C - Data type

sizeof

    printf("sizeof(char) = %d\n", sizeof(char));        // 1
    printf("sizeof(short) = %d\n", sizeof(short));      // 2
    printf("sizeof(int) = %d\n", sizeof(int));          // 4
    printf("sizeof(long) = %d\n", sizeof(long));        // 4
    printf("sizeof(float) = %d\n", sizeof(float));      // 4
    printf("sizeof(double) = %d\n", sizeof(double));    // 8

    printf("sizeof(char) = %d\n", sizeof(char*));       // 4
    printf("sizeof(short) = %d\n", sizeof(short*));     // 4
    printf("sizeof(int) = %d\n", sizeof(int*));         // 4
    printf("sizeof(long) = %d\n", sizeof(long*));       // 4
    printf("sizeof(float) = %d\n", sizeof(float*));     // 4
    printf("sizeof(double) = %d\n", sizeof(double*));   // 4

Reason:
    char
        -128 ~ 127 -> 8 bits needed = 1 byte
    unsigned char
        0 ~255
    short
        -32768 ~ 32767 -> 16 bits needed = 2 bytes
    unsigned short
        0 ~ 65535

    int 

        -2147483648 ~ 2147483647
    float 
        32 bit (s + exp + fraction = 1 + 8 + 23)

    double
        64 bit (1 + 11 +52)

    pointer
        4 bytes = 1 word

Precision Problem
    1. estimate the precision you want.
    2. use big number.