Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 24 Nov 2013 20:01:09 +0100
From:      Daniel =?ISO-8859-1?Q?Pat=F3n?= <d.paton.d@gmail.com>
To:        freebsd-questions@freebsd.org
Subject:   Re: Many thanks to Jason Bacon
Message-ID:  <20131124200109.9ee6e380ca5dd9f41aa9cc04@gmail.com>
In-Reply-To: <1385308505.4081.51434569.06A345C6@webmail.messagingengine.com>
References:  <20131124031940.c78cd86d27f21fa0a6115c57@gmail.com> <1385308505.4081.51434569.06A345C6@webmail.messagingengine.com>

Next in thread | Previous in thread | Raw E-Mail | Index | Archive | Help

> I'm genuinely curious why it works better for you on FreeBSD than
> Debian. Can you elaborate? I have no idea what this software does.

OK. This software analyze mathematically the sound files. For example: spec=
trum, harmonics, formants, pitch, similitude between sounds, periodicity, e=
tc... Sound files have a high quantity of information. Animal sounds can ha=
ve frequencies between 20 and 20000 Hz or even more if you are studying spe=
cies that emit ultrasounds such as cetaceans. A simple file of 5 minutes of=
 a dolphin that a range between 20 and 150000 Hz that vary continuously. A =
simple spectra of this file in a normal computer can require 15 minutes in =
the calculus. The first problem for me is determining the software that I n=
eed. Audacity is quite good editing but is limited in the mathematical info=
rmation that analyze. R have sound libraries that are very useful but curio=
usly can not with long sound files. Even I had tested by SSH a computer in =
the Physics Department with R under a Debian system, with 32 processors, 64=
 GB and the cost in computing time is very high. Probably this is due to th=
e design in the construction of the software that can not use the parallel =
processing of information. Praat is the best election for the mathematical =
analysis of very long audio files, almost in my case. The secondary questio=
n is the operating system. As I only can use Praat in my laptop I need the =
maximum performance. I had observed that FreeBSD respond very good to high =
consume rates. Linux is slower. Probably this difference is minor for norma=
l work but empirically I observe important differences. With Debian in many=
 occasions my laptop is completely blocked. When I use FreeBSD I can prepar=
e the analysis of a sound file and continue with the rest of activities (ma=
il, web, text processor, etc...). My laptop never is blocked with FreeBSD a=
nd the time of analysis is minor. This is my experience but I have not benc=
hmarks or detailed explanations.

Regards
--=20
********************************************************
 Daniel Pat=F3n Dom=EDnguez
 Numerical Ecology. Ecology Unit
 Department of Plant Biology, Ecology and Earth Sciences
 Faculty of Sciences. University of Extremadura
 Avda. Elvas s/n 06071 Badajoz (Spain)
 http://unex.academia.edu/DanielPatonDominguez
 https://www.researchgate.net/profile/Daniel_Paton/
 http://sites.google.com/site/numericalecologyuex/home
********************************************************



Want to link to this message? Use this URL: <http://docs.FreeBSD.org/cgi/mid.cgi?20131124200109.9ee6e380ca5dd9f41aa9cc04>