Forum

Expr Node


#1

round(0.5) = 0 but it has to be 1 !

arctan() is not implementet

DIV and MOD gives wrong results!

ARG has bugs

COTAN has bugs… etc etc


#2

ok. this node has some secrets… there is a little help page at FAQ Expr

round(0.5) = 0 but it has to be 1 !

i think this the default behaviour of the used rounding function - which does commercial rounding. values right in the middle between two integers are rounded to the next even number: round(0.5)=0, round(1.5)=2, round(2.5)=2, round(3.5)=4. while this helps bankers to distribute rounding errors between creditors and debitors, it may be somehow unintuitive to all mathematicians. trunc is usually your friend

arctan() is not implemented

try arctan2(x, y) which calculates arctan(x/y).
arctan2 handles the y=0 case quite nice.

DIV and MOD gives wrong results!
ARG has bugs
COTAN has bugs… etc etc

okok… i had some similar experiences sometime. can you give an example?


#3

A DIV B should be
|A/B|

A MOD B should be
A - |A/B| *B
while | | are the GaussianBrackets aka floor()

Ex: in vvvv
3.6 MOD 2 is 0 but it has to be 1.6
3.6 DIV 2 is 2 but it has to be 1

oschatz:
ok. this node has some secrets… there is a little help page at FAQ Expr

round(0.5) = 0 but it has to be 1 !

i think this the default behaviour of the used rounding function - which does commercial rounding. values right in the middle between two integers are rounded to the next even number: round(0.5)=0, round(1.5)=2, round(2.5)=2, round(3.5)=4. while this helps bankers to distribute rounding errors between creditors and debitors, it may be somehow unintuitive to all mathematicians. trunc is usually your friend

arctan() is not implemented

try arctan2(x, y) which calculates arctan(x/y).
arctan2 handles the y=0 case quite nice.

DIV and MOD gives wrong results!
ARG has bugs
COTAN has bugs… etc etc

okok… i had some similar experiences sometime. can you give an example? [[[[[[[[[[[[[[[[[[[[ùíí


#4

thanks, i´ll look into that…