search
Log In
0 votes
82 views
In Single Precision floating point number when E=0  and M not equals to 0 it represents a fractional form..(BiAS=+127)

when writing the decimal form why do we right (-1)^S * (1.M) * 2^(-126) instead of  (-1)^S * (1.M) * 2^(-127)??
in CO and Architecture 82 views
0

@ I feel you should take it as (-1)^S * (1.M) * 2^(E-127) for single precision because this is the format prescribed by IEEE .

The actual exponent is biased by 127 to get E that is the actual value of the exponent is E − 127. This gives the range: $2^{1-127}=2^{-126} \, to \,\, 2^{254-127}=2^{127}$  .

are you preparing for GATE'20 ?

1 Answer

0 votes

@ I feel you should take it as (-1)^S * (1.M) * 2^(E-127) for single precision because this is the format prescribed by IEEE .

The actual exponent is biased by 127 to get E that is the actual value of the exponent is E − 127. This gives the range: $2^{1-127}=2^{-126} \, to \,\, 2^{254-127}=2^{127}$  .

 

Related questions

0 votes
0 answers
1
90 views
How do we perform 0 – 1 in 2’s complement subtraction using 1-bit register??
asked Mar 21, 2019 in CO and Architecture Doraemon 90 views
0 votes
3 answers
2
113 views
How to do:(1000)base3 – (121)base3 without converting them to base 10 the final answer should be in base3
asked Feb 4, 2019 in CO and Architecture DIYA BASU 113 views
0 votes
0 answers
3
125 views
what is (120)base 10=( ? )base 64 ??
asked Feb 4, 2019 in CO and Architecture DIYA BASU 125 views
1 vote
2 answers
4
87 views
How addition of $+3$ and $-3$ in sign magnitude form is $-5?$
asked Mar 19, 2019 in Digital Logic srestha 87 views
...