Changelog for accumulate~, v. 0.2 (OSX‑MachO):
0.1: First version
0.2: Added argument syntax, updated mode controls

Changelog for analyzer~, v. 1.4.1 (OSX‑MachO):
1.4.1: fixed twiddle bug in fft code - mzed
1.4: Sample rate agnostic - mzed
1.3.1: Port to Universal Binary, assist strings, changed free() routine to call dsp_free() *before* freeing memory. - mzed
1.3: implements an altivec-optimized FFT and adds more windows
1.2.2: Matt Wright more debug info: addr of most recent input sig vector and gettime()
1.2.1: Matt Wright fixed pitch_getit() bounds error and added debug info

Changelog for cambio~, v. 0.2 (OSX‑MachO):
0.2: Renamed to "cambio~"
0.1: Matt's initial version

Changelog for cambio~, v. 0.2 (OSX‑CFM):
0.2: Renamed to "cambio~"
0.1: Matt's initial version

Changelog for cambio~, v. 0.2 (WIN):
0.2: Renamed to "cambio~"
0.1: Matt's initial version

Changelog for deinterleave, v. 1.0 (OSX‑MachO):
1.0: 2/20/02 Matt Wright changed to right-to-left output order
0.1: Tim's original version

Changelog for deinterleave, v. 1.0 (OSX‑CFM):
1.0: 2/20/02 Matt Wright changed to right-to-left output order
0.1: Tim's original version

Changelog for deinterleave, v. 1.1 (WIN):
1.1: Matt changed to use global version() procedure and compile on Windows/GCC
1.0: 2/20/02 Matt Wright changed to right-to-left output order
0.1: Tim's original version

Changelog for divcount~, v. 309.1 (OSX‑MachO):
309.1: Force Package Info Generation
309: Built for UB by Matt Wright
000308: Takahiko's final version.

Changelog for divcount~, v. 309 (OSX‑CFM):
309: Built for UB by Matt Wright
000308: Takahiko's final version.

Changelog for divcount~, v. 309.1 (WIN):
309.1: Force Package Info Generation
309: Built for UB by Matt Wright
000308: Takahiko's final version.

Changelog for interleave, v. 1.1 (OSX‑MachO):
1.1: Force Package Info Generation
1.1: UB recompile
1.0: Tim's original version

Changelog for interleave, v. 1.0 (OSX‑CFM):
1.0: Tim's original version

Changelog for interleave, v. 1.1 (WIN):
1.1: UB recompile
1.0: Tim's original version

Changelog for lcm, v. 0.2 (OSX‑MachO):
0.2: First release; fixed bug with typed-in arguments
0.1: Matt's initial version, 3/29/98.

Changelog for lcm, v. 0.2 (WIN):
0.2: First release; fixed bug with typed-in arguments
0.1: Matt's initial version, 3/29/98.

Changelog for list‑interpolate, v. 1.8 (OSX‑MachO):
1.8: Force Package Info Generation
1.7: Zero Pad mode
1.6: Merged Windows changes into real version
1.2: Made compilable in CW 8.3, Matt Wright 12/5/2
1.1: Bug fixed to really work with floats, Matt Wright 1/4/01

Changelog for list‑interpolate, v. 1.7 (OSX‑CFM):
1.7: Zero Pad mode
1.6: Merged Windows changes into real version
1.2: Made compilable in CW 8.3, Matt Wright 12/5/2
1.1: Bug fixed to really work with floats, Matt Wright 1/4/01

Changelog for list‑interpolate, v. 1.7 (WIN):
1.7: Zero Pad mode
1.6: Merged Windows changes into real version
1.2: Made compilable in CW 8.3, Matt Wright 12/5/2
1.1: Bug fixed to really work with floats, Matt Wright 1/4/01

Changelog for matlabcommunicate, v. 1.1.2 (OSX‑CFM):
1.1.2: Fixed bug so Matlab disp() works
1.1.1: increased Matlab text output capacity from 1000 to 10000 characters
1.1: buffer~ I/O, also fixed bug when evaluating an expression prints nothing
1.0: Initial hacking by Matt

Changelog for midifile, v. 3.0.2 (ANY):
3.0.2: Now outputs /text message for meta events like markers
3.0.1: Sets all note off velocities to 0 by default
3.0: Totally redesigned to make much better use of javax.sound.midi
2.2.2: Understands text meta events
2.2.1: Fixed a bug that caused all tracks of a multi-track file to start together even if one of them had a delay at the beg.
2.2: Now handles meta messages
2.1.4: Play now outputs the channel number if there is one.
2.1.3: Gratuitous increment of the version number.
2.1.2: Fixed a bug where an error would be reported if play was called and there was no data in one of the tracks.
2.1.1: Added play, play_from, and read messages and implemented open and save dialogs.

