Objects			page : 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
liveGranul~	External		live input granulation DSP object Advanced live input granulation DSP engine for MaxMSP. Features are buffer based enveloppe, float/signal parameter control, internal spatialisation of each grain on 2-4-6-8 channels, high polyphony ( up to 512 simultaneous grains ).
LJController	External		LJ Controller is a manager for the USB measurement and automation peripheral LabJack U12 LJController external is based on the Labjack U12 driver (which can be downloaded on labjack.com). This Max/Msp external encapsulate the driver, in order to provide integration in Max/Msp applications, ease of use and advanced functionnalities. The LabJack U12 is a measurement and automation peripheral that enables the connection of a PC/Mac to the real-world. Labjack U12 is very convenient to build custom controllers for software, without any soldering (full box version, as in the picture), or to integrate into a hardware equipment (electronic circuit only without box, plugs, ...). It's small size (around 8x12 cm for version without the red cover) allows it to be embeded in small devices. Important: the Labjack hardware doesn't require separate power supply, it is USB powered.
lp.coshy	External		Generate random numbers from a Cauchy distribution The Cauchy distribution has a single parameter, É—. The standard Cauchy distribution is symmetrical. In some literature reference is made to a positive Cauchy distribution, and the lp.coshy object can optionally produce this variant. Although not found in the literature, lp.coshy supports a negative variant for the sake of symmetry to the positive form. Although the standard Cauchy distribution is symmetrical around zero, its mean does not converge but is undefined. Sooner of later one gets used to this.
lp.ernie	External		Select items from an urn ("Finite urn" probability model) The lp.ernie object implements a very flexible general-purpose "finite urn" model. The typical finite urn model deals with colored balls in an urn. For instance, there might be three red balls, two black balls, and four white balls. Balls are removed one at a time. They are not replaced after removal. As balls are removed, the probabilities of picking each color change. The characteristics of the urn model are the concern with the changing probabilities and the fact that the total number of balls is known.
lp.frim~	External		Frequency domain interval mutation The Frequency-domain Interval Mutator is a variant of the Time-domain Interval Mutator (lp.tim~) object, modified in two ways to better handle frequency-domain signals (for instance, signals produced by the fft~ object).
lp.frrr~	External		Low-frequency noise Low-frequency noise is generated from a sequence of random values chosen at a constant rate slower than the sampling rate. In its simplest form, it functions as a noise generator passed through a sample-and-hold module. However, lp.frrr~ also allows the samples between the randomly generated values to be interpolated, either linearly or quadratically.
lp.kg	External		Map I Ching values to nonstandard ranges One common way to consult the I Ching when there are less than 64 alternatives is to ask the I Ching how much weight to assign the individual choices. The I Ching is consulted for each of the choices wanted, and the hexagram values chosen are assigned as weights to each choice. Then a range of hexagram values proportional to the weight of each choice will be assigned. Thereafter, when a hexagram number is chosen from the I Ching, it is mapped to the choice with the relevant range. This process is computed for both "present" and "future" I Ching values. In lp.kg the choices are numbered starting at one.
lp.lili	External		Parametric linear congruence method The linear congruence algorithm has been the standard method of generating pseudorandom numbers since the late 1950s. More recent statistical literature (i.e., since about 1960) has pointed out numerous shortcomings with the algorithm. Despite this, linear congruence remains the method provided by practically all operating systems and programming libraries. Max is no exception. With carefully chosen parameters, LC can produce sequences of numbers that at least appear random at first glance. However, even with the most carefully chosen parameters, LC shows a number of correlations that are not in any sense random. For this reason, the Litter Power Package uses more modern methods that are more measurably random and robust. The algorithm used by default in the Litter Power Package is faster, to boot.
lp.lll~	External		Parametric linear congruence "noise" "White" noise using the Linear Congruence algorithm, while allowing you to specify values for the LC parameters. See lp.lili for more information on parametric linear congruence. The lp.lll~ object works very much like lp.lili, except that the integral values produced are scaled to the range -1 < x. < 1 for signals. Note that the scaling factor is calculated relative to the mod parameter, so the maximum power range is always produced (except for LC cycles that get stuck at a constant… this can happen!). For many parameter combinations, this cycle of numbers generated may be very short. In other words, the result may be much closer to pitch than noise. There are many intermediate signals.
