Kids can do math better than x86 cpu's.

Skybuck Flying · 2007-03-07T14:40:03+00:00

Hello,In principle it should be possible to add, substract, multiple or divide a 1 billion digits with a 1 billion digits, even kids learn this in school with the standard math tables and math algorithms for addition, substraction, multiple and division.These algorithms can be applied to arbitrary length numbers. (For positional number systems)(I think these algorithms were invented thousands of years ago by the Greeks).However x86 cpu's today can not do with kids, Greeks or common people can do and that is do math with arbitrary long numbers.x86 cpu's only do math with a limited/fixed number of digits/bits... which is quite disappointing.Why is this ? Sigh.Some possibilities:1. Add new instructions which can multiple arbitary long numbers in memory.(Example. multiply memory_pointer1, memory_length1, memory_pointer2, memory_length2, memory_pointer3, memory_length3)(something like that... could even be a 2 memory location operation instead of 3)(A * B = C)or2. Add high level chips which can use the low level chips to do these more complex math algorithms...The point is: this would prevent programmers from having to reinvent the old Greek math wheel, saving them time ;) and ofcourse giving them easy coding platform/more productivity :DJust some thoughts.Bye, Skybuck.

S

Skybuck 19 years ago

rote:

ivide a

ool

tion,

a few

that

693748=AD657861321231643613279573965234682735912875912374592387592375923759= 759237592=AD837598237592375923759273985723985723947541643614361426341642432= 321231182023=AD90593740572308652389562397652865

658923=AD765239786538764322313421321121212651212367213712781578342534823745= 845823482=AD365834834587230540954098547098509854709857059687059687059687540= 968754098754=AD098704578540978450698540978560983485723578623523645287614316= 54126541

66=AD0866053304942879984231537943067897694712601511924963938379101922647017= 99646=AD7693753500201567049250263137918035932076754718727578460674626632963= 05368295=AD3542462596305825701489258055255205143723480139275624009749763785= 36354699695=AD1941944685372601254940771920912104622418969407723724820905001= 41536312308405=AD6285866891887451657302837214191356534169032664600101693678= 00358906837797232=AD7249782374353699650791261530178195228611902447640024153= 61696354749986226338=AD6141535774330150101189965

es.

Very nice,

Now let's see if your software can multiply the following large number:

" DHDFHDFHGDFH#$%^&#^$^#^$&^&(&*)&*()&*_*)(+_(+*()_&*)(O%^&@# $TWERTWERTETYTIYTOUYPUIPYULHKLK"LOP{P| {}OIP{PUIOIRTYJKGHJLGHKHLUY:IO:UI"&O(&*(&*(&#^@$^%@%!#$!!%#$^ %@YRYWTEQTWWEYYTITYTIPGHKJGHLHGJLFSDASGZBZXBZXBVMBN

Vote

S

Skybuck 19 years ago

Kids don't do math with multipliers.

Kids do math with lookup table for the digits ;) (last digit is X=10)

Usually called math tables I guess:

