Q If a short int can run out of room and wrap around, why not always use long integers?
A Both short integers and long integers will run out of room and wrap around, but a long integer will do so with a much larger number. For example, an unsigned short int will wrap around after 65,535, whereas an unsigned long int will not wrap around until 4,294,967,295. However, on most machines, a long integer takes up twice as much memory every time you declare one (four bytes versus two bytes), and a program with 100 such variables will consume an extra 200 bytes of RAM. Frankly, this is less of a problem than it used to be because most personal computers now come with millions (if not billions) of bytes of memory. Optical computers may one day have gigabytes (or even terabytes!) of memory.
Q What happens if I assign a number with a decimal point to an integer rather than to a float? Consider the following line of code:
int aNumber = 5.4;
A A good compiler will issue a warning, but the assignment is completely legal. The number you've assigned will be truncated into an integer. Thus, if you assign 5.4 to an integer variable, that variable will have the value 5. Information will be lost, however, and if you then try to assign the value in that integer variable to a float variable, the float variable will have only 5.
Q Why not use literal constants; why go to the bother of using symbolic constants?
A If you use the value in many places throughout your program, a symbolic constant allows all the values to change just by changing the one definition of the constant. Symbolic constants also speak for themselves. It might be hard to understand why a number is being multiplied by 360, but it's much easier to understand what's going on if the number is being multiplied by degreesInACircle.
Q What happens if I assign a negative number to an unsigned variable? Consider the following line of code:
unsigned int aPositiveNumber = -1;
A A good compiler will warn, but the assignment is legal. The negative number will be assessed as a bit pattern and assigned to the variable. The value of that variable will then be interpreted as an unsigned number. Thus, -1, whose bit pattern is 11111111 11111111 (0xFF in hex), will be assessed as the unsigned value 65,535. If this information confuses you, refer to Appendix A.
Q Can I work with C++ without understanding bit patterns, binary arithmetic, and hexadecimal?
A Yes, but not as effectively as if you do understand these topics. C++ does not do as good a job as some languages at "protecting" you from what the computer is really doing. This is actually a benefit because it provides you with tremendous power that other languages don't. As with any power tool, however, to get the most out of C++ you must understand how it works. Programmers who try to program in C++ without understanding the fundamentals of the binary system often are confused by their results.