lp.lonnie	External		Generate random numbers from a log-normal distribution The log-normal distribution is derived from the normal (or "Gaussian") distribution. By definition, if the logarithm of a set of random variables has a normal distribution, then the variable has a log-normal distribution. Conceptually, one can think of the lognormal distribution is as the product of many independent uniform distributions (in contrast to the normal distribution, which is derived from the notion of summing independent uniform distributions). The log-normal distribution is often used to model characteristics such as income distribution, distribution of grain sizes in geological contexts, and distribution of weight or height in biological contexts. The log-normal distribution has two parameters: mean and standard deviation. Values from a log-normal distribution are positive and skewed to the right (i.e., the median is greater than the arithmetic mean.)
lp.pfishie	External		Generate random numbers from a Poisson distribution The Poisson distribution has one parameter, É…, which happens to be both the expected mean and variance. (Standard deviation is therefore É…). The Poisson distribution generates non-negative integers only. It is defined for positive real values of É…. The Poisson distribution was originally developed as an efficient means of approximating the Bernoulli distribution for special cases (to wit, when the product np is small even when n is large). It has gained considerable popularity for use in algorithmic composition, particularly due to the influence of Iannis Xenakis, who used it extensively.
lp.ppp~	External		Popcorn (dust) noise This noise generator, known variously as popcorn or dust noise, generates exponentially distributed pulses of varying amplitude and pulse width. It resembles kinds of noise frequently found in telecommunications lines and sometimes in radio broadcast. Curiously, in most naturally occurring circumstances, the pulses are all of the same sign, either positive or negative. The lp.ppp~ object supports both, as well as a symmetrical variant in which positive and negative pulses are mixed at random. When the density of pops becomes high and pulse width also increases, it becomes possible for pops to overlap. The current implementation makes no provision for overlapping pops; one pop must be completed (i.e., the signal must return to 0) before the next one can begin. Thus, the actual frequency of pops may fall slightly underneath the specified mean.
lp.stacey	External		Collect statistics Count the input values, track cumulative minimum and maximum, and calculate mean, standard deviation, skew, and kurtosis.
lp.tata	External		Generate random numbers using the Tausworthe 88 algorithm The lp.tata object implements the Tausworthe 88 random number generator. This is currently the fastest algorithm that passes all standard statistical tests for randomness. It has a cycle of approximately 288 (that’s about 3 ÅE 1026) and generates random values across the entire range of 32-bit numbers (i.e., -2,147,483,648 < x <2,147,483,647). The lp.tata object allows you to scale the output to a given range.
lp.titi	External		Generate random numbers using the TT800 algorithm The lp.titi object implements the TT800 random number generator proposed by Makoto Matsumoto and Yoshiharu Kurita. This algorithm passes all standard statistical tests for randomness. It has a cycle of 2800 - 1 (that’s approximately 6 ÅE 10240) and generates random values across the entire range of 32-bit numbers (i.e., from -2,147,483,648 to 2,147,483,647).
page : 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Libraries
artificial tango	debug: SELECT prenom, nom FROM auteurs RIGHT JOIN auteur_libraries USING (id_auteur) WHERE auteur_libraries.id_library='104'Olaf Matthes	The artificial tango library is a collection of externals for Max/MSP dealing with recognition, analysis and generation of musical structures and events. Most objects take MIDI data as input. In order to use the objects from the artificial tango library FTM 2.0 has to be installed on your system.
Cosm	debug: SELECT prenom, nom FROM auteurs RIGHT JOIN auteur_libraries USING (id_auteur) WHERE auteur_libraries.id_library='147'Wesley Smith Graham Wakefield	Cosm is an integrated collection of externals and abstractions to assist the construction of navigable, sonified virtual worlds using Max/MSP/Jitter. Cosm has been designed to require only minimal changes to existing Max/MSP/Jitter patches to support a number of features valuable in the creation of virtual worlds. Supports six-degrees-of-freedom (6DoF) navigation using quaternions, spatial audio using 3rd order Ambisonics, distance filtering and doppler, collision detection using spherical intersection (query sphere), world boundaries, stereographic control, 3D field interaction, and a strategy for remote rendering.
