/[svn]/linuxsampler/trunk/ChangeLog
ViewVC logotype

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 3573 - (show annotations) (download)
Tue Aug 27 21:36:53 2019 UTC (4 years, 7 months ago) by schoenebeck
File size: 82005 byte(s)
NKSP: Introducing floating point support.

* NKSP language: Added support for NKSP real number literals and
  arithmetic operations on them (e.g. "(3.9 + 2.9) / 12.3 - 42.0").

* NKSP language: Added support for NKSP real number (floating point)
  script variables (declare ~foo := 3.4).

* NKSP language: Added support for NKSP real number (floating point)
  array script variables (declare ?foo[3] := ( 1.1, 2.7, 49.0 )).

* NKSP built-in script function "message()" accepts now real number
  argument as well.

* Added built-in NKSP script function "real_to_int()" and its short
  hand form "int()" for casting from real number to integer in NKSP
  scripts.

* Added built-in NKSP script function "int_to_real()" and its short
  hand form "real()" for casting from integer to real number in NKSP
  scripts.

* Bumped version (2.1.1.svn6).

1 Version SVN trunk (?)
2
3 * general changes:
4 - Fixed compiler error in Pool.h.
5 - Require C++11 compiler support.
6 - Autoconf: Added m4/ax_cxx_compile_stdcxx.m4 macro which is used
7 for checking in configure for C++11 support (as mandatory
8 requirement) and automatically adds compiler argument if required
9 (e.g. -std=C++11).
10
11 * Real-time instrument scripts:
12 - Added method ScriptVM::setExitResultEnabled() which allows to
13 explicitly enable the built-in exit() function to optionally accept
14 one function argument; the value of the passed exit() function
15 argument will then become available by calling
16 VMExecContext::exitResult() after script execution.
17 - 64 bit support for NKSP integer scripts variables (declare $foo).
18 - Variable names, function names and preprocessor condition names must start
19 with a regular character (a-z or A-Z); starting them with a digit or
20 underscore is not allowed.
21 - NKSP parser fix: equal comparison operator "=" and not equal comparison
22 operator "#" must only accept integer operands.
23 - NKSP language: Implemented support for standard units like Hertz, seconds,
24 Bel including support for metric unit prefixes; so one can now e.g.
25 conveniently use numbers in scripts like "5us" meaning "5 microseconds",
26 or e.g. "12kHz" meaning "12 kilo Hertz", or e.g. "-14mdB" meaning
27 "minus 14 Millidecibel", or e.g. "28c" meaning "28 cents" (for tuning).
28 - NKSP language: Introduced "final" operator "!" which is specifically
29 intended for synthesis parameter values to denote that the synthesis
30 parameter value is intended to be the "final" value for that synthesis
31 parameter that should explicitly be used by the engine and thus causing
32 the sampler engine to ignore all other modulation sources for the same
33 synthesis parameter (like e.g. LFO, EG); by simply prefixing a value,
34 variable or formula with this new "!" operator the expression is marked as
35 being "final".
36 - NKSP script editor API: Added support for detecting standard unit tokens
37 and their potential metric prefix token.
38 - NKSP language: Added support for NKSP real number literals and
39 arithmetic operations on them (e.g. "(3.9 + 2.9) / 12.3 - 42.0").
40 - NKSP language: Added support for NKSP real number (floating point) script
41 variables (declare ~foo).
42 - NKSP language: Added support for NKSP real number (floating point) array
43 script variables (declare ?foo[]).
44 - Built-in script function "message()" accepts now real number argument as
45 well.
46 - Added built-in script function "real_to_int()" and its short hand form
47 "int()" for casting from real number to integer in NKSP scripts.
48 - Added built-in script function "int_to_real()" and its short hand form
49 "real()" for casting from integer to real number in NKSP scripts.
50
51 * test cases:
52 - Fixed compiler errors in test cases.
53 - Updated README for how to compile & run test cases.
54 - Updated test case
55 MutexTest::testDoubleLockStillBlocksConcurrentThread() to latest
56 expected behaviour of the Mutex class implementation (recursive
57 mutex type).
58 - Added test cases for NKSP core language aspects and core built-in
59 functions.
60 - Fixed thread tests segfaulting on Linux.
61
62 Version 2.1.1 (27 Jul 2019)
63
64 * Real-time instrument scripts:
65 - Fixed behavior of built-in NKSP functions change_sustain(),
66 change_cutoff_attack(), change_cutoff_decay(), change_cutoff_sustain()
67 and change_cutoff_release().
68
69 * general changes:
70 - Only play release trigger samples on sustain pedal up if this behaviour
71 was explicitly requested by the instrument (otherwise only on note-off).
72 - Fixed compiler warnings.
73 - Fixed compilation error when cross-compiling to Mac.
