1 is today considered to be the first freely programmable computer of theworld! t was completed in 1938 and financed completely from private funds!Konrad Zuse:s first computer+ built between 1936 and 1938+ was destroyed inthe bombardment of erlin in FF + together with all construction plans! n1986+ Konrad Zuse decided to reconstruct the Z1!$he Z1 contained all parts of a modern computer+ e! g! control unit+ memory+micro seCuences+ floating point logic!



Z1 in the apartment of Konrad Zuse:s parents in 1936Konrad Zuse constructed the Z1 in the apartment of his parents+ in fact+ he wasallowed to use the living room for his construction! n 1936 Zuse Cuitted his ?obin airplane construction to build the Z1! 4is parents were not enthusiastic+ butthey did support him anyway they could!

$he rebuilt Z1 in the Deutsche $echni' #useum erlin in 1989n 1986 Konrad Zuse decided to rebuilt the Z1! $he reason was that the Z1contained almost all the important features of a modern computer! n 1986 heconstructed thousands of elements of the Z1 again+ because the original Z1was destroyed by allied airraids in 19*3! n 1989 the rebuilt Z1 was finished andcan be visited in the Deutsche $echni' #useum erlin)Kreu7berg!



0ogic structure of the Z1

&iew of the Z1

Data sheet



#ame of the

machine

"1

<mplementation )hin metal plates, wor%ed with fret saw

7reHuenc 1 *ert$

#umeric unit 7loatin! point unit, 22 Bit word len!th

vera!e calculation

speed

@ultiplication appro+ 8 seconds

<nput ecimal %eboard, automatic binar codin!

Futput ecimal di!its

Word len!th 2' Bit mantissa, = bit eponent, 1 si!n

#umber of relas #o relas, thousands of metal plates, appro+ 20,000 parts

#umber of step

switches

#one

@emor ' cells L 22 Bit

.ower consumption ppro+ 1000 watts for the electric ccle motor

Wei!ht ppro+ 800 %!

rea of application (perimental model, no application, developed for scientific

calculations

#umber of

machines sold

0

3ost in (@ #o price

3omments )he "1 was not reliable+ reconstruction is on displa in theeutsche )echni%museum Berlin+ )he metal plate construction was

too comple+ #et were the "2 and the "&+

zusecs.tu!berlin.de

The Z2

Konrad Zuse had bad luc' with his original Z1! $he Z1 did contain allcomponents of a modern day computer+ however+ the Z1 was not reliable! tsmechanical construction was too comple/ and error prone! $herefore+ KonradZuse decided to construct the Z=! 4e wanted to find out whether the use ofrelays yielded better results! " .umeric unit (no picture available with 8,, relays did wor' well! $he Z= wascomplete in 1939! Zuse now decided to build the Z3+ a computer completelybuilt with relays! t later became to be 'nown as the first fully functionalprogrammable computer!

http://irb.cs.tu-berlin.de/~zuse/Konrad_Zuse/en/Rechner_Z2.html


mailto:[email protected]










Data sheet #ame of the

machine

"2

<mplementation @emorC thin metal plates, wor%ed with fret saw; #umeric unitC

relas

7reHuenc & *ert$

#umeric unit 7i point unit, 1 bit word len!th

vera!e calculation

speed

@ultiplication appro+ 8 seconds

<nput ecimal %eboard, automatic binar conversion

Futput 7laps, binar

Word len!th 1 Bit, fi point

#umber of relas 200

#umber of step

switches

M

@emor 1 cells L 1 Bit

.ower consumption ppro+ 1000 watts

Wei!ht ppro+ 800 %!



rea of application emonstration model, no applications

#umber of

machines sold

0

3ost in (@ #o price

3omments )he "2 was a demonstration model for the rela techniHue, whichwas the basis for the "&+ )he "2 was destroed b bombardment

around 19'0+

zusecs.tu!berlin.de

The Z3

$oday it is undisputed that Konrad Zuse:s Z3 was the firstfully functional+ program controlled (freely programmablecomputer of the world! $he #ar' + the 2."; and the;olossus followed 19*3 and later! $he Z3 was presentedon #ay 1=+ 19*1 to an audience of scientists in erlin! $hedemonstration was a success!$he Z3 as the Z1 contained practically all features ofmodern day computers! $he Z3 was built with relays! $he Z3 did not have a ?ump instruction! Konrad Zuse+ however+ did 'now the ?ump instruction+ as heimplemented it in the micro code for floating point calculations!

.o picture of the original Z3 e/ists! $he above picture was ta'en from areconstruction made in early 196, by Zuse K-+ ad 4ersfeld! $hereconstruction was made for the 196* nterdata industry fair in #unich! n 196O+it was on display at the #ontreal 2/po+ where it received great attention! t is

now owned by the Deutsche #useum in #unich!









)he picture above shows the "& on displa in the eutsche @useum in @uenchen+ )he

ri!ht cupboard contains the relas for the arithmetic unit+ lso the relas on the top of

the left cupboard belon! to the arithmetic unit+ )he relas in the left cupboard with the

li!ht color belon! to the memor 4' words a 22 bits6+ Fn the bottom of the ri!ht

cupboard the micro seHuencers can be seen, which are the heart of the control unit of the

"&+ )he micro seHuencers are reali$ed with stepwise relas+ )he are used to reduce

complicated arithmetic operations to additions or subtractions+ )he input- and output

devices can be seen on the left side+ )he decimal numbers are converted to binar

floatin! point numbers and vice versa+ )he punch tape reader close to the input device

ma%es the "& to a freel$ pro!rammable machine+

)he bloc% structure of the "&+ )here is a separation of the components of the machine,

li%e the control unit, the memor, the arithmetic unit, and the input- and output devices+

)he two >e!ister >1 and >2 are short memories of 22 bits+ n arithmetic operation is

performed as followsC >1 CN >1 I >2+ the contents of >e!ister >1 is added to the



contents of >e!ister >2 and the result is assi!ned to >1+ >2 is set to empt+ )he "& was

a cloc%ed machine of 8+& *ert$+

Data sheet

#ame of the machine "&

<mplementation 00 relas numeric unit, 100 relas stora!e unit

7reHuenc 8-10 *ert$

#umeric unit 7loatin! point unit, 1 steps for multiplication, & steps for addition,

1= steps division

vera!e calculation

speed

@ultiplication & seconds, division & seconds, addition 0+ seconds

<nput ecimal %eboard with 20 di!its, automatic binar codin!