EAMIR - the Electro-acoustic Musically Interactive Room	debug: SELECT prenom, nom FROM auteurs RIGHT JOIN auteur_libraries USING (id_auteur) WHERE auteur_libraries.id_library='133'V.j. Manzo	The EAMIR project is an open-source effort to enable educators with technology for music education. Software is designed around a common configuration which can be easily implemented by music educators around the world by downloading source code and standalone applications through eamir.org. Users who understand programming can edit the source code and post their new source and standalone applications back to the site for the EAMIR community to use. Much of the EAMIR software is also designed to allow students with physical and mental disabilities to create meaningful music using interfaces that are accessible to their needs. The EAMIR SDK allows users to easily create their own EAMIR-like applications by connecting preassembled interface modules to musical modules. The interfaces include cameras with color tracking, guitar hero controllers, wii remotes and devices, dance dance revolution pads, and other gaming controllers as well as traditional MIDI instruments like keyboards. created by V.J. Manzo www.vjmanzo.com | www.eamir.org
loadbang.net SQL	debug: SELECT prenom, nom FROM auteurs RIGHT JOIN auteur_libraries USING (id_auteur) WHERE auteur_libraries.id_library='99'Nick Rothwell	A Java library for communicating with SQL databases from MXJ. We currently support MySQL and HSQLDB. The HSQLDB system includes an embedded database instance, so it runs automatically from text files in Max\'s search path; no external database server configuration is necessary.
MaxAlea	debug: SELECT prenom, nom FROM auteurs RIGHT JOIN auteur_libraries USING (id_auteur) WHERE auteur_libraries.id_library='96'Carl Faia	MaxAlea contains various objects for random distributions and functions. MaxAlea was begun as a Max port of an existing PatchWork Library created in 1991-2 by Mikhail Malt. While the distributions and functions found in MaxAlea are similar to those found in the Patchwork version ,there are many differences in their functioning. The environment of Patchwork is static and is not designed for real-time work. Part of the incentive for creating these objects to work with Max was to have a dynamic and real-time environment with which to experiment and work with these algorithms in a manner as simple and straightforward as possible. One can change variables and manipulate the output in many ways in real-time. There are several different versions of the various stochastic models/processes best presented in the now classic references by Denis Lorrain and Charles Dodge. Carl Faia has used a variety of sources for the creation of this library which include the Lorrain, Dodge and Malt implementations as well as sources found on the WorldWideWeb. The externals found in the package include several random distributions, examples of random walks and 1/f noise algorithms, as well as one or two utilities written specifically for the MaxAlea library. Carl Faia wanted to make a coherent collection (as he thought Malt had managed to do in PatchWork) of these various algorithms and provide an interface easily accessible using the Max environment for real-time control. All these algorithms have been created using a seeded version of the random function found in the standard AINSI library. That is, each time the function is first run there will always be a different set of random numbers (unlike the random funtions found in Max, PatchWork and other versions of random number generators).
Panaiotis Objects	debug: SELECT prenom, nom FROM auteurs RIGHT JOIN auteur_libraries USING (id_auteur) WHERE auteur_libraries.id_library='122' Panaiotis	The Mac version is UB. These Max objects have been enhanced since the documentation to the left was written. Help files for the objects provide information on enhancements. The matrix object has been substantially upgraded. It now combines features of unpack, spray, funnel, append, and prepend into one object. This makes a great object to place between controllers and jit objects because it acts like a multi-prepend. There are new configuration commands and enhancements to the old: even, odd, mod,and range, among others). Most commands can be applied to inlets of outlets. There is also a mute function that adds another layer of control. Matrixctrl support has been enhanced. See the help file for full details and examples. Most other objects now fully support floats. RCer and autocount will count in float values, not just integers. Notegen16 is a 16 channel version of its predecessor: notegen. It is more generalized and much more efficient.
PeRColate	debug: SELECT prenom, nom FROM auteurs RIGHT JOIN auteur_libraries USING (id_auteur) WHERE auteur_libraries.id_library='5'Dan Trueman	PeRColate is an open-source distribution of a variety of synthesis and signal processing algorithms for Max, MSP, and Nato. It is centered around a (partial) port of the Synthesis Toolkit (STK) by Perry Cook (Princeton) and Gary Scavone (Stanford CCRMA). Like the STK, it provides a fairly easy to use library of synthesis and signal processing functions (in C) that can be wired together to create conventional and unusual instruments. Also like the STK, it includes a variety of precompiled synthesis objects, including physical modeling, modal, and PhISM class instruments; the code for these instruments can serve as foundations for creating new instruments (one example, the blotar, is included) and can be used to teach elementary and advanced synthesis techniques. Given it's STK heritage and educational function, PeRColate is largely un-optimized, though all the objects run on a 80MHz 7100, which is pretty good. PeRColate also includes a number of objects not from the STK; some are from RTcmix and others are our own evil creations, designed to crash your computer, but only after making some kind of interesting sound or image.
