In case you are interested in the technical details...
"Normal" integer variables can only store whole numbers.
Let's start with unsigned numbers. Unsigned means that the number doesn't have a sign - it's always positive. You don't really have these in Pawn IIRC but that's a different thing. In AMX Mod X, a variable is either 32 or 64 bit, depending on the architecture you run on.
Anyway, with unsigned numbers, the variable simply contains the binary representation of the number it holds. For the decimal number 9, this would be 1001 for example. 7 would be 111. And so on. If you need to be able to store negative numbers as well, you usually reserve one bit for "sign"; usually the most significant bit; if this bit is 1, the number is negative. There are more used methods of storing negative integers; the most widely used one is the "two's complement" thing, I guess google will help you on these.
Anyway, you can compute the decimal value of a binary number by doing:
bit0 * 1 + bit1 * 2 + bit2 * 4 + bit3 * 8 + bit4 * 16 + bit5 * 32 + ...
bit0 is the LSB, least significant bit, which you usually draw on the far right. ie. the number 13 would be stored like this in a 8 bit number:
00001101
this is: 1*1 + 0*2 + 1*4 + 1*8 + 0*16 + ... = 1 + 4 + 8 = 13
If you want to store numbers with decimal points, you have to pick a different storage method. Usually people use either fixed point numbers or floating point numbers. For some reason I associate fixed point numbers with banking software and such; in the PC and graphics world they're very rare.
So, fixed point numbers store the whole part and the rest. ie. 13.37 would be stored as two separate numbers: 13 and 37. Floating point numbers, on the other hand, store the "mantissa" and the exponent. ie. they store 1337 as a whole and then tell you where to place the decimal point.
Usually, computers use the "IEEE 754" standard for floating point numbers; for 32 bit numbers, it specifies 1 bit for sign, 8 bits for the exponent and 23 bits for mantissa. So if you store a floating point number and then interprete it as a "normal" integer, you will get weird, usually big results (big because the exponent is stored in the highest 9 bits (following the sign), so when it is non-zero, your binary number will get some pretty big value).
I hope I could clarify this
__________________