74 - FX Sends: Provide more useful error messages on routing problems
75 (see bug #169).
76 - LSCP doc: Be more clear describing the two distinct approaches
77 of using external vs. internal effects (see bug #169).
78 - "optional" class: Fixed comparison operators.
79
80 * Gigasampler/GigaStudio format engine:
81 - Format extension: If requested by instrument then don't play release
82 trigger sample on note-off events.
83
84 * SFZ format engine:
85 - Fixed memory leak when releasing samples
86 (fixes bug #307, patch by Jacek Roszkowski)
87 - Fixed potential crash when a sample is shared by more than one region
88 (fixes bug #308, patch by Jacek Roszkowski).
89 - Opcode 'sample': Added support for built-in sample '*silence'
90 (fixes bug #310, patch by Jacek Roszkowski).
91
92 Version 2.1.0 (25 Nov 2017)
93
94 * SFZ format engine:
95 - added support for <global>, <master> and #define (patch by Alby M)
96 - Removed code duplication in SFZ file loading code.
97 - Added support for sfz extension opcode 'script' which may be used to
98 load real-time instrument script file (NKSP script language).
99 - Implemented opcode set_ccN (initial patch by Giovanni Senatore).
100 - Fixed unintended volume fade-in of voices under certain conditions.
101 - sfz parser: allow missing space between header and opcode
102
103 * Gigasampler/GigaStudio format engine:
104 - Fixed clicks and pumping noise with Lowpass Turbo filter on very low
105 cutoff settings.
106 - Got rid of resembling an ancient GSt misbehavior which did not pitch at
107 all if an up-pitch of more than 40 semi tones was requested (I don't
108 think there is any stock gig sound that requires this behavior to
109 resemble its original sound).
110 - Added support for controlling whether the individual EGADSR stages may
111 be aborted (as LinuxSampler extension to the original GigaStudio 4
112 format).
113
114 * general changes:
115 - fixed printf type errors (mostly in debug messages)
116 - use unique_ptr instead of auto_ptr when building with C++11
117 - Added RTAVLTree class which is a real-time safe ordered multi-map, thus
118 allowing to sort data efficiently in real-time safe manner.
119 - RTList class: added methods for moving/inserting elements to arbitrary
120 position within a list.
121 - RTList class: added method fromPtr() for creating an Iterator object from
122 a raw element pointer.
123 - Attempt to partly fix resetting engine channels vs. resetting engine, an
124 overall cleanup of the Reset*(), ConnectAudioDevice(),
125 DisconnectAudioDevice() API methods would still be desirable though,
126 because the current situation is still inconsistent and error prone.
127 - Active voices are now internally grouped to "Note" objects, instead of
128 being directly assigned to a keyboard key. This allows more fine graded
129 processing of voices, which is i.e. required for certain instrument
130 script features.
131 - Fix: Release trigger voices were not spawned on sustain pedal up
132 (CC #64) events.
133 - Fix: Release trigger voices can now also distinguish correctly between
134 note-off and sustain pedal up events.
135 - All engines: Increased ramp speed of volume smoother and pan smoother
136 (while slow rate of crossfade smoother is preserved) to allow quick
137 volume and pan changes by instrument scripts for instance.
138 - gig/sf2/sfz: Fixed aftertouch channel pressure events being processed
139 properly and thus were ignored for certain uses.
140 - Added new C++ API method
141 VirtualMidiDevice::SendChannelPressureToSampler().
142 - windows, 32-bit: fixed potential crashes by making sure the stack in
143 sub threads is 16-byte aligned
144 - fixed numerous compiler warnings
145 - Fixed invalid (note-on) event ID being assigned to new Note objects.
146 - Revised fundamental C++ classes "Thread", "Mutex" and "Condition" which
147 fixes potential undefined behavior.
148 - Fixed Note object leak when triggering notes on keys which did not
149 have a valid sample mapped (fixes bug #252).
150 - Fixed compilation errors when compiling with CONFIG_DEVMODE enabled.
151 - linuxsampler binary fix: option --create-instruments-db ignored
152 subsequent optional argument due to glibc's implementation oddity
153 which expects a "=" sign, but no space between them.
154
155 * packaging changes:
156 - removed unnecessary dependency to libuuid
157 (originated by libgig's usage of it)
158 - Automake: set environment variable GCC_COLORS=auto to allow GCC to
159 auto detect whether it (sh/c)ould output its messages in color.
160 - Debian: Fixed packaging error about invalid "Source-Version"
161 substitution variable.
162 - Debian: Raised Debian compatibility level to Debian 9 "Stretch".
163 - Debian: Added build dependency to libsqlite3-dev for building
164 linuxsampler with instruments DB support.
165
166 * Real-time instrument scripts:
