>> in decimal, then I might have to store a digit in each char.

That technique is called BCD (binary coded decimal) and is a good technique for storing numbers if the math you will be doing is minimal and you will be doing lots of inputing and ouput of the numbers. This is because BCD is very easy to read and write in decimal. However it is not as efficent for using in mathematical calculations. (But it is not too bad, I use it myself.)

However, if you want greater computational speed, then stick with your mult-byte binary value. To convert it to decimal, you need to have a modulos operator. (remainder). What you do is start with the initial number and take the modulos with the highest power of ten that "fits" in it This gives the first digit, then you "remove" the value of that digit and continue with the next lower power of ten to get the next digit. This continues until the power fo ten is 1.

continues.