Futput With lamps, ' decimal di!its with decimal point

Word len!th 22 Bit, floatin! pointC mantissa, eponent and si!n

#umber of relas 2000

#umber of step

switches

10 for micro code control of the floatin! point unit

@emor 1'00 relas, ' words L 22 Bit

.ower consumption ppro+ '000 watts

Wei!ht ppro+ 1000 %!

rea of application Win! construction #umber of machines

sold

0

3ost in (@ none

3omments )he "& was destroed throu!h bombardment in 19''+

reconstruction is on displa in the eutsche @useum in @unich+

Details of the first inary For'ing ;omputer Z3 ;ontributions of Konrad Zuse to the #odern ;omputer 4ow to ma'e the computer Z3 to an universal computerG

atent of the Z3

O onrad "uses .atent "&91 for the @achine "& from 19'1 has been proposed for the

#(53F .ro!ram @emor for the World at u!ust &1, 2000+ Please look here, too.

[email protected]

Konrad Zuses Z3 in Detail

Konrad Zuse built the machine Z3 from 1939 to 19*1 in the #ethfesselstraQe Oin erlin)Kreu7berg with some friends and a small support by the government!

http://irb.cs.tu-berlin.de/~zuse/Konrad_Zuse/Z3-detail.htm

http://irb.cs.tu-berlin.de/~zuse/Konrad_Zuse/z-contr.html

http://www.computer.org/annals/an1998/a3051abs.htm

http://www.unesco.de/Aktuell/Ua.htm


http://www.spiegel.de/kultur/gesellschaft/nf/0,1518,91436,00.html


http://irb.cs.tu-berlin.de/~zuse/Konrad_Zuse/Z3-detail.htm

http://irb.cs.tu-berlin.de/~zuse/Konrad_Zuse/z-contr.html

http://www.computer.org/annals/an1998/a3051abs.htm



http://www.spiegel.de/kultur/gesellschaft/nf/0,1518,91436,00.html




Fith the Z3 Konrad Zuse wanted to show+ that it is possible to build a reliablewor'ing machine for very complicated arithmetic calculations+ which is freelyprogrammable and is based on a binary floating point number and switchingsystem! 5or reliability reasons he used relays for the entire machine!

icture 1! $he design of the Z3 by Konrad Zuse in 1939! $here does not e/ist aphoto of the Z3 because all documents and photos were destroyed by alliedairraids during the war!

Konrad Zuse rebuilt the Z3 in 196,61 in his Zuse K- in order to show thepatent ?ustice and the world the performance of this machine and todemonstrate this machine for advertising purposes!

icture 2! 0eft< $he rebuilt computer Z3 in 196* with Konrad Zuse! $hememory is on the left side ($he logo Z of the Zuse K- can be seen and thearithmetic unit with the stepwise relays are on the right side! 0eft on the front theconsole with the punch tape reader can be seen! Iight< $he diary of KonradZuse at #ay 1=+ 19*1< Konrad Zuse presented the wor'ing Z3 to scientists inerlin!

$oday+ it is undisputed+ that the Z3 was the first freely programmable+ based ona binary floating number and switching system+ really wor'ing computer of the



world! $he Z3 did not store the program in the memory+ but it contained in 19*1almost all the components of a modern computer as reCuired by Mohn von.eumann et al! in 19*6 @IK*6! >ne e/ception is the ability to store theprogram in the memory together with the data! Konrad Zuse did not implementthis feature in the Z3 because the memory of 6* words was too small for this!

4e wanted to calculate thousands of instructions in a meaningful order! $hemachine he constructed should be freely programmable! 5or this reason+ heonly used the memory for values or numbers!





icture 3! $he bloc' structure of the Z3 and the rebuilt machine Z3!

$he bloc' structure of the Z3 is very similar to a modern computer! $he Z3disposed of separate units+ li'e punch tape reader+ control unit+ floating pointarithmetic unit and input) output devices! Fith respect to this theme+ ur's et

al! @IK*6 wrote as late as in 19*6< "n as much as the completed device #illbe a general!purpose computing machine it should contain main organs relatingto arithmetic, memor$!storage, control and connection #ith the human operator."t is intended that the machine be full$ automatic in character, i.e. independentof the human operator after the computation starts. t is important to notice+ thatthe Z3 contained organs li'e an arithmetic unit+ memory)storage+ control andconnection with the human operator! $he machine Z3 contained a specialoperating modus! Fith the instruction %u the input device was activated and theprogram was stopped! $he human operator could chec'+ among others+ theIegisters I1 and I= of the arithmetic unit and he could ma'e intermediatecalculations using the Iegisters I1 and I=! $hen he could proceed the

program! $his is a very early interactive communication with the machine!

0et us consider the discussion of floating point numbers in the paper of ur's etal! @IK*6< "t #ould therefore seem to us not at all clear #hether the modestadvantages of a floating binar$ point offset the loss of memor$ capacit$ and theincreased comple&it$ of the arithmetic and control circuits. n .2@#* the useof the binary system for computers is proposed! t is very important to notice+that floating point numbers were discussed as late as in 19*6 by ur's et al!and they were not really convinced that this is a good concept! Konrad Zusedescribed in 193* the concept of a floating processor and implemented in 1936such a floating point processor! $his is ten years before ur's et al! @IK*6postulated it as a concept for modern computers (not a built machine! "sshown above and will be shown below+ Konrad Zuse also implemented apowerful arithmetic e/ception unit! 0et us consider the logical concepts of theZ3 in detail!

arallel Machine! $he Z3 was a parallel wor'ing machine! $he == bits from thememory to the Iegister I1 and vice versa were moved in one step (cycle! $hesame holds for the binary arithmetic unit+ where+ among others+ two paralleladders (e/ponent+ mantissa were used!

