Sponsored Content
Top Forums Programming Problem with rounding using lrint Post 302985096 by Don Cragun on Friday 4th of November 2016 05:29:41 PM
Old 11-04-2016
You need to determine what rounding rules you want to use for tie-breaking cases (see Wikipedia's discussion on rounding)
and you need to realize that even in double precision floating point, the result of a floating point calculation is not always exact (even when computing a value using decimal calculations would be exact). For example, the awk program (which uses double precision floating point for its calculations):
Code:
printf '.1 .5\n.3 .5\n' | awk '{printf("%.40f\n", $1 * $2)}'

produces the output:
Code:
0.0500000000000000027755575615628913510591
0.1499999999999999944488848768742172978818

not the output you might expect:
Code:
0.0500000000000000000000000000000000000000
0.1500000000000000000000000000000000000000

Which shows two examples where decimal arithmetic produces exact results, but binary arithmetic produced one result that was a little bit high and one result that was a little bit low.
 

10 More Discussions You Might Find Interesting

1. Shell Programming and Scripting

Rounding off using BC.

Hello again. I'm trying to use BC to calculate some numbers in a shell script. I want to have the numbers rounded off to 1 decimal place. for example: initsize=1566720 zipsize=4733 I'm trying to get the ratio between them. the equation is: (($initsize-$zipsize)/$initsize)*100 so... (3 Replies)
Discussion started by: noodlesoup
3 Replies

2. UNIX for Dummies Questions & Answers

Rounding problem

Hi, Can any one help me in finding a solution for rounding off to 2 decimal places. I am using the following code: VAR1=.01292105263157894736 VAR2=`echo "scale=2; $VAR1 * 100" | bc -l` The result I 'm getting is 1.29210526315789473600 But I need the output as 1.29 Thanks Shash (2 Replies)
Discussion started by: shash
2 Replies

3. Shell Programming and Scripting

Rounding off the value of Floating point value

Hello, i have some variables say: x=1.4 y=3.7 I wish to round off these values to : x = 2 (after rounding off) y = 4 (after rounding off) I am stuck. Please help. (7 Replies)
Discussion started by: damansingh
7 Replies

4. Shell Programming and Scripting

Rounding off to the next whole number

Hello, I searched a lot on this Forum. Please help me with the below problem. I want to divide two numbers and the result should be the next nearest whole number. E.G. Dividing 10.8/5 ideally gives 2.16. But the result should be 3 i.e. rounded off to the next whole number. Any help will... (2 Replies)
Discussion started by: damansingh
2 Replies

5. Linux

Rounding Script Help

I need some help with my rouding script. I have started pretty much from scratch and have no idea if its correct or even close but I have been trying and have gotten to this point. i keep getting syntax errors and im not sure what is wrong. Here is what I got let value=$1; while do let... (0 Replies)
Discussion started by: kingrj46
0 Replies

6. Shell Programming and Scripting

Rounding Script Help

I need some help with my rouding script. I have started pretty much from scratch and have no idea if its correct or even close but I have been trying and have gotten to this point. i keep getting syntax errors and im not sure what is wrong. Here is what I got let value=$1; while do let... (4 Replies)
Discussion started by: kingrj46
4 Replies

7. UNIX for Dummies Questions & Answers

Rounding a decimal

Hi, I am currently using tcsh I am trying to round a decimal number to the ten-thousandths place For instance: 1.23456 is rounded up towards 1.2346 I am not looking for truncation, but for rounding. Anyone know how to do this with awk or expr? Thanks (2 Replies)
Discussion started by: miniwheats
2 Replies

8. Shell Programming and Scripting

Rounding number, but....

Dear Experts, I'm trying to find a way to round a number but in this way: 14367.577 ---> 14000 I used the following to round the number to the closer integer: echo $var|awk '{print int($1+0.5)}' and also: xargs printf "%1.0f" However, they don't work for my above... (9 Replies)
Discussion started by: Gery
9 Replies

9. UNIX for Dummies Questions & Answers

Rounding up to nearest whole number

Hi all of you, Would be great if you help me with how to round up to whole number from my input values like 2.99996,2.17890,3.00002,-2.3456,-2.7890 o/p should be like 3,2,3,-2,-3 thnks in adv!!!! regards (3 Replies)
Discussion started by: Indra2011
3 Replies

10. UNIX for Dummies Questions & Answers

Rounding off a decimal

How to round off a decimal number to higher whole number using ceil command in unix? Eg. 4.41 or 4.11 or 4.51 should be rounded off to 5. (11 Replies)
Discussion started by: SanjayKumar28
11 Replies
fesetprec(3M)						  Mathematical Library Functions					     fesetprec(3M)

NAME
fesetprec, fegetprec - control floating point rounding precision modes SYNOPSIS
cc [ flag... ] file... -lm [ library... ] #include <fenv.h> int fesetprec(int prec); int fegetprec(void); DESCRIPTION
The IEEE 754 standard defines rounding precision modes for systems that always deliver intermediate results to destinations in extended double precision format. These modes allow such systems to deliver correctly rounded single and double precision results (in the absence of underflow and overflow) with only one rounding. The fesetprec() function sets the current rounding precision to the precision specified by prec, which must be one of the following values defined in <fenv.h>: FE_FLTPREC round to single precision FE_DBLPREC round to double precision FE_LDBLPREC round to extended double precision The default rounding precision when a program starts is FE_LDBLPREC. The fegetprec() function returns the current rounding precision. RETURN VALUES
The fesetprec() function returns a non-zero value if the requested rounding precision is established and 0 otherwise. ATTRIBUTES
See attributes(5) for descriptions of the following attributes: +-----------------------------+-----------------------------+ | ATTRIBUTE TYPE | ATTRIBUTE VALUE | |Architecture |Intel (see below) | |Availability |SUNWlibms | |Interface Stability |Stable | |MT-Level |MT-Safe | +-----------------------------+-----------------------------+ These functions are not available on SPARC systems because SPARC processors deliver intermediate results to destinations in single or dou- ble format as determined by each floating point instruction. SEE ALSO
fegetenv(3M), fesetround(3M), attributes(5) Numerical Computation Guide SunOS 5.10 1 Oct 2003 fesetprec(3M)
All times are GMT -4. The time now is 03:48 AM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy