PARI's commands

[science_man_88]

Originally Posted by CRGreathouse

Ah. You should only use one, but pre-set it to the appropriate size, filled with 0s, then *add* rather than concatenate. That way you can go from

[0, 0, 0, 0, 0]
to
[0, 0, 0, 0, 6]
to
[0, 0, 0, 4, 6]
to
[0, 0, 0, 13, 6]
to
[0, 0, 6, 13, 6]
which can be simplified to
[0, 0, 7, 3, 6]
and then
[7, 3, 6]
if desired.

I was thinking of using while loops and another vector, but your way works I think. The problem is if I get 9*9 then it makes it can make it even longer still.

[CRGreathouse]

Originally Posted by science_man_88

I was thinking of using while loops and another vector, but your way works I think. The problem is if I get 9*9 then it makes it can make it even longer still.

If you multiply an n digit number by an m digit number, n + m digits will suffice (though the most significant digit may be 0).

[science_man_88]

Originally Posted by CRGreathouse

If you multiply an n digit number by an m digit number, n + m digits will suffice (though the most significant digit may be 0).

Okay I think I get the preset stuff, if I just add I get 4+9+6+6 if I have it as is.

[science_man_88]

answer[#answer-(x*y-1)]=(answer[#answer-(x*y-1)]+n);carry=floor((c[y]*d[x])/10)

getting closer I got to:

[6, 0, 13, 6]

I know why it's happened, going to have to work out how to fix it. and it all has to do with: (x*y-1) because:
-(1*1-1) gives -0,
-(2*1-1) and -(1*2-1) give -1,But
-(2*2-1) gives -(3) so it skips -2 the place it should go in.

[CRGreathouse]

I wouldn't worry about carries; just fix them in the end.

[science_man_88]

Originally Posted by CRGreathouse

I wouldn't worry about carries; just fix them in the end.

It's not a carry problem it's a placement formula error problem because I should of had a formula to place the 6 in the 2nd from right not the very right.

[CRGreathouse]

Originally Posted by science_man_88

It's not a carry problem it's a placement formula error problem because I should of had a formula to place the 6 in the 2nd from right not the very right.

Every time you move one digit to the left on the multiplier you need to move the result one digit to the left.

[science_man_88]

Originally Posted by CRGreathouse

Every time you move one digit to the left on the multiplier you need to move the result one digit to the left.

Instead of multiplying I should be adding lol and subtracting 1 I think. I'll try it.

[science_man_88]

Got it working for 2 digits I'll check for 3. If it works I'll just need to chop off a 0 on the end.

[science_man_88]

Originally Posted by science_man_88

Got it working for 2 digits I'll check for 3. If it works I'll just need to chop off a 0 on the end.

oh and then the carry has to be done in the additions.

[science_man_88]

longmultiply(a,b)= c=eval(Vec(Str(a)));d=eval(Vec(Str(b)));carry=0;answer=vector(length(d)+length(c));forstep(x=#d,1,[-1],forstep(y=#c,1,[-1],n=((c[y]*d[x])+carry)%10;answer[#answer-((#d-x+1)+(#c-y+1)-1)]=(answer[#answer-((#d-x+1)+(#c-y+1)-1)]+n);carry=floor((c[y]*d[x])/10)));forstep(w=#answer,2,-1,answer[w-1]=answer[w-1]+floor(answer[w]/10);answer[w]=answer[w]%10)

One thing I can't seem to do is shift it maybe I'll have to add it in as another loop, this code is making me loopy.