Memory! $he memory of the Z3 consisted of 6* words R == bits! 2ach wordwas directly addressable by the instructions r 7 or s 7+ where 7 is the address

in the range< 6* ≤ 7 ≤ 1! 5or each bit a relay was needed!

Floating oint "umbers! Konrad Zuse used a semi)logarithm representationof numbers! 0et us ta'e the number 1,,! n the binary system this number canbe written as<

1,, S 1/=6T 1/= T ,/=* T ,/=3T 1/== T ,/=1 T ,/=,S 6* T 3= T *!

$he number 1,, can be created by the power of = related to the numbers< =6+ =

and == S 6* T 3= T *! n the binary representation we use the 1 or , in order torepresent the number! ($hese are e/actly the factors 1 and , before the



multiplication sign! n the binary digit system the number 1,, can be written as<11,,1,,! $his is a binary number without a comma! $he number 1,,+ can berepresented as<

1,,+ S 1/=6T 1/= T ,/=* T ,/=3T 1/== T ,/=1 T ,/=, T 1/=)1S 6* T 3= T * T ,+!

n the binary system we write this as< 11,,1,,+1! $he most right 1 of thecomma is =)1S ,+! $hese are binary numbers with a comma+ we also say fi/edpoint numbers! $o represent numbers in the , or 1 notation has big advantagesfor computers! $his has not changed till today! ince only two states or twonumbers are considered+ we can use pair circuits to store numbers and tooperate with them! n 1939 relays were such building bloc's! 4owever+ theabove representation of numbers has a disadvantage! Konrad Zuse used ==bits for his number+ but with == bits we can only represent small numbers in a

range ≤ ===)1! 5or this reason Konrad Zuse used the binary floating point

representation (semi)logarithmic representation for the numbers! 0et us

consider the number 1,, again!

1,, S 1/=6T 1/= T ,/=* T ,/=3T 1/== T ,/=1 T ,/=,S 6* T 3= T *!

Konrad Zuse divided the number in an e/ponent and a mantissa! "s thee/ponent he used the power of two of the highest e/ponent+ in our case 6! $hisnumber was converted into the binary system and we got 11,! "s the mantissawe used the binary number 11,,1,,! $he new number U a floating pointnumber U is now<

,,,,,11, 11,,1,,,,,,,,,,!

$he first digit of the mantissa always is 1! 5or this reason the first 1 is left outand the mantissa is 1,,1,, instead of 11,,1,,! Doing this+ Konrad Zuse didwin one digit for a higher accuracy of his numbers! nstead of 1* bits he had 1bits for the mantissa! $he number 1,, is now<

,,,,,11, 1,,1,,,,,,,,,,!

$he number 1,,! in the binary system is< 11,,1,,+1! $he floating pointnumber is<

,,,,,11, 11,,1,,1,,,,,,,!

n the Z3 the number is<

,,,,,11, 1,,1,,1,,,,,,,!

@sing this concept+ Konrad Zuse was able to handle very big and very small

numbers in a range of appro/imately ± =)63 to ± =63!



icture #! 5loating point numbers of the Z3! &Z means the sign for themantissa+ 2O to 21 are the bits of the e/ponent and the numbers 1* to 1describe the 1* bits of the mantissa! $he picture is from the Konrad Zuse#ultimedia how Z@298!

$nstructions! $he Z3 disposed of the following instruction set and cycles! Feintroduce the following notation< I1 and I= are Iegisters R == bits! $henotation< I1 <S I1 T I= means< $he contents of Iegister I1 is added to thecontents of Iegister I= and the result is stored in Iegister I1! Iegister I= isset to empty after arithmetic the arithmetic operation!

$nstruction Function %ycles

r 7 $o read the contents of the memory cell 7 into the Iegisters I1 or I=! 1

s 7 $o write the contents of Iegister I1 in the memory cell 7! ,)1

0s1 "ddition< I1 <S I1 T I= 3

0s= ubtraction< I1 <S I1 T I= *)

0m #ultiplication< I1 <S I1 T I= 16

0i Division< I1 <S I1 T I= 18

0w Cuare root< I1 <S VI(I1 =,

0u $o call the input device for decimal numbers! 9)*1

0d $o call the output device for decimal numbers! 9)*1

$he two basic arithmetic operations of the Z3 are the addition and substractionof the e/ponent and signs! "ddition and subtraction reCuire more than one cyclebecause+ in the case of floating point numbers+ care has to be ta'en to set the

si7e of the e/ponent of both arguments to the same value! $his reCuires somee/tra comparisons and shifting! $he number of cycles needed for the 0u and 0d



instructions is variable+ because it depends on the e/ponent of the argument! "number can be stored in memory (s in 7ero cycles when the result of the lastarithmetic operation can be redirected to the desired memory address! n thiscase+ the cycle needed for the store instruction overlaps the last cycle of thearithmetical operations!

t is a little bit surprising+ that the conditional branch is missing! Konrad Zusedescribed in his papers from 1936 till 19* a lot of scientific and numericalproblems+ which he wanted to solve with his machines! 5or this calculations hedid not need the conditional branch! "t least in 19**+ he mentioned theconditional branch indirectly in the conte/t of free programs Z@2*+ p16!Konrad Zuse introduced 'ree (rograms )'reie *echenpl+ne< $he definition is<-ei den freien *echenpl+nen beeinflussen die eigentlichen ariablen den Ablauf der *echnung. /un+chst k0nnen die bei den 1uasistarren *echenpl+nenbesprochenen (lanvariablen #ie variable 2perationszeichen, 3trukturzeichenus#. 'unktionen der eigentlichen ariablen sein. 4s kann z.-. die Art der

2peration in einer *echengleichung erst errechnet #erden.