Changelog for OSC‑bcf2000, v. 0.1.3 (ANY):
0.1.3: Fixed line-break problem
0.1.2: Gratuitously incremented the version number
0.1: First release

Changelog for OSC‑unroute, v. 0.3 (ANY):
0.3: "correct" mode checks whether input already begins with an OSC address.
0.2: Added "correct"/"cheap" modes; made "correct" the default.
0.1.1: Moved name/value block from infosource.txt to js source file
0.1: First release

Changelog for peqbank~, v. 2.2 (OSX‑MachO):
2.2: Added "bank" message as a synonym for "list", dsp_free fixed -mz
2.1: Fixed bug of overwriting input signal vector with the filtered output
2.0.1: Force Package Info Generation
1.3: Fixed fast mode, 7/11/2000
1.2: Major fix of smooth mode disaster, Matt Wright 5/4/2000
1.1: Minor polishing by Matt Wright, 12/10/99 (version, tellmeeverything)
1.0: Tristan's initial version

Changelog for peqbank~, v. 2.1 (OSX‑CFM):
2.1: Fixed bug of overwriting input signal vector with the filtered output
2.0.1: Force Package Info Generation
1.3: Fixed fast mode, 7/11/2000
1.2: Major fix of smooth mode disaster, Matt Wright 5/4/2000
1.1: Minor polishing by Matt Wright, 12/10/99 (version, tellmeeverything)
1.0: Tristan's initial version

Changelog for peqbank~, v. 2.3 (WIN):
2.3: Tried to make peqbank_compute() be reentrant.
2.2: Added "bank" message as a synonym for "list", dsp_free fixed -mz
2.1: Fixed bug of overwriting input signal vector with the filtered output
2.0.1: Force Package Info Generation
1.3: Fixed fast mode, 7/11/2000
1.2: Major fix of smooth mode disaster, Matt Wright 5/4/2000
1.1: Minor polishing by Matt Wright, 12/10/99 (version, tellmeeverything)
1.0: Tristan's initial version

Changelog for printit, v. 0.4 (OSX‑MachO):
0.4: Added support for binary OSC packets sent as "FullPacket" messages (i.e., from the OpenSoundControl object)
0.3: Added support for many more a_type possibilities found in ext_mess.h
0.2.1: Force Package Info Generation
0.2: Using new version system
0.1B: Earliest version I could find

Changelog for printit, v. 0.3 (OSX‑CFM):
0.3: Added support for many more a_type possibilities found in ext_mess.h
0.2.1: Force Package Info Generation
0.2: Using new version system
0.1B: Earliest version I could find

Changelog for printit, v. 0.4.1 (WIN):
0.4.1: Added min and max OSC Packet sizes as a heuristic protection against crashing from non-OSC input.
0.4: Added support for binary OSC packets sent as "FullPacket" messages (i.e., from the OpenSoundControl object)
0.3: Added support for many more a_type possibilities found in ext_mess.h
0.2.1: Force Package Info Generation
0.2: Using new version system
0.1B: Earliest version I could find

Changelog for res‑transform, v. 1.78 (OSX‑MachO):
1.78: Force Package Info Generation
1.77: (MW) replaced NewPtr() with sysmem_newptr() so it will compile on Windows too. Plugged memory leak.
1.76: I can't get the alias feature to work
1.75: replaced getbytes by NewPtr and tested
1.74: fixed amplitude comparison so that non-zero gains were output, changed to larger model size (1024), added alias for sin-transform, didn't test or even compile anything
1.73: Updated tellmeeverything to disclose info about new features
1.72: AF changes setone to setonesinusoid and adds setoneresonance, removes numresonances changes matt's amprange and freqrange semantics
1.7: aliased messages without "-" for compatibility with Javascript, support for sinusoidal models,drop-partials byo from the CAST era , exponential decay maker for percussion effect or clean decay,all calculations in double precision
1.6.2: Added amprange and freqrange messages
1.6.1: Added "more_resonances" method
1.6: cleaned up so it works for Mac too by Matt Wright
1.5ALPHA: for windows compiled by Michael Zbyszynski
1.5: debugs setone and adds the -1 index feature, adds numresonances
1.4: by Matt Wright allows float midi-pitch
1.3: by Matt Wright has "setoneamplitude", "setonefrequency", "setonerate", and "setone"
1.2: by Matt Wright: compiles with CW 7 and new Max/MSP SDK