167 - Implemented scheduler for delayed MIDI events and for suspended scripts.
168 - Built-in script function "wait()": implemented support for function's
169 "duration-us" argument, thus scripts using this function are now
170 correctly resumed after the requested amount of microseconds.
171 - Built-in script function "play_note()": implemented support for
172 function's "duration-us" argument, thus notes triggered with this
173 argument are now correctly released after the requested amount of
174 microseconds.
175 - Fix: script events were not cleared when engine channel was reset,
176 potentially causing undefined behavior.
177 - Fixed crash which happened when trying to reference an undeclared
178 variable.
179 - Built-in script function "play_note()": Added support for passing
180 special value -1 for "duration-us" argument, which will cause the
181 triggered note to be released once the original note was released.
182 - Instrument script classes now exported with the liblinuxsampler C++ API.
183 - Added new API method ScriptVM::syntaxHighlighting() which provides
184 a convenient syntax highlighting backend for external instrument
185 script editor applications.
186 - Added new C++ API class "ScriptVMFactory".
187 - Extended parser issues to provide not only first line and first
188 column, but also last line and last column of issue (thus marking
189 the precise span of the issue within the source code).
190 - Reload script automatically after being modified by an instrument
191 editor.
192 - NKSP language grammar correction: allow empty event handler bodies
193 like "on note end on".
194 - Implemented built-in script function "change_vol()".
195 - Implemented built-in script function "change_tune()".
196 - Implemented built-in script function "change_pan()".
197 - Implemented built-in script function "change_cutoff()".
198 - Implemented built-in script function "change_reso()".
199 - Implemented built-in script function "event_status()".
200 - Added built-in script constants "$EVENT_STATUS_INACTIVE" and
201 "$EVENT_STATUS_NOTE_QUEUE" both for being used as flags for
202 "event_status()" function.
203 - NKSP language: Added support for bitwise operators ".or.", ".and."
204 and ".not.".
205 - NKSP language scanner: Fixed IDs matching to require at least one
206 character (i.e. when matching function names or variable names).
207 - NKSP language scanner: disabled unusued rules.
208 - Fixed behavior of play_note() and note_off() functions which must
209 be distinguished engine internally from "real" MIDI note on/off
210 events in order to avoid misbehaviors like hanging notes.
211 - Implemented built-in script variable "$KSP_TIMER".
212 - Implemented built-in script variable "$NKSP_REAL_TIMER".
213 - Implemented built-in script variable "$NKSP_PERF_TIMER".
214 - Implemented built-in script variable "$ENGINE_UPTIME".
215 - Implemented built-in script function "inc()".
216 - Implemented built-in script function "dec()".
217 - NKSP language fix: division expressions were evaluated too often.
218 - NKSP language fix: string concatenation operator was right
219 associative instead of left (to right).
220 - NKSP language correction: allow empty statements in entire language
221 (i.e. "if end if").
222 - Implemented built-in script function "stop_wait()".
223 - Implemented built-in script variable "$NI_CALLBACK_ID".
224 - Implemented built-in script variable "$NI_CALLBACK_TYPE".
225 - Implemented built-in script variable "$NKSP_IGNORE_WAIT".
226 - Added support for read-only built-in variables (respectively
227 handled by the script parser).
228 - Added built-in script constant "$NI_CB_TYPE_INIT".
229 - Added built-in script constant "$NI_CB_TYPE_NOTE".
230 - Added built-in script constant "$NI_CB_TYPE_RELEASE".
231 - Added built-in script constant "$NI_CB_TYPE_CONTROLLER".
232 - NKSP Language: Added support for user defined script functions.
233 - Implemented built-in script function "change_attack()".
234 - Implemented built-in script function "change_decay()".
235 - Implemented built-in script function "change_release()".
236 - Fixed all change_*() built-in script functions to apply their
237 synthesis parameter changes immediately in case the respective note
238 was triggered at the same time, instead of scheduling the parameter
239 change, especially because it would cause some parameter types's
240 changes either to be ramped (i.e. change_vol()) or other types even
241 to have not effect at all (i.e. change_attack()).
242 - Implemented built-in script function "sh_left()".
243 - Implemented built-in script function "sh_right()".
244 - Implemented built-in script function "min()".
245 - Implemented built-in script function "max()".
246 - NKSP Fix: "init" event handler was not always executed when a script
247 was loaded.
248 - NKSP built-in wait() script function: abort script execution if
