What is the output of this c code?

Replies

• 

  simplycoder

  Did it compile?
  I should throw exception on int a=0.7;
• 

  theAvinash

  yes dude
• 

  simplycoder

  What compiler are you using?
  Prefer DevC++ or Code blocks.(See the list of compilers they are using.)
  Please donot use Turbo C/C++.
  
  As well, conio is not a standard c library function.
• 

  VIJAYSY

  out put is -not equal
  because in first case data type of a is int so,a is stored as 0;
  in second case even data type of a is float,but internally a is stored as 0.699999
• 

  theAvinash

  how you can say in 2nd case a is stored as 0.699999?
• 

  theAvinash

  in 2 nd case.
  i think 0.7 is treated as double..so the if condition fails...
  if we type cast 0.7 to float as (float)0.7.... the answer is equl
• 

  simplycoder

  Avinash Waghole

  in 2 nd case.
  i think 0.7 is treated as double..so the if condition fails...
  if we type cast 0.7 to float as (float)0.7.... the answer is equl

  This can be justified by the computer organization-architecture.
  Read how floats are saved in computer. You would also require some knowledge on mantissa-exponent form.
  Try to represent 0.7 in binary, you would get the answer.
• 

  grsalvi

  1st case is clear
  but for 2nd case a is getting stored as .700000 still why 'not equal' getting printed ?
• 

  VIJAYSY

  yes a is stored as 0.700000
  (up to my knowledge)
  the 0.7 which is used to compare with 'a' is taken as double
  you can check with sizeof(0.7) it shows 8.
  if you type cast 0.7 to float it will convert to float ,
  the data type of 0.7(which is used to compare with a) is not in hands of user,
• 

  sulochana anand

  the answer is not equal.because int data type only stores without decimal values.so it will not undetand first part and controll will move to else part.
• 

  sulochana anand

  its a type of logical error.
• 

  grsalvi

  Okay...
  But suppose float a=0.7; (compiler stores : a=0.700000)
  And double b=0.7; (compiler stores : b=0.70000000000000)
  
  Still they should be equal.And why difference between a and b is not zero?
  Any idea what way the difference is getting calculated?
• 

  VIJAYSY

  i guess,
  out put is 0.000000 ,0,or 0.00000000000000 depends on format specifiers used in printf function ,while subtraction implicit type conversation will take place ,a is converted to double,
• 

  sulochana anand

  we know that when we declare any data type our system allocate space in memory accordingly.double and float both support decimal values.but only size is diffrent.float occupy 4 bytes and double 8 bytes.if we would store value which eceeds 4 byte in float it will give an error.but 0.7 is limited value so it will give right answer.
• 

  theAvinash

  Computers use a binary (0's and 1's) system to store decimal numbers. This leads to some inaccuracy, since some decimal values can't be stored exactly in binary....0.7 is stored as the value 0.69999999.Since that value isn't 0.7, the condition fails..
  
  The float function takes care of this problem -
  float(0.7)
  it rounds the value 0.69999999 to 0.7. Many decimal values are stored accurately in binary, for example 0.5, but many are not.
  
  its good to use the float function...
• 

  grsalvi

  Avinash Waghole

  Computers use a binary (0's and 1's) system to store decimal numbers. This leads to some inaccuracy, since some decimal values can't be stored exactly in binary....0.7 is stored as the value 0.69999999.Since that value isn't 0.7, the condition fails..
  
  The float function takes care of this problem -
  float(0.7)
  it rounds the value 0.69999999 to 0.7. Many decimal values are stored accurately in binary, for example 0.5, but many are not.
  
  its good to use the float function...

  I told above that there is no problem during storage as seen during debugging.
  The problem is difference between a and b is not zero,hence machine says they are unequal.
  
  see below,
  
  float a=0.7 (compiler stores 0.700000)
  double b=0.7 (compiler stores 0.70000000000000)
  
  During debugging : a-b=-9.2559631349318e+061