Changelog for res‑transform, v. 1.77 (OSX‑CFM):
1.77: (MW) replaced NewPtr() with sysmem_newptr() so it will compile on Windows too. Plugged memory leak.
1.76: I can't get the alias feature to work
1.75: replaced getbytes by NewPtr and tested
1.74: fixed amplitude comparison so that non-zero gains were output, changed to larger model size (1024), added alias for sin-transform, didn't test or even compile anything
1.73: Updated tellmeeverything to disclose info about new features
1.72: AF changes setone to setonesinusoid and adds setoneresonance, removes numresonances changes matt's amprange and freqrange semantics
1.7: aliased messages without "-" for compatibility with Javascript, support for sinusoidal models,drop-partials byo from the CAST era , exponential decay maker for percussion effect or clean decay,all calculations in double precision
1.6.2: Added amprange and freqrange messages
1.6.1: Added "more_resonances" method
1.6: cleaned up so it works for Mac too by Matt Wright
1.5ALPHA: for windows compiled by Michael Zbyszynski
1.5: debugs setone and adds the -1 index feature, adds numresonances
1.4: by Matt Wright allows float midi-pitch
1.3: by Matt Wright has "setoneamplitude", "setonefrequency", "setonerate", and "setone"
1.2: by Matt Wright: compiles with CW 7 and new Max/MSP SDK

Changelog for res‑transform, v. 1.78 (WIN):
1.78: Force Package Info Generation
1.77: (MW) replaced NewPtr() with sysmem_newptr() so it will compile on Windows too. Plugged memory leak.
1.76: I can't get the alias feature to work
1.75: replaced getbytes by NewPtr and tested
1.74: fixed amplitude comparison so that non-zero gains were output, changed to larger model size (1024), added alias for sin-transform, didn't test or even compile anything
1.73: Updated tellmeeverything to disclose info about new features
1.72: AF changes setone to setonesinusoid and adds setoneresonance, removes numresonances changes matt's amprange and freqrange semantics
1.7: aliased messages without "-" for compatibility with Javascript, support for sinusoidal models,drop-partials byo from the CAST era , exponential decay maker for percussion effect or clean decay,all calculations in double precision
1.6.2: Added amprange and freqrange messages
1.6.1: Added "more_resonances" method
1.6: cleaned up so it works for Mac too by Matt Wright
1.5ALPHA: for windows compiled by Michael Zbyszynski
1.5: debugs setone and adds the -1 index feature, adds numresonances
1.4: by Matt Wright allows float midi-pitch
1.3: by Matt Wright has "setoneamplitude", "setonefrequency", "setonerate", and "setone"
1.2: by Matt Wright: compiles with CW 7 and new Max/MSP SDK

Changelog for resonators~, v. 1.998 (OSX‑MachO):
1.998: Improved help file
1.997.1: Force Package Info Generation
1.997: (MW) Changed free() routine to call dsp_free() *before* freeing memory.
1.996: (MW) Changed NewPtr to sysmem_newptr(), added free() to plug memory leak
1.995: AF: changed getbytes back to NewPtr and increased resonances to 1024
1.99: AF: fixed typo in above changed NewPtr to getbytes and increased number of resonances to 512
1.98: MW+AF re-fixed coefficient interpolation bug to zero state variables and interpolate a1 aka a0.
1.97: fixed double precision version strange high frequency sound by turning off unrolling (Compiler now seems to do a good enough job on PowerPC anyway, fixed coefficient interpolation bug in case where smoothing is used and resonance models of different sizes are loaded
1.96: Doesn't crash when making a new object (filterstate array taken out of t_resonators)
1.95: added output amplitude vector interpolation, found unfixed bug in double stuff
1.9: added output amplitude vector
1.8: double precision mode, second outlet for filter state, ping completed and tested, repaired amplitude interpretation
1.7A: first windows compile
1.7: Doesn't get smooth/unsmooth backwards
1.6: Compiles under 7/02 Max SDK and CW 7.0