$ranslation< 5ith the free programs the actual variables influence the e&ecutionof the calculation. 'irst of all the variables, like operation signs, structure signs,as defined #ith the 1uasi rigid )1uasi starr programs, etc., can be functions ofthe actual variable. "t is, for e&ample, possible to calculate the t$pe of operationin a calculation e1uation. 4owever+ all the other machines at this time did nothave the conditional branch!

&rithmetic 'nit and %arry &head! $he arithmetic unit of the Z3 is KonradZuses masterstro'e! 4e reduced all the arithmetic operations to addition orsubstraction! 5or the reali7ation of the addition (subtraction is an addition of thecomplement of one number and the number Konrad Zuse implemented aspecial switch because he wanted to avoid too many cycles for the addition oftwo binary floating point numbers! @sing the special switch+ he could reduce theaddition from at least 1* cycles with a serial addition down to three cycles with aparallel addition! "lthough there were only five instructions (0s1+ 0s=+ 0m+0i+ and0w for arithmetic operations+ some more operations were implemented whichcould be called from the input device! 4e also simplified the e/ecution of thearithmetic operations with micro)seCuences controlled by stepwise relays!

icture (! 0eft< $he arithmetic unit of the Z3! t could be used from the programbut also for intermediate operations from the input device! Iight< " relay of theZ3!



Konrad Zuse used a self developed carr$ look!ahead circuit of relays for theaddition of floating point numbers! Fith this concept he could add two floatingpoint numbers in three cycles! "ssuming the numbers to be added were still inthe memory+ then he needed five cycles+ two for the instructions r and threefor the instruction 0s1! Fith the first cycle+ the first operand was transferred to

Iegister I1 and with the second cycle+ the second operand was transferred toIegister I=! $hen+ the e/ponent and mantissa were added and the result wasstored in Iegister I1! "fter this Iegister I= is set empty! Fe write this as< 0s1<I1 <S I1 T I=! Konrad Zuse used additional internal registers which are notshown on the picture above! $hese registers were two bits longer than themantissa in order to get a higher accuracy of his results!

$he multiplication algorithm of the Z3 is li'e the one used for decimalmultiplication by hand+ that is+ it is based on repeated additions of themultiplicator according to the individual didgits of the multiplicand!

$he division algorithm is similar to the multiplication one+ but the repeatedsubstraction is used!

$he instruction algorithm behind the instruction 0w is a ?ewel of the Z3! $hemain idea is to reduce the sCuare root operation to a division (ee also the#achine Z*!

&rithmetic )*ception +andling! $he Z3 disposed of an arithmetic e/ceptionhandling! Konrad Zuse implemented the e/ception handling because he wantedto be sure that the Z3 calculates the numbers correctly even when the Z3 iswor'ing without a supervisor! $he Z3 recogni7ed the following operations<

• @nderflow of the range of numbers is< ,!

• Fverflow of the ran!e of numbers isC ∞ +

• 7or the calculation with 0 holdsC 0 I N , 0 P N 0, 0 N ∞ , 0 0 N M

• 7or the calculation with ∞ holdsC ∞ I N ∞ , ∞ P N ∞ , 1 ∞ N 0+

• ll operations with M have the result M+

• 00 N undefined

• ∞ -∞ N undefined

• ∞ ∞ N undefined

• 0 ∞ N undefined

$he undefined state was shown on the output device on the left side with small

lights! 5or the numbers , and ∞ Konrad Zuse used special bit codes in the

e/ponent! "n e/ponent of U6* is the decimal ,! "n e/ponent of U63 or T63

represents ± ∞ ! $he Z3 calculates always correctly+ if an argument is , or ∞

and the other argument is in the allowed range!



icture ,! 0eft< $he input) and output devices of the Z3! "t the front thenumbers could be put in by buttons! $here were four buttons for the mantissaand 1O buttons for the e/ponent (from )8 to T8! $he results were shown bylamps! Iight< $he output device of the Z3 with the lamps for the decimalnumbers (right and the arithmetic e/ception handling on the left side!$he binary floating point numbers were converted to decimal floating numbers!5or this conversions he needed between 9 and *1 cycles depending on thee/ponent! $he mantissa consisted of four decimal digits (five digits for the 1and the e/ponent was between U8 and T8! $he biggest decimal number whichcould be shown was 1999921,8!%loc-ed Maschine . %loc- Fre/uency! $he Z3 is a cloc'ed machine! Konrad

Zuse used this principle to synchroni7e the different components of themachine! n order to do this he implemented a special impuls generator with adrum!



icture 0! $he impuls generator for the Z3! $he speed of the capstan could becontrolled in steps! t is an electric motor which drives a shaft+ upon which areattached a number of arms (or protruding levers+ where each arm is used toclose a switch+ and the angular separation between the arms caused differentswitches to be closed at different times+ thereby allowing the system to controlthe flow of data between the various units!$he impuls generator stepwise provides the different relay groups with electricalpower! $he cloc' freCuency was around !3 4ert7 depending on the Cuality ofthe telephone relays+ so the faster the cloc' then the faster the machine! n tal'swith Io?as I>M"98a and me Konrad Zuse mentioned that a multiplication of

two floating point numbers too' about three seconds! $he Z3 needed 16 cyclesfor a multiplication! 5or this reason the cloc' freCuency is 16 3 ≈ !3 4ert7! n

order to avoid spar'ing of the relay contacts when switching them+ he used thedrum to drop off the electricity at the switching time! $his principle avoidedmaterial rust and guaranteed a long functionality of the relay contacts!Micro e/uencer and ipelining! $he heart of the control unit of the Z3 arethe micro seCuencers! 2very cycle in the Z3 was divided in five steps to &!