249 a negative or zero wait time was passed as argument (since this is
250 a common indication of a bug either of the script or even of the
251 engine, which could lead to RT instability or even worse).
252 - ScriptVM: Implemented automatic suspension of RT safety
253 threatening scripts.
254 - Provide more user friendly error messages on syntax errors.
255 - Fixed NKSP parser warning "Not a statement" when assigning an
256 initializer list to an array variable.
257 - Implemented built-in script array variable "%ALL_EVENTS".
258 - Implemented built-in script function "in_range()".
259 - Implemented built-in script function "change_amp_lfo_depth()".
260 - Implemented built-in script function "change_amp_lfo_freq()".
261 - Implemented built-in script function "change_pitch_lfo_depth()".
262 - Implemented built-in script function "change_pitch_lfo_freq()".
263 - Implemented built-in script function "change_vol_time()".
264 - Implemented built-in script function "change_tune_time()".
265 - Implemented built-in script function "fade_in()".
266 - Implemented built-in script function "fade_out()".
267 - Fixed acceptance of wrong data type of parameters passed to built-in
268 script functions "change_vol()", "change_tune()", "change_pan()",
269 "change_cutoff()", "change_reso()", "change_attack()", "change_decay()",
270 "change_release()", "change_amp_lfo_depth()", "change_amp_lfo_freq()",
271 "change_pitch_lfo_depth()" and "change_pitch_lfo_freq()".
272 - Added built-in script function "get_event_par()" and implemented some
273 of its possible parameter selections.
274 - Added built-in script function "set_event_par()" and implemented some
275 of its possible parameter selections.
276 - Fixed a bunch of scheduler time related bugs.
277 - Fixed polyphonic variables not being reset to zero after usage.
278 - Built-in "ignore_event()" function: argument is now optional, like with
279 built-in function "ignore_controller()".
280 - Implemented built-in script function "change_velo()".
281 - Implemented built-in script function "change_note()".
282 - Adjusted behavior of "change_vol()" and "change_tune()" to a more
283 intuitive behavior if used in combination with "change_vol_time()" or
284 "change_tune_time()" respectively: now tuning/volume changes are only
285 assigned (without delay) immediately to a new note if the respective
286 timing function has not been called before, otherwise the volume/tuning
287 changes are automatically faded (before, only the event's time stamp was
288 relevant).
289 - Implemented built-in script function "array_equal()".
290 - Implemented built-in script function "search()".
291 - Implemented built-in script function "sort()".
292 - NKSP Fix: Never suspend "init" event handlers.
293 - Implemented built-in script function "same_region()" (currently only
294 available for gig format engine).
295 - Added built-in script constant "$NKSP_LINEAR".
296 - Added built-in script constant "$NKSP_EASE_IN_EASE_OUT".
297 - Implemented built-in script function "change_vol_curve()".
298 - Implemented built-in script function "change_tune_curve()".
299 - built-in "play_note()" function now supports a sample playback start
300 offset with argument 3, where special value -1 means to use the regular
301 sample offset as defined by the instrument file.
302 - Built-in array variable %KEY_DOWN[] is now a read-only variable.
303 - Built-in variable $EVENT_NOTE is now a read-only variable.
304 - Built-in variable $EVENT_VELOCITY is now a read-only variable.
305 - built-in "play_note()" function now accepts -2 for its fourth argument
306 (note duration) which means the life time of the note shall be sticked
307 to the requested note number of argument 1.
308 - Fix: built-in "play_note()" function now returns 0 as result value if
309 -1 was passed for its fourth argument (note duration) and the respective
310 parent note is already gone.
311 - Implemented built-in script function "change_play_pos()".
312 - NKSP language: Added support for user declared const array variables.
313 - NKSP language: Raise parser warning if array variable is accessed with
314 an index that exceeds the array's size.
315 - NKSP language fix: Unknown characters were not handled correctly.
316 - NKSP language: Added support for "synchronized .. end synchronized"
317 code blocks.
318 - Implemented built-in script function "abort()" which allows to abort
319 another script handler by passing its callback ID.
320 - Fixed potential memory access bug and potential undefined behavior of
321 "init" event handlers.
322 - Print a time stamp along to each call of built-in function "message()".
323 - ScriptVM API: Added VMParserContext::preprocessorComments() which allows
324 to retrieve all code blocks filtered out by the preprocessor.
325 - Added built-in script function "fork()".