SFA Max/MSP Library	debug: SELECT prenom, nom FROM auteurs RIGHT JOIN auteur_libraries USING (id_auteur) WHERE auteur_libraries.id_library='172'Stefano Fasciani	The SFA-MaxLib is a collection of Max/MSP objects developed in the context of the VCI4DMI. It includes functions and utilities in the form of FTM externals, FTM abstractions and Max abstractions. FTM is a shared library for Max/MSP developed by IRCAM, which provides a small and simple real-time object system and a set of optimized services to be used within Max/MSP externals. List of FTM Externals: sfa.eig - eigenvalues; sfa.inputcombinations - combination generator; sfa.levinson - levinson-durbin recursion; sfa.lpc2cep - lpc to cepstra conversion; sfa.rastafilt - rasta filter; sfa.rmd - relative mean difference; sfa.roots - polynomial roots; List of Abstractions: sfa.bark.maxpat - energy of the Bark bands from time domain frame;sfa.bark2hz_vect.maxpat - Herts to Bark conversion;sfa.barkspect.maxpat - energy of the Bark bands from spectrum; sfa.ceil.maxpat - ceil function; sfa.featfluxgate.maxpat - gated distance on stream of feature vectors; sfa.fft2barkmx.maxpat - utility sub-abstraction of sfa.bark; sfa.fft2barkmxN.maxpat - utility sub-abstraction of sfa.barkspect; sfa.hynek_eq_coeff.maxpat - hynek equalization coefficients; sfa.hz2bark.maxpat - Hertz to Bark conversion; sfa.hz2bark_vect.maxpat - Hertz to Bark conversion for vectors; sfa.hz2mel.maxpat - Hertz to Mel conversion; sfa.idft_real_coeff.maxpat - utility sub-abstraction of sfa.rasta-plp; sfa.maxminmem.maxpat - minimum and maximum of a stream of data; sfa.mfcc.maxpat - MFCC coefficients; sfa.modalphafilter.maxpat - 1st order IIR lowpass on a stream of vectors; sfa.nonlinfeqscale.maxpat - linear spectrum to Bark or Mel scale conversion; sfa.rasta-plp.maxpat - PLP and RASTA-PLP coefficients; sfa.spectmoments.maxpat - 4 spectral moments (centroid, deviation, skewness, kurtosis); sfa.3spectmoments+flatness.maxpat - 3 spectral moments (centroid, deviation, skewness) and the spectral flatness; sfa.spectralflux.maxpat - spectral flux on stream of spectrum vectors; sfa.spectralfluxgate.maxpat - gated spectral flux on stream of spectrum vectors; sfa.std.maxpat - standard deviation; sfa.win_to_fft_size.maxpat - smaller FFT size given frame size; sfa.GCemulator.maxpat – 3D gestural controller emulator;
suivi	debug: SELECT prenom, nom FROM auteurs RIGHT JOIN auteur_libraries USING (id_auteur) WHERE auteur_libraries.id_library='88' Ircam	Two externals performing score following on soloist performances using Hidden Markov Models (HMM) Suivi is based on FTM and requires the shared library FTMlib for Max/MSP. Both externals use an FTM track object - a sequence of time-tagged FTM values - to store the score of the soloist performance to be followed. Notes, trills and other elements of the score are represented by FTM score objects (FTM scoob class). For the moment, scores can be imported from standard MIDI files only. An editor for the FTM track class, which will also provide a graphical control interface for the score follower is under development as well as the import of MusicXML files. The suivi object set is distributed within the IRCAM Forum.
tapemovie	debug: SELECT prenom, nom FROM auteurs RIGHT JOIN auteur_libraries USING (id_auteur) WHERE auteur_libraries.id_library='144'Tom Mays Olivier Pfeiffer Renaud Rubiano	tapemovie is a modular software environment for controlling, processing, and analyzing various media in realtime (sound, video, 3D, lighting). It has stood the test of numerous productions since 2007, for theater, dance, concert performance and installation - enabling precise control and sequencing of media and their interactions while at the same time allowing connections with multiple peripheral controllers and interfaces. It is programmed with Max/MSP/Jitter and exists as a standalone application (free download), as well as in patch version for advanced users.
Teleo Max Objects for Teleo Introductory Module	debug: SELECT prenom, nom FROM auteurs RIGHT JOIN auteur_libraries USING (id_auteur) WHERE auteur_libraries.id_library='55' MakingThings LLC	This set of software will stand alone for users of the new Teleo Introductory Module
Tristan Externals	debug: SELECT prenom, nom FROM auteurs RIGHT JOIN auteur_libraries USING (id_auteur) WHERE auteur_libraries.id_library='4'Tristan Jehan	FFT-based (optimized for the G4 processor)