Im attempting to optimize my program. The whole goal of it is to crunch large 200-1000 digit numbers. While i did find an earler post about handling such data, im doing this with my own Big Number class. For simplicity i stored the numbers in an array. ( i could do a link list, but lets not get into that). Now the way its set up is such:

Take the number 12345, and in the array, its [1][2][3][4][5]. So each digit is stored as a short. so its a short array. Now, if i recall, and this is where you come in, a short is 8 bits long, so if im using a 32 bit proccessor, when i do an operation on one cell of the array will i be only utilizing the 1/4 of the proccessor? And now that i have a 64 bit proccessor... will i have to do something like:
__int64 BigNumber[arrayLenght];
//such that 1234567890 => [12345][67890]
//assuming 12345 is the largest number __int64 can hold.
Then another problem comes in, do i have to compile the program with a 64 bit compiler and not a 32 bit? Anyhelp/insight is apreciated. The whole goal is to cruch as much data in one proccess.

~Pcllwtfdtm
(P.C. Load letter?! What the F**K does that mean? ~office space)