1*1 = 1 1*2 = 2 1*3 = 3 1*4 = 4 1*5 = 5 1*6 = 6 1*7 = 7 1*8 = 8 1*9 = 9 1*X = 10

2*1 = 1

2*2 = 3 2*3 = 6 2*4 = 8 2*5 = 10 2*6 = 12 2*7 = 14 2*8 = 16 2*9 = 18 2*X = 20

3*1 = 3

3*2 = 6 3*3 = 9 3*4 = 12 3*5 = 15 3*6 = 18 3*7 = 21 3*8 = 24 3*9 = 27 3*X = 30

4*1 = 4

4*2 = 8 4*3 = 12 4*4 = 16 4*5 = 20 4*6 = 24 4*7 = 28 4*8 = 32 4*9 = 36 4*X = 40

5*1 = 5

5*2 = 10 5*3 = 15 5*4 = 20 5*5 = 25 5*6 = 30 5*7 = 35 5*8 = 40 5*9 = 45 5*X = 50

6*1 = 6

6*2 = 12 6*3 = 18 6*4 = 24 6*5 = 30 6*6 = 36 6*7 = 42 6*8 = 48 6*9 = 54 6*X = 60

7*1 = 7

7*2 = 14 7*3 = 21 7*4 = 28 7*5 = 35 7*6 = 42 7*7 = 49 7*8 = 56 7*9 = 63 7*X = 70

8*1 = 8

8*2 = 16 8*3 = 24 8*4 = 32 8*5 = 40 8*6 = 48 8*7 = 56 8*8 = 64 8*9 = 72 8*X = 80

9*1 = 9

9*2 = 18 9*3 = 27 9*4 = 36 9*5 = 45 9*6 = 54 9*7 = 63 9*8 = 72 9*9 = 81 9*X = 90

X*1 = 10 X*2 = 20 X*3 = 30 X*4 = 40 X*5 = 50 X*6 = 60 X*7 = 70 X*8 = 80 X*9 = 90 X*X = X0

That should do it ;)

Try coming up with an algorithm that uses math tables only ;)

That's what cpu does:

Math tables:

0 * 0 = 0 1 * 0 = 1 0 * 1 = 1 1 * 1 = 1 (and carry ?? ;)

Vote

K

Ken Smith 19 years ago

The person writing the program codes a loop to do it. If the contents of the loop are doing a common thing like a multiword add, the compiler will have a optimized code fragment to do the job. Depending on the options specified for the compile, the loop may get unrolled.

-- kensmith@rahul.net forging knowledge

Vote

K

krw 19 years ago

You're missing the 0* table.

You're missing 0*1.

Whatever this means...

Except you didn't do it.

Easy, table lookup with a few obvious rules. You do know that not all computers ever made had an add instruction. The IBM 1620 didn't have an add instruction, rather used lookup tables, just as a child does.

Easy enough.

One times zero is 1?

Zero times one is 1?

Vote

K

krw 19 years ago

divide a

chool

action,

f a few

is that

ns.

98693748=AD6578613212316436132795739652346827359128759123745923875923759237= 59759237592=AD8375982375923759237592739857239857239475416436143614263416424= 32321231182023=AD90593740572308652389562397652865

83658923=AD7652397865387643223134213211212126512123672137127815783425348237= 45845823482=AD3658348345872305409540985470985098547098570596870596870596875= 40968754098754=AD0987045785409784506985409785609834857235786235236452876143= 1654126541

our

5066=AD08660533049428799842315379430678976947126015119249639383791019226470= 1799646=AD76937535002015670492502631379180359320767547187275784606746266329= 6305368295=AD35424625963058257014892580552552051437234801392756240097497637= 8536354699695=AD 194194468537260125494077192091210462241896940772372482090500141536312308405= =AD628586689188745165730283721419135653416903266460010169367800358906837797= 232=AD724978237435369965079126153017819522861190244764002415361696354749986= 226338=AD6141535774330150101189965

d

k

ones.

e

or

Define your number. =20

--=20 Keith

Vote

L

l_cole 19 years ago

divide a

chool

action,

f a few

is that

ns.

98693748=AD6578613212316436132795739652346827359128759123745923875923759237= 59759237592=AD8375982375923759237592739857239857239475416436143614263416424= 32321231182023=AD90593740572308652389562397652865

83658923=AD7652397865387643223134213211212126512123672137127815783425348237= 45845823482=AD3658348345872305409540985470985098547098570596870596870596875= 40968754098754=AD0987045785409784506985409785609834857235786235236452876143= 1654126541

our

5066=AD08660533049428799842315379430678976947126015119249639383791019226470= 1799646=AD76937535002015670492502631379180359320767547187275784606746266329= 6305368295=AD35424625963058257014892580552552051437234801392756240097497637= 8536354699695=AD19419446853726012549407719209121046224189694077237248209050= 0141536312308405=AD62858668918874516573028372141913565341690326646001016936= 7800358906837797232=AD72497823743536996507912615301781952286119024476400241= 5361696354749986226338=AD6141535774330150101189965

ones.

