Codes between Compilers

[1wvX92yv]

Thankyou sir.........

I tried replacing void with int,

I removed conio and everything related to it
I put std::infront of cin and cout

.......but the site still says the output is wrong

[TheIdeasMan]

And the for loop? Do you change that?

Can you update the code in the LiveWorkspace as well - then we can see what you have done exactly.

[Cubbi]

Once you get your program to compile with a C++ compiler, put it on http://ideone.com (which happens to be an online compiler that allows arbitrary input) and give it the input similar to what your contest gives you.

For example, run it with

7
5
10
1
2
75
99
100

in the input field, and if you wrote it correctly, it should show

120
3628800
1
2
24809140811395398091946477116594033660926243886570122837795894512655842677572867409443815424000000000000000000
933262154439441526816992388562667004907159682643816214685929638952175999932299156089414639761565182862536979208272237582511852109168640000000000000000000000
93326215443944152681699238856266700490715968264381621468592963895217599993229915608941463976156518286253697920827223758251185210916864000000000000000000000000

[1wvX92yv]

after changes...............

http://liveworkspace.org/code/3mQTbm$7

[1wvX92yv]

its not calculating 100! now!!!!!!!!!!!!!!!!!!!!!!

[1wvX92yv]

i changed it a bit more.......................

http://liveworkspace.org/code/3mQTbm$25

but it still wont calculate 100!

[TheIdeasMan]

This is where you have to be aware of the limits of your data types - you need a different method because with the standard types you are not going to get anywhere near factorial 100. Even a 64 bit unsigned int only gives 20 significant figures.

http://msdn.microsoft.com/en-us/library/s3f49ktz%28v=vs.80%29.aspx

These online judge problems are often not straight forward.

Spotted another thing in your code:

for(j=1,k=1; j<=t; j++,k++)
{
  std::cout<<factorial(testcase[k-1]);
  std::cout<<"\n";
}

k is the only variable used, so it could be written like this:

for(k=0; k < t; k++) { //I just edited that
      std::cout<<factorial(testcase[k]) << std::endl;  
}

You should get used to using arrays from zero and not subtract 1 everywhere.

[TheIdeasMan]

A different method could be to store numbers as bits, then do multiplication by bit shifting. Shifting left by 1 is the same as multiplying by 2. So do code to factor your multiplying number into 2's.

Example: 50 * 33

= 50 * 2 * 2 * 2 * 2 * 2 + 50
= (50 << 5) + 50

You can use the log function to see how factors of 2 there are in a number, rounding down.

Any way I am just writing this off the cuff, so I had better leave you in the expert hands of cubbi - as I haven't actually done it. (Maybe I should try it myself !!)

Sorry I didn't mention this earlier - I was focussing on getting your code to compile & didn't ask what the actual question was.

Hope all goes well :)

Edit:

http://www.cprogramming.com/tutorial/bitwise_operators.html

[1wvX92yv]

ok, you lost me there..........i have no clue whatsoever about bit shifting multiplication (Remember, i'm a student who started programming about 7 months ago)

[1wvX92yv]

i still dont know how to get 100!.................

in tc++ it worked.......

i tried using long long, unsigned __int64, and unsigned long long ....it dint work.

but it worked for lond double.............and the site wont accept it.

[TheIdeasMan]

As I said unsigned long long (a 64 bit number) will only give 20 significant places - as cubbi's post shows you need a lot more than that.

long double will hold a big enough number, but will only give you 18 significant places, so you need another method. As I said these online judge things aren't that easy.

ok, you lost me there..........i have no clue whatsoever about bit shifting multiplication (Remember, i'm a student who started programming about 7 months ago)

Did you see the link I posted? I added it afterwards. It explains what bit shifting means, & I gave a quick example.

You can google to get more examples - hell there is probably the answer to the question somewhere.

Any way I am going to bed - it's 04:45AM here. Have a go at the new method, then cubbi or someone else may help you.

Good Luck

[Cubbi]

..oh, right, LWS allows input too now.
Calculating factorials up to 100 is not a one-liner program in C++, that's what makes it a competition, I guess (in practice, it's trivial with boost-1.53 or with other multiprecision libraries)

Or you could just hardcode all 100 possible answers as an array of strings.