[Floating Point Representation Floating Point Representation Representehion of rood numbers Decined um GC 14-526" 14-626 C.. — Monrrssa the signlicant- digit of FPN Exponent Poser = 2 ry Brve 2 Brow « Nokes . ms @ Rw sy 2se b 1 Jt _ = rg 32 1b BY 12 OS O02 O-12E 0-062 0-02126 Coes L: Decrmn Nuwere to Bawney Conversion Where will be certain Hhiags Wal ould be given iA Your Qwcation. No of bits used for Mankissa 12 bibs No of bits used for fponenls 4 bisQuestion + Represenl + W+75 in Moabing point representation Step 1: Waite dane We memorized Oise eg O32 be BY 1 OS O25 O-l2C 0-0628 0-03126" Scop 2+ Rb_1 on manbers uihich usll_be added bo fulfil qrestion Cequicoment- trro- y \ TVA Fite point Feprecesbelion 28 G32 16 B 4 2 1 6 OF OS O-l2E 09-2626 O-03126° Step 3: Move he decimal pice ath the wy to Hye Qeft ang unite the exponent volue - Step 4: Igsore decierwh and convert the exponent into bisa Gon Yer O1o0 Sey2y ONO looStep 5: Store Ihem 10 given spaces, DOO S) TOoOoOoBAEL oO & MANTISSA SxPonenr + Mole + Alwoys stove mortissa From Left side Alueys store exeonent fiom right sike- Trea Gres Oyesrion 1 Ina particular computer system, real numbers are stored using floating-point representation wit: + 10 bits for the mantissa + Gils for the exponent J + _ two's complement form for both mantissa and exponent, — Mantissa Exponent — O | {olojolojojo}) 6 jolt Jol ojo VARA 1.20.N%O."0..°0...0.9. \ {@) Calculate the normalised floating-point representation of +192.5 in this system. Show your working Rs O32 1b B42. 1.4 OF 0-2 O-l2E 0-O62E O-03126° O\O00000) x 2° owe S424 i) —————_ G@se 2, Neaawe Decco Nuwe Io Brnnay Fonrme Rewe Question ; Convert _- 14-75 to Birwey Foot. 5 Bint Representalls on Stop 1s Igroce gegebive sig and fallow the steps of erse 4 iN step 4OsVory\ C100 on mantissa only and sloce it Stop 2: Exponent would cewain Some and opel tuss's compliment metod O1rory e100 Nertigsa, Sepanent Or1iroit Brnncy Aoortron Ors amlnen 1 © 0 0 100 3 A Q a oO 4000104 ° 1 ° Store. this J t 4 oO n oO. 1 4000 104 1 4 4 HBOOOOBHOBERH IDOE MAnTIssA SxPonene Case 3: Binony To Dewnes Greveesns JTOODHeOOHE a6 HOE MaNTIssa SxPonense Oyi,i } Os1 100000 Pod rt 0100 ey2yOLLLO- Id Sxponent - ¥ ) Move decinee Q 4 places 1fio0-tt BH VA, os ors W765 Cece 4: Neamive Brnasy 10 Denney DORE JHE EERE IOOE MANTIssA SyPonene 4 Exponcat = ¥ attout S 424. OF erg NUTS NA Answer Sccws Sine Qusstion(b) Calculate the normalised floating-point representation of ~192.5 in this system. Show your working. Mantissa Exponent alofo|s [4 [4 [s [et [a [a] O}9 |f | 9/9 [0 Working O11 000099); 901000 Eyponnk:8 Voor htt 10 sos 4 4 foot site (c) The floating-point representation has changed. There are now 12 bits for the mantissa and 4 bits for the exponent as shown. Mantissa Exponent anne [o{1|s [o Negakive Exgotent Explain why +192.5 cannot be accurately represented in this format. =3 + Zepenent:... toe. Qoege, bo! » Exponent. butn, oegelire thece fore... decened porat wove... the. sang, 19895 (3) Noasecazartod Nowrelircisn ig a betinique tuk is weed to make your dake more aqucdleOn Represents positive, number Lo Repeseals nagalive auaber How wit we now /figuie ol that 0 binary cepresentabin is vymebirca! - Fist ant seend bit should never be same - Manirssa Seeomensy BOEOHOoOOOE IO Pewaooooo OOBE Soy na O-LLOssn vo 0100 Segoe = 5 = 4 u Srponeat 100 Ladtel= probleas cul occur if the biewy cepresalation is rot _oomalized ? |. Multiple representation of singe Auntber + Pedsion Qout- Redandest Yeading aias 0 Nentisen. What ace Yhe problems ia Moating Paint «opreseatation 2 2 O-2 for/ou ese numbers can wl be exadly coptesated Ane solution for this problem is Rounding WAG ould couse toundiag ertor 7 + O.2 WS been cepresenttd by velre gradkir tran 0-2+ O-2 WS been cepresenttd by velre gradkir tran 0-2 0-4 has been repesented by value geealer Van O-4 +S alter calcublbiag with thes cumbers he differcece well increase Aad the AMerence vill be saaificaat: Why rounding esror occurs ? 4 Because there ig ao exact repeosentabiva For some bitary tnbers “Tenne. On eswesn Moasrissa Pwo Loveontur Trade off means relabionsti: BOL 7AHOBOaSEEG NOE Mantissa Sapons gr Precision Reawoe 12 biks For enonbisse Bbils for montion. Ness precision uy bits For expoaut —, Slits for exponent better age Questiow + Whol _is jhe }rade- off betwenn Mankissa ond eaporent |The Wadeoff is belwen predsion od ca|The Wadeoff is belwen predsion od ca | WPmore bits ove used for Mantissa that menos better precision TE more bits are used for _expnent thal means oetber eonae Moreno af bis for mantison meang Yess no of bits for exponent: Larner Bernie Nemate € Smatlest: Bsitve Number Laswesr Barre Swunesr Porte Ort ost too 1000 Meat Exponent Moitisso _ Ceponent~ wo (©) (Write the largest postive number that can be stored as a normalised floating-point number using this format. Mantisea Exponent ols] fal [sft ets ae ojala tt (1) Wrte the smallest non-zero positive number that can be stored as a normalised ‘oating-poit number using this format, a Mantisa Exponent ol tfelofojelolefe[ojoe +Tolep | e OvERFLow Uno enrrow Ness space more More space Qe bils bo store bits be tore,