While subtracting to values I'm getting the wrong answer. try this:

Connect three number boxes to a [-] object and enter the following values (first to the right inlet; second to the left, of course):

60 and 60.1 (the result is wrong = 0.099!)

60 and 60.2 (the result is right = 0.200)

Is there any explanations for that? Is it a bug or just something to do with C language? How do I fix that?

I tried to do the same operation with [expr], same results.

Its got to do with the precision of the datatype: float

Hi Dotche!

I have nothing clever to say regarding your problem, it's way beyond my limited knowledge of computing. But, I must say, just to avoid confusion, that your second example gives wrong result too. To see it, you must increase the width of the number box. I think it's better for pd to show consistency, even if it's a bad one

Recently I made one abstraction which serves me for truncating decimal numbers to arbitrary number of decimal places (see attachment).

sinewave

http://www.pdpatchrepo.info/hurleur/truncate_f.pd

60.1 - 60 won't actually be 0.099, it will be more like 0.09999999 or something (whatever the float precision is)

so, for most purposes, the difference is insignificant.

however, if you really need to get the exact number represented, then the patch sinewave just posted should do the trick.

Thanks everybody for the help! Especially to sinewave for the patch.