asked on

truncate

VC5

Is there function that i can truncate
a double of value 3.333333333333333......
to a fewer number of decimal point?

double D = 3.33333333..... // return by some function
double MyNewD = truncate_D_to_2_decimal_point(D) // ???

arnond

one way is to do:

double truncate (double D,int n)
{
long multiplied;
double to_return;

multiplied = (D*pow(10,n)); // save only the first n digits after the decimal point . (as a long integer).
to_return = (multiplied / (pow(10,n)); // return to the trtunced (original ) number.

return (to_return);
}

for example:
in truncate (0.123456789,5) you'll get:
multiplied = 12345
to_return = 0.12345

nietod

That would work, but you have to understand that because of the limitations of the floating point format, not all decimal numbers will be expressible, thus when you truncate the number, you may see the last digit or two change in value. This is not problem with arnond's method, this is a limitation of the floating point system. For this reason, most programmer's don't truncate the value at all. They may limit the number of digits that print, but will try to retain full precision internally.

arnond

nietod is right, you can limit the number of digits you print using the format specifications of your favorite output function. for example: in printf() you should use "%.pf".
the 'p' stands for precision, i.e.: how many decimal places to display.

arnond

leowlf,
did my suggestion help ?
would you like me to post an answer for this question ?

leowlf

ASKER

arnond,
Your suggestion is good enough for me,
please post an answer.
Thank you.

ASKER CERTIFIED SOLUTION

arnond

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

pibarale

Hi Arnod,
I am using C language because I can not use C++ due to some desing issue.
how do we proceed if the double has this value
UINT s = 10;
int n = 16;
double dx = (double)(s/100.0);
here is dx = 0.10000000000000001
The above logic does not work.

ozo

> you have to understand that because of the limitations of the floating point format, not all decimal numbers will be expressible, thus when you truncate the number, you may see the last digit or two change in value.

pibarale

Thanks OZO. This is also my understanding.
One concern we are use C code in such a huge platform like Win32 (using VS 2008 compiler). Can this irreguarity be fix by the vendor in near future?
With this I stop with satisfaction.