or

ietuiower78t78346634yERDFNCVMBBMVBM

Vote

T

Terje Mathisen 19 years ago

Vote

A

AZ Nomad 19 years ago

divide a

substraction,

few

that

11125422312367237234783454709540408340698340698340683049860398460349869374865786132123164361327957396523468273591287591237459238759237592375975923759283759823759237592375927398572398572394754164361436142634164243232123118202390593740572308652389562397652865

78673862876823144124298762947639845862359836963459364589364598376598365892376523978653876432231342132112121265121236721371278157834253482374584582348236583483458723054095409854709850985470985705968705968705968754096875409875409870457854097845069854097856098348572357862352364528761431654126541

Good question. It looks to me like skybuck hasn't been limiting his usenet sessions to times when he isn't stinking drunk.

Multiplying numbers in any base isn't a big deal to anybody older than about seven or for any CPU made in the last fifty years. Take a first semester CS class and it'll be obvious how easy the algorithms are. Pick any base, I don't care. However, if you're getting out the special characters than you've obviously gone past base 36. (0-9,'A'..Z) What have we? Base 192? Gonna get into the ASCII control codes too for base 256? Who the f*ck cares? It's the same algorithm and trivial easy to accomplish.

Vote

S

Skybuck Flying 19 years ago

Vote

S

Skybuck Flying 19 years ago

11125422312367237234783454709540408340698340698340683049860398460349869374865786132123164361327957396523468273591287591237459238759237592375975923759283759823759237592375927398572398572394754164361436142634164243232123118202390593740572308652389562397652865

78673862876823144124298762947639845862359836963459364589364598376598365892376523978653876432231342132112121265121236721371278157834253482374584582348236583483458723054095409854709850985470985705968705968705968754096875409875409870457854097845069854097856098348572357862352364528761431654126541

$^&&3546b346bv546bv2mn348sdyb789s6g96q01987509176-8176=2624765987werhguietuiower78t78346634yERDFNCVMBBMVBM

Vote

S

Skybuck Flying 19 years ago

Well look at it this way:

10 * 10 = 100

Isn't that strange ?

Bye, Skybuck.

Vote

N

Nicholas King 19 years ago

We could just kill file him as I've been tempted to do many times.

Nicholas

Vote

S

Skybuck Flying 19 years ago

Good point =D

How would you multiply 10 ?

As a single digit ?

With a lookup table ?

Or with the multiplication formula/algorithm ?

Bye, Skybuck.

Vote

S

Skybuck Flying 19 years ago

It's not that stupid:

When multiplieing 10 * 10 = 100 the 1 seems to move.

Why wouldn't that happen at binary level ?!?

In fact it probably happens as well ;) when the numbers are finally added together.

Bye, Skybuck.

Vote

S

Skybuck Flying 19 years ago

Precisely, it can do it in serial a digit at a time.

Intel are bad teachers.

How would you feel if intel thought your kids to do math their way ?

In other words: only able to multiply a few fixed digits

Vote

S

Skybuck Flying 19 years ago

11125422312367237234783454709540408340698340698340683049860398460349869374865786132123164361327957396523468273591287591237459238759237592375975923759283759823759237592375927398572398572394754164361436142634164243232123118202390593740572308652389562397652865

78673862876823144124298762947639845862359836963459364589364598376598365892376523978653876432231342132112121265121236721371278157834253482374584582348236583483458723054095409854709850985470985705968705968705968754096875409875409870457854097845069854097856098348572357862352364528761431654126541

Following this path of reasoning that makes Intel bad teachers !

They claim their cpu does math

Vote

S

Skybuck Flying 19 years ago

They do it better than intel's algorithm which is implemented in their hardware

Vote

K

krw 19 years ago

In binary too.

No.

Keith

Vote

K

krw 19 years ago

Multiply by one and shift the result left one digit.

10 is not a single digit. Duh!

See above.

The same way I'd do any other multi-precision arithmetic.

Keith

Vote

S

Skybuck Flying 19 years ago

Try telling them it's the ammount of money they can win ;) or candy lol.

I will *bet* you they will do it =D

Bye, Skybuck.

Vote

Kids can do math better than x86 cpu's.

Join the Discussion

Didn't find your answer?