icture ! $he e/ecution pipeline of the Z3!$he steps & and & are used to transfer information from one part of the Z3 toanother one+ for e/ample the contents of a register to a memory cell! During thesteps and an addition or subtraction in both parts of the arithmetic unit ise/ecuted! $his is the e/ecution phase of an instruction! " typical instructionta'es the arguments of the operation+ e/ecutes an operation and writes theresult bac'! Konrad Zuse considered it as very important to save e/ecutiontime+ that he e/ecuted the instruction in an overlapping way (ee the picture! tis 'ind of a pipelining principle!t is not well 'nown+ that Konrad Zuse in 19*9 applied for a patent for apipelining computer in 19*9!



icture ! $he first pipeline computer of the world! Konrad Zuse applied for a

patent for it at Mune 3,+ 19*9 and got it at December 6+ 196=!! "t Mune 3,+ 19*9 Konrad Zuse applied for a patent of the first pipeline computerof the world and got it as late as on December 6+ 196=! $he title of the patent is<*echenmaschine zur Durchführung arithmetischer *echenoperationen )z.-.Multiplikationen. $ranslation< Calculator for the e&ecution of arithmeticoperations, for e&ample multiplication. t was a machine consisting of eightindependent arithmetic units which could e/ecute partial arithmetic operationson certain data! Fith this machine it was possible to e/ecute arithmeticoperations for a large set of data!Fe now are coming bac' to the micro)seCuencers of the Z3!

icture 14! 0eft< $he micro)seCuencers of the Z3 reali7ed with stepwise relays!Iight< " stepwise relay of the Z3!Fith stepwise relays Konrad Zuse implemented so)called micro seCuencers!$he seCuencers were part of the control unit and were used to enable aseCuence of repeated micro)instructions! 5or e/ample+ the instruction 0m isreali7ed by a seCuence of additions! n the Z3 we can find micro)seCuences forthe instructions 0w+ 0m+ 0i+ 0u and 0d! t is not micro)programming+ but it is ahardware wired micro)seCuencing!

&n )*ample rogram 5olynomial6! Fe now present an e/ample program ofthe Z3! Fe want to calculate the polynomial< ((a*/ T a3/ T a=/ T a1! $he



numbers for a*+a3+a= and a1 are already stored in the memory cells *+ 3+ =+ 1+which had to be done by the input device! $he number for / is reCuired by theinput device with the instruction 0u!Lu To call the input device for the variable x.

Ps 5 To store variable x in memory cell 5.

Pr 4 Load a1 in Register R1.

Pr 5 Load x in Register R2.

Lm ultiply! R1 !" R1 x R2.

Pr # Load a# in Register R2

Ls1 $dd! R1 !" R1 % R2.

Pr 5 Load x in R2.


Pr 2 Load a2 in Register R2.

Ls1 $dd! R1 !" R1 % R2.

Pr 5 Load x in Register R2.


Ppr 1 Load a1 in Register R2.

Ls1 $dd! R1 !" R1 % R2.

Ld &ho's the result as a decimal number.

"fter the e/ecution of the program the Z3 stopped and showed the result via theinstruction 0d!Minimal Design rinciple and 'ni7ersal %omputer 5Turing6! Konrad Zusefollowed a minimal design principle with the construction of his computers! 4etried to build a powerful computing machine with minimal effort and costs! 4ehad no other way because his parents were not rich and he had not muchmoney! 5or Konrad Zuse+ the Z3 was the last machine in a series of trialmachines (Z1)Z3 in order to build a machine which was able to solve themathematical problems of engineers and scientists! "lthough he wasconvinced+ that his machine could calculate all mathematical problems+ hecould not prove it! 4e tried to show this with many e/amples and chess! n 19*1he said to friends+ that his machine can play chess! 4owever+ he had the rightfeeling+ that his machine could calculate everything! 5rom 19*1 till his death in199 he did not 'now+ that his computer Z3 was an universal computer in thesense of $uring! n 1998 Iaul Io?as I>M"98a formulated the proof+ that theZ3 is an universal computer!

The 1

Data sheet

#ame of the machine 51

<mplementation (lectromechanical

7reHuenc 1-8 *ert$

#umeric unit 7i point

vera!e calculationspeed

@ultiplication appro+ 8 5ec+



<nput ecimal %eboard

Futput rra of lamps


#umber of relas 00

#umber of step

switches

ppro+ 10 for hard-wired pro!ram control

@emor M

.ower consumption M

Wei!ht M

rea of application 5pecial application for win! calculations

#umber of machines

sold

0

3ost in (@ #one3omments *ard-wired pro!ram with step switches+ )he 51 was used from

19'2 to 19'' at *enschel-7lu!$eu!wer%e+

zusecs.tu!berlin.de

The 2

Data sheet

#ame of the

machine

52

<mplementation (lectromechanical

7reHuenc 1-8 *ert$

#umeric unit 7i point

vera!e calculation

speed

@ultiplication appro+ 8 sec+



<nput ecimal %eboard

Futput rra of lamps


#umber of relas 00

#umber of step

switches

ppro+ 10 for hard-wired pro!ram control

@emor M

.ower consumption M

Wei!ht M

rea of application 5pecial application for win! calculations

#umber of

machines sold

0

3ost in (@ #one3omments *ard-wired pro!ram with step switches+ 3onnected to measurement

dials with analo! to di!ital conversion+ )he first process control

com puter+

zusecs.tu!berlin.de

The Z#

$he Z*+ the second general purpose computer+ was completed in 19**! $he Z*was reassembled in the years following 19*! unched tape readers were thenadded! 5rom Muly 11+ 19,+ this configuration was used for five years at thenstitute of "pplied #athematics at 2$4 Zurich! n 191+ the Z* was the onlyoperational computer in 2urope! n 19*+ the Z* was transferred to the "nstitut'ranco!Allemand des *echerches de 3t. %ouis in 5rance+ where it was in useuntil 199! $oday+ the Z* is on display in the Deutsche #useum in #unich!

