Re: Calculation issue with octave (data unexpectedly go to zero at some point)

[rchretien]

Thank you all for your answers. I will try to answer every question that was
asked to me.

@ Gordon : You are right, I am drawing pseudo-random numbers at a certain
point in the code. I am not aware of Mersenne-Twister and /dev/random stuff,
so that I have simply used the rand() function the simplest way possible, in
order to get a pseudo-random number whose distribution is uniform in [0,1].
For the entropy of the system, here is what I got (I am running Linux, no
problem of translation under other OS) :

address@hidden ~]$ cat /proc/sys/kernel/random/entropy_avail
3464

@ Juan-Pablo : Indeed, I have printed the value taken by "n", value which
stops increasing at 27 under octave and goes as expected to 51 under matlab,
which I have explained in a previous message. I should have mentioned it
again, sorry for that. Concerning the outputs you asked for, they look to be
the same (visual constatation).

@ Bill : This is exactly what I expect from the code. Your suggestion about
the modulo is interesting. However, since your example using a comparison to
an epsilon yields different results, have you got another idea on how I
could perform the modulo operation ?

@ Dmitri and Briankaz : Why calling modulo on non-integer numbers is a bad
idea ? Does it come from roundoff errors in the numerical representation of
real numbers ? Have you got a suggestion for performing such an operation
with no risk of failure ? Basically, I want to record the simulation results
only at some multiple integers of the time step dt.

@ Dmitri (again) : can you please develop your explanation in the message
posted at 1:21 am

@ Juan-Pablo : your "mod_ml" function seems to work perfectly (both on the
much simplified and older version I posted and on the latest version of the
code). I wait before becoming too enthusiastic (I have to perform further
checkings tonight) but I think your function fixed the problem :) Now, is
the problem I had before related to a bad usage of modulo or related to a
bad coded modulo function ?

Thanks to all of you,

Renaud




--
View this message in context: 
http://octave.1599824.n4.nabble.com/Calculation-issue-with-octave-data-unexpectedly-go-to-zero-at-some-point-tp4662177p4662512.html
Sent from the Octave - General mailing list archive at Nabble.com.