Quote (mcfighter @ Sep 12 2012 06:45pm)

I have some questions about unsigned binary- decimal and what not.

Just a few of the questions.
An explanation of how you got the answer would be helpful too.

If you're give 5 bits how many distinct numbers can be represented if you only have 2 symbols?

Given 10 bits, what's the decimal equivalent of the smallest unsigned binary integer that can be represented?

What's the decimal equivalent of 0001 1101?

Thank you.

Quote (carteblanche @ Sep 12 2012 11:18pm)

you have to explain what system you're using to generate floating points. how many bits are used to represent > 1 vs < 1? IEEE has some standards, but i'm not sure you're using them.

http://en.wikipedia.org/wiki/IEEE_floating_point

you could simply writing floating points similar to how you write in decimal, with a period separating the two sections. everything to the left increases by a power of 2, everything to the right decreases by a power of 2.

eg: 11.01 = 3 + 0*(1/2) + 1*(1/4) = 3.25

the first question is easy though. there are as many distinct numbers are there are possible values.  so 5 bits is from 00000 to 11111, which is 2^5 = 32

Quote (mcfighter @ Sep 12 2012 07:29pm)

Thank you, not sure by what system you mean?
We are using C, and Linux if that helps.

What's the decimal equivalent of the largest unsigned binary integer that can be represented using 'n' bits?

Quote (carteblanche @ Sep 12 2012 11:44pm)

I meant standard, not system sorry.

What does each digit represent? Unlike integers, there is no straightforward logical way to handle it afaik. If you have n bits, are the first x going to be the integer portion and the last n-x going to be the decimal portion? are all n bits going to be the decimal portion? is the first x digits going to be the base (assuming leading 1) and last n - x digits going to be an exponent? etc etc

iirc C does not have an unsigned floating point out of the box. their float is system dependent, though usually IEEE which is signed. you could, of course, follow IEEE's format then treat the sign bit as an additional base or exponent digit. i could be wrong but i dont think IEEE has a standard for any generic n bits, but rather very specific numbers of bits (eg 16, 32, 64, and 128 are defined, but 37 bits is not defined)

do you handle infinity, negative infinity, and NotANumber?

bottom line, you have to ask your teacher how he wants you to write your unsigned floating numbers.

Quote

If you're give 5 bits how many distinct numbers can be represented if you only have 2 symbols?

Quote

the first question is easy though. there are as many distinct numbers are there are possible values. so 5 bits is from 00000 to 11111, which is 2^5 = 32

Quote

Given 10 bits, what's the decimal equivalent of the smallest unsigned binary integer that can be represented?

Quote

What's the decimal equivalent of 0001 1101?

Quote (carteblanche @ Sep 12 2012 11:58pm)

Shit, i just noticed by decimal you meant base 10. i thought you meant floating point (eg 1.234). It's been a long day :/

ok ignore everything i said before.



actually, dont ignore this one. my answer was right.



all zeros, which is zero.


counting i from the right starting with i = 0, each i'th digit represents 2^i

so 0001 1101 is
0*2^7+
0*2^6+
0*2^5+
1*2^4+
1*2^3+
1*2^2+
0*2^1+
1*2^0
= 16 + 8 + 4 + 1

Quote (carteblanche @ Sep 12 2012 11:58pm)

Shit, i just noticed by decimal you meant base 10. i thought you meant floating point (eg 1.234). It's been a long day :/

ok ignore everything i said before.



actually, dont ignore this one. my answer was right.



all zeros, which is zero.


counting i from the right starting with i = 0, each i'th digit represents 2^i

so 0001 1101 is
0*2^7+
0*2^6+
0*2^5+
1*2^4+
1*2^3+
1*2^2+
0*2^1+
1*2^0
= 16 + 8 + 4 + 1

Quote (rockonkenshin @ Sep 13 2012 01:37pm)

You aren't even trying to help yourself with your homework questions.