$he Z*!





#emory unit of the Z*!$his picture was 'indly made available by Dr! Kistermann!t was ta'en at an e/hibition at 2$4 Zurich in 1981!

Z# 8 lanfertigungsger9t

$he Z* had a (lanfertigungsger+t ! t was a unit in order to produce punch tapesas a program with instructions for the Z* in a very easy way! 5or this reason itwas possible to learn the programming of the Z* in at least three hours! $helanfertigungsgerEt already was a part of the Z* in 19*!

$he Z* (lanfertigungsger+t !t was possible to use symbolic memory cells+ instructions and symbolicarithmetic operations for the creation of a program! t also was possible to copyprograms and to ma'e corrections!

erformance of the Z#

5rom 19,)19 the Z* processed appro/! 1,,different pro?ects! $he Z* e/ecuted appro/!

1,,+,,, instructions! 5or e/tern pro?ects the costswere one Iappen (,!,1 ;45 per instruction!





$nstruction et of the Z#

$he Z* had a large instruction set in order to calculate complicated scientificprograms! $he instruction set of the Z* was formulated in 19*=! $he Z* had aperformance of 1,,, instructions per hour!

• <nstruction " nC 7or eample " 1O+ <t reads the contents of memor cell 1 into

>e!ister I1+ <f >e!ister I1 is occupied, the contents is loaded into >e!ister I=+

• <nstruction nC 7or eample 18+ <t writes the contents of >e!isters I1 into the

memor cell 1=+

• s ar!uments for the arithmetic operations the >e!isters I1 and I= were used+

7or monadic operations the >e!ister I1 is used and the result occurs in >e!ister

I1, too+ 7or dadic operations both >e!isters I1 and I= are used and the result

occurs in >e!ister I1+ )he contents of >e!ister I= is deleted+

• @onadic operations areC T ) W #"M (#a/imum #in+

• adic operations areC /X= VI(/ 1/ / sgn(/ /Y /Y= /Y()1 /Y1, /Y3 /Y13 /Y1

/Y1O /Ypi /Y1pi+• <nstructions for comparison 4N0, QN0, << N infinit6 are testin! the number in

>e!ister I1 and set >e!ister I1 to I1 if the condition is fulfilled, if not, then the

contents of >e!ister I1 is -1+ )his instruction was planned in 19'2 but not

reali$ed because of the war+

• )he conditional branch I was a special reHuirement b the ()*+ )he

instruction I can be used after a test instruction+ <t s%ips the punch tape to the

instruction $, if the contents of >e!ister I1 is I1+ <f the contents of the re!ister

is -1 then there is no impact+

• <nstruction @C )he "' had two punch tape readers+ <n the ori!inal version up to

such readers were planned+ )he instruction @ switches from the main punch

tape reader 406 to the sub punch tape reader 4t16+ )he instruction 5. causes aswitch bac% to 0+

• <nstructions for FutputC )[ D 0 etc+C )hese instructions cause the conversion of

binar numbers to decimal numbers with li!hts, @(>3((5 tpewriter of the

punchtape+

• <nstructions for <nputC \) "t1 etc+C <t allows to read numbers from the punch tape+

0i'e the Z3+ the Z* has an arithmetic e/ception handling! f there are numbersoutside the range of 1,X)=, \ / \ 1,X=, then the machine gives the rangewhere the result is+ for e/ample<

• Very big + very big = very big.

• Very big - very big = undefined.

• ! = undefined.

f an undefined value is combined with a number+ the result is again undefined!@sing this method it can be avoided+ that the Z* calculates wrong numberswhen it is wor'ing without attendance! $his was often the case at the 2$4!

Data sheet

#ame of the

machine

"'



<mplementation ppro+ 2200 relas #umeric unit, mechanical memor unit

7reHuenc &0 *ert$

#umeric unit 7loatin! point unit, &2 Bit Word len!th, 28-&0 Fperations per sec+

vera!e

calculation speed

ppro+ 11 @ultiplications per 5ec+

<nput ecimal %eboard or punch tape

Futput @ercedes tpe writer

Word len!th &2 Bit, floatin! point, Bit eponent, 1 si!n, 2' Bit mantissa

#umber of relas 2200

#umber of step

switches

ppro+ 21 4micro seHuences6

@emor @echanical memor, metal plates, ' words L &2 Bit, 800 words

planned

.ower

consumption

ppro+ '000 watts

Wei!ht ppro+ 1000 %!

rea of

application

1980-1988 ()* "urich

#umber of

machines sold

1 rented to ()* "urich

3ost '0,000 3*7

3omments )he "' was similar to the "&, however improved in terms of speed+ <talso implemented unconditional branchin!+ )he "' was wor%in! da

and ni!ht at ()* "urich+ )he ori!inal machine is on displa in the

eutsche @useum in @unich+

)he computer "' at the ()*-"/rich - report b + 5peiser

zusecs.tu!berlin.de

Zuse &pparatebau Berlin 51#481#(6and Zuse K: 51#81,6

$ntroduction



0ogo of the Zuse K-$his is a short story of the two companies Zuse "pparatebau erlin (19*1)19* and the Zuse K- (19*9)1969! Konrad Zuse had a vision! 4e wanted tobuild computers in order to release engineers from their hard wor' of calculatingthousands of numbers! n 19*3 he formulated this vision and calculated theprice of one word for a memory of thin metal sheets up to I#! $he si7e ofsuch a memory should be one thousand words a 3= bits! $he ;omputer Z* wasplanned as the prototype of a serial machine for engineers! 4owever+ the wardestroyed Konrad Zuses visions!

