Quiz submission record for quiz3-2-1 at Tue Jul  6 23:58:06 2004:

Your Answer for Question 1:
Since data is almost all the same, it is easily to get the underlying representation and to manipulate the data. Data is almost universal. It can be used in anyway like in C. It doesn't really check the types because they can  all be swtiched because they are essentially the same. I think one way around it is having data that is more strongly typed like that is found in java. By creating objects and such the data is more unique and therefore harder to manipulate. The tradeoff maybe in designing something complicated, it would probably take more time and would make something simple a lot harder to understand.

Your Answer for Question 2:
Subnormal are denormalized numbers. They have the same exponent as zero, but a nonzero significand. They allow a number to degrade in significance until it becomes 0. This allows for really small numbers.

Your Answer for Question 3:
It was to compensate for the lack of a guard digit. This allowed for better precision, because the guard digit gives an extra digit for rounding.

Your Answer for Question 4:
The IEEE 754 standard allows for a good way to express partial numbers (non-whole numbers). It was a clever way of using scientific notation. Those are really big advantages. The only disadvantage is the inconvience I suppose is the limitation. I know that big of numbers will not be used, but it looks we're heading in that direction anyway. If they didn't force that implicit 1 as the first digit, it would allow for a larger range of numbers. Other than that it's a good standard.


Your unique submission ID is quiz3-2-1-cs61c-al-1089183486-1670.