Hi,
The screenshot presented below shows that there are 2 pairs that negates each other which should equal to 0 on column1, same with column2. With that in mind, the total value should be 0, right? But upon rounding it up to 13th decimal value and beyond, it no longer equates to 0.
I've attached a search query that will replicate the said issue stated above.
| makeresults
| eval column1=600.0000
| append
[| makeresults
| eval column1=390.0000]
| append
[| makeresults
| eval column1=355.0000]
| append
[| makeresults
| eval column1=575.0000]
| append
[| makeresults
| eval column1=355.0000]
| append
[| makeresults
| eval column1=590.0000]
| append
[| makeresults
| eval column1=600.0000]
| append
[| makeresults
| eval column1=355.0000]
| append
[| makeresults
| eval column1=-600.0000]
| append
[| makeresults
| eval column1=-590.0000]
| append
[| makeresults
| eval column1=-355.0000]
| append
[| makeresults
| eval column1=-575.0000]
| append
[| makeresults
| eval column1=-355.0000]
| append
[| makeresults
| eval column1=-390.0000]
| append
[| makeresults
| eval column1=-600.0000]
| append
[| makeresults
| eval column1=-355.0000]
| eval column2=round(column1/1.12,20) , column1=round(column1,20)
| addcoltotals
Thank you.
Regards,
Raj
My first guess would be that this is the magic of floating point numbers at work. If you're having real life issues like the example you shared, you might want to look at using the sigfig() function: https://docs.splunk.com/Documentation/Splunk/8.0.5/SearchReference/MathematicalFunctions#sigfig.28X....
So it seems it's really a 'magical' number then. Thank you sir for your reply.