Zuse &pparatebau Berlin 51#481#(6

$he /use Apparatebau in erlin were founded in erlin in 19*1 in the#ethfesselstraQe O and 1,! n 19* when the company was destroyed by alliedairraids+ the company had appro/! =, employees! $he plot of land in the#ethfesselstraQe O is till today a bombed site!

0eft< #ethfesselstraQe 1, in erlin (ca! 199,! n a house on this plot of land theZ3 and Z* were constructed! $oday+ there is a vinyard! Iight< Konrad Zuse atthe entrance of the #ethfesselstraQe O in 1989+ where he built the famous Z3

and started the development of the Z*!


http://www.inf.ethz.ch/news/speiser.html




Zuse &pparatebau Berlin 51#481#(6and Zuse K: 51#81,6

%omputers from 13, to 1#(

n the #ethfesselstraQe O and 1, and from 19** in the >ranienstraQe 6+ KonradZuse built the following computers<

• "1 - thin metal sheets 4.rivatel financed6

• "2 - relas and thin metal sheets 4.rivatel financed6

• "& - relas 45upport b the !overnment6,

• 51 - relas 45upport b the !overnment6,

• 52 - relas 45upport b the !overnment6,

• "' - relas, memor consistin! of thin plates 45upport b the !overnment6+

n 19* the Zuse "pparatebau were destroyed by allied airraids and only the#achine Z* survived the attac's by the allied airraids!

randenburger $or (randenburg -ate in 19*!Konrad Zuse fled from erlin in #arch 19* and arrived in 4interstein ("llgEu+avaria in "pril 19*! $here+ Konrad Zuse met Fernher von raun in the 4ausngeborg! t was the first and last time in his life that he met Fernher von raun!4e left the group of Fernher von raun and went to 4interstein and hided partsof the Z* in a barn of the Alpenhotel 3teinadler ! $he 5rench and "mericantroops loo'ed for the Z*+ but they did not find it!

n this barn of the 4otel teinadler in 4interstein ("llgEu+ avaria the Z* washidden! "t this time the Z* was the only one computer in 2urope!








5rom 19* till 19*O it was almost impossible to restore the not finished andslightly damaged Z*! "t this time+ the Z* was the only one computer in 2urope!Konrad Zuse used tins from the allies to repair parts of the mechanical memoryof the Z*! $he first problem was to survive the years after the war! n order toget some food+ Konrad Zuse made woodcuts and sold them to the farmers and

the "merican troops! $he ne/t picture (D. "* shows a woodcut by KonradZuse+ which shows 4interstein ("llgEu+ avaria! $he ;omputer Z* was hiddenclose to the church!

" woodcut of Konrad Zuse of 4interstein ("llgEu! 6he image above is linked toan enlargement )789 :-!

$he following picture is a color woodcut of the ZipfelsfElle (Zipfel waterfall in4interstein (D. "3 by Konrad Zuse! 4e used three separate wood plates+ onefor the yellow+ one for blue and one for red color! 2ach wood plate was pressedon the same paper!

$he waterfall ZipfelsfElle close to 4interstein! t is a three color woodcut (*, cm/ 6, cm by Konrad Zuse (19*!

n a paper from 19*O Konrad Zuse wrote the following statement related tocomputers in the future<

Die in diese" #usa""en$ang auftretenden %rob&e"e sind so

u"fangreic$' da( eine )eneration von *issensc$aft&ern' ec$ni,ern und

*irtsc$aft&ern erforder&ic$ sein wird' u" sie ersc$pfend u bearbeiten.






http://irb.cs.tu-berlin.de/~zuse/Konrad_Zuse/en/Rechner_S1.html

http://irb.cs.tu-berlin.de/~zuse/Konrad_Zuse/en/Rechner_S2.html









$ranslation<

$e prob&e"s in t$is context are so extensive' t$at a generation of

scientists' engineers and econo"ists wi&& be necessary' to process t$e"

sufficient&y.

Fith problems Konrad Zuse meant the introduction of computer s in science andcompanies! 4is plan was to build logistic computers! $hese computers weree/tended algebraic computers+ li'e the Z*+ in order to solve problems of theengineers! $hese problems should be formulated in the lan'al'Jl+ which todayis accepted as the first programming language of the world! "lready in 19*3 hewrote+ that from his view the problems of algebraic computers are solved!

$he wor'shop of Konrad Zuse in 4opferau ("llgEu in 19*O+ where he installedthe Z*!

n 19*8 the Z* was build up in a former stable for horses in 4opferau ("llgEu+avaria! ome calculations could be done+ but it was very difficult to 'eep themachine wor'ing! 2lectrical power was only available for some hours a day andmostly it was not sufficient for the Z*! n 19*8 rof! tiefel from the 2$4(2idgenLssische $echnische 4ochschule ZJrich visited Konrad Zuse in

4opferau and wanted to see the Z*! $ill today it is not clear from whom heheard about this #achine Z* in 4opferau!



http://irb.cs.tu-berlin.de/~zuse/Konrad_Zuse/img/ZKG_Artikel_6_large.gif













$he input and output device of the Z* in 19*O in 4opferau!rof! tiefel gave Konrad Zuse a differential eCuation as a problem to solve!Konrad Zuse formulated the program for the Z*+ created a punch tape and theZ* began to wor'! $he result was correct! 0ater+ a contract was made betweenthe 2$4 and the Zuse K-+ although rof! tiefel visited the machines 2.";+#"IK and #"IK in the @"! $he ma?or reasons to lease the Z* for yearswere<

• <t was the onl one available computer in (urope+

• <t was a reliable rela computer althou!h there was a bad feelin! about the

mechanical memor of the "'+ )he reliabilit of relas was well %nown+

*owever, a memor such as of the "' nobod had seen at an time before+• )he .lanferti!un!teil, which was a feature to write pro!rams in a ver eas wa+

• onrad "use promised to ma%e some etensions with the "'+

• )he 5 computers were not available+

n order to f ulfil the contract with the 2$4+ Konrad Zuse founded the Zuse K-!

Zuse K: 51#81,6

n 19*9 Konrad Zuse founded the Zuse K- with five employees in .eu'irchen(Kreis 4Jnfeld+ 1=, 'm north of 5ran'furt! $he building in .eu'irchen (Kreis4Jnfeld was formerly a trading post!










n this building the first computer company in -ermany was founded by KonradZuse! t was an old trading post! $here the Z* was restored and the Z+ Z11

and Z== were constructed there!$he first tas' of the Zuse K- in 19*9 was to restore the ;omputer Z*+ whichsurvived the air attac's of the allies in erlin and on the way from erlin to4interstein ("llgEu! n 19*9, the Z* was e/tended by the following featuresand delivered to ZJrich on Muly 11+ 19,<

• 3onditional branch+

• <nstructions for printin! the results on a @ercedes tpewriter+

• 5torin! numbers on the punch tape in order to transfer them into the memor of

the "'+

• >estorin! the furniture+

• 5eparation of the service unit and the rela cupboards+

$he restored Z* in .eu'irchen Kreis 4Jnfeld in 19,! $he Z* consisted of ca!ten relay cupboards with ==,, relays! $he memory was a mechanical one with6* words a 3= bits! $he processor was a very powerful floating processor! 6he

image above is linked to an enlargement );89 :-!$he Z* was a big machine with ==,, relays and a memory of 6* words ] 3= bitsconsisting of metal sheets!












Zuse wor'ing at the Z* (19,!Iestoring the Z* charged the Zuse K- with appro/! 6,+,,, D2#+ but the 2$4

ZJrich paid for the Z* an amount *,+,,, ;45! Fith this money it was possibleto found the Zuse K- and to restore the Z*! $he Z* was a great success for the2$4 and the Zuse K-! $he speed of the Z* was about 1,,, instructions perhour!n 19,+ the 0eit7 ;ompany in Fet7lar ordered a big relay computer+ the Z!$he picture shows the employees of the Zuse K- in 19=+ who built the Z! $heZ also was a great success for the Zuse K-! $he Z was sold for 3,,+,,,D2#!

$he Z was an e/tended version of the Z*! Konrad Zuse used modern relays inorder to overcome the bad reliability of tubes! $he Z was the first commerial

computer in 2urope!

4owever+ the Z was the only one big mandate+ that the Zuse K- got between19*9 and 19! "f ter the second world war the situation in -ermany and2urope was very difficult! $he companies had no money to buy computers!-ermany was destroyed up to 6,^! $he capacity of the -erman economy wasless than =,^ compared to 1939! 5or this reason+ Konrad Zuse tried to ma'esome business with @ companies! 4e successfully made some business withthe Iemington Iand ;orporation! $ill 19+ in -ermany was no support by thegovernment for buying or building computers in a company! "n e/ception wasthe construction of the -1 in -Lttingen+ which was a research computer! 5rom19 the D5- (Deutsche 5orschungsgemeinschaft gave money to installcomputers in the universities! $he computer Z11+ a relay computer+ was sold

from 19 to -erman companies+ mostly to the optical industry and universities!n 19O the universities started to order computers of the type Z==! $his was a





http://irb.cs.tu-berlin.de/~zuse/Konrad_Zuse/img/ZKG_Artikel_11_large.jpg























very fle/ible and cheap computer based on tubes with a drum of >K words! "fter 19O many other computers followed+ for that see Konrad Zusecomputers!2arly in 19 Konrad Zuse began with the development of the -raphomat Z6*!t was a drawing machine with an accuracy of ,!, mm! $his machine was

controlled by a program created by the ;omputers Z==+ Z=3+ Z=+ or Z31!

$he Zuse -raphomat 9, in 196*! 6he image above is linked to an enlarg ement)<99 :-!

5rom 19*9 to 1969+ the Zuse K- produced more than =, computers with avalue of 1,, #illion D#! "lthough in 196* the Zuse K- had orders of more than= #illion D#+ it was for Konrad Zuse impossible to 'eep the Zuse K- as hisown company! $he ne/t picture shows the Zuse K- (1=,, employees in 196*in ad 4ersfeld!

$he Zuse K- in ad 4ersfeld in 196*! 6he image above is linked to anenlargement )=9 :-! "fter 1961 the Zuse K- got more or less in financial trouble! $here were manyreasons for that<

• )here was no financial support of the computin! industr b the !overnment+

• )here were an increasin! competition b <B@ and other computer manufactures+

• )he patent situation of onrad "uses computers were not solved at this time+

decision of onrad "use patents were made in 19+

• @ismana!ement+

• )oo fast !rowin! of the compan+

• 5oftware unbundlin!+ )he development of sof tware char!ed the compan more

and more, but the customer did not want to pa man for that+• #o financial spares because computers were the onl production sub?ect+
























• 5oftware production costs+

Zuse K- advertising slogan since 196,!n 196* the Zuse K- was bought by Iheinstahl+ then by ; und finally in196O completely by the iemens company! n 1969 the logo Zuse K- wasdeleted! Konrad Zuse left in 1969 the Zuse K-!5rom my view+ Konrad Zuse is not a collapsed entrepreneur! t is easier to

critici7e people than to ma'e it better! Konrad Zuse was years ahead with hiscompany+ computers and ideas related to computers!

More than 244 machines deli7eredby Zuse K: 5around 1,#6

















The Z31

"reas of application< an's+ insurance companies+ research+ industry+ trade+public service! ;ompilers for -erman formula code and "0->0 6, are availablefor the Z31!








Documents

The History of the Atomic Bomb