326 - Added built-in array variable %NKSP_CALLBACK_CHILD_ID[].
327 - Added built-in variable $NKSP_CALLBACK_PARENT_ID.
328 - Fixed potential crash when accessing dynamic built-in array variables.
329 - Added built-in script function "callback_status()".
330 - Added built-in constant $CALLBACK_STATUS_TERMINATED.
331 - Added built-in constant $CALLBACK_STATUS_QUEUE.
332 - Added built-in constant $CALLBACK_STATUS_RUNNING.
333 - Removed max. value limitation of built-in functions "change_attack()",
334 "change_decay()" and "change_release()" to i.e. allow passing 2000000
335 for doubling the respective time.
336 - NKSP script editor syntax highlighting API: Fixed app termination due
337 to a lexer start condition stack underrun.
338 - NKSP preprocessor: Fixed wrong behavior on nested USE_CODE_IF() and
339 USE_CODE_IF_NOT() preprocessor statements.
340 - NKSP: Added built-in preprocessor condition NKSP_NO_MESSAGE, which
341 can be set to disable all subsequent built-in "message()" function calls
342 on preprocessor level.
343 - Implemented built-in script function "change_sustain()".
344 - NKSP script editor syntax highlighting API: catch all fatal lexer errors,
345 to avoid the editor app to crash on ill-formed text input.
346 - Added built-in script function "change_pan_time()".
347 - Added built-in script function "change_pan_curve()".
348 - Added built-in script function "change_cutoff_attack()".
349 - Added built-in script function "change_cutoff_decay()".
350 - Added built-in script function "change_cutoff_sustain()".
351 - Added built-in script function "change_cutoff_release()".
352 - Added built-in script function "change_cutoff_lfo_depth()".
353 - Added built-in script function "change_cutoff_lfo_freq()".
354
355 * Instruments DB:
356 - Fixed memory access bug of general DB access code which lead to
357 undefined behavior.
358 - Cleanup of instruments DB file creation and opening code.
359 - The instrument DB path of linuxsampler's --create-instruments-db argument
360 is now optional, if it is missing, then a default location is used.
361 - Added support for scanning SFZ (.sfz) files.
362 - Added support for scanning Sound Font (.sf2) files.
363 - Fixed undefined DB transaction behavior.
364
365 Version 2.0.0 (15 July 2015)
366
367 * packaging changes:
368 - fixed building with newer MinGW-w64
369 - Mac OS X: support the new dir for Core Audio SDK
370 - Mac OS X: fixed building outside source directory
371 - made sure all source files for hostplugins are included when
372 doing "make dist"
373 - removed empty directories left from the cvs to svn migration
374 - fixed compilation with gcc 4.6.1
375 - another "make dist" fix, for the LV2 plugin
376 - made --enable-pthread-testcancel default on Mac OS X
377 - minor Makefile fix for building lscpparser when build and source
378 directories are separate
379 - Mac OS X: made it possible to specify plugin installation dir to
380 configure
381 - Mac OS X: Makefile fix for the install-strip target
382 - fixed compilation with gcc 4.7
383 - fixed configure script error with old autoconf versions
384 - lsatomic.h: use gcc provided atomic functions if building with
385 gcc 4.7 and C++11
386 - modernized configure script
387 - fixed linkage error when building with
388 LDFLAGS="-Wl,--no-undefined" (#190)
389 - fixed compilation with Clang 3.2
390 - removed usage of deprecated Automake variable INCLUDES
391 - fixed building with C++11
392 - build fix: ChangeFlagRelaxed.h was missing in makefile
393 - build fix: libsndfile compiler flags were missing in some
394 makefiles
395 - fix for building with bison 3.0 (#202)
396 - Mac OS X: added temporary hack allowing to spawn gigedit as callback
397 on the process's main thread
398 - fixed build error on newer MinGW
399 - support building with older jack versions
400 - support building with spaces in vst sdk path
401 - enabled automake 'subdir-objects' option and moved external
402 source references (vst, au, asio) from makefiles to cpp files,
403 in order to get rid of warnings from automake 1.14
404
405 * general changes:
406 - Refactoring: moved the independent code from
407 the Gigasampler format engine to base classes
408 - command line option '--profile' is currently disabled, since the
409 respective profiling code is currently broken
410 - Introduced support for optional environment variable
411 "LINUXSAMPLER_PLUGIN_DIR" which allows to override the directory
412 where the sampler shall look for instrument editor plugins
413 (patch by Luis Garrido, slightly modified).
414 - implemented Roland GS NRPN 1ArrH which allows to set volume per note
415 - implemented Roland GS NRPN 1CrrH which allows to set pan per note
416 - implemented Roland GS NRPN 1DrrH which allows to set reverb send per
417 note (in this implementation of the sampler its simply hard routed to
418 the 1st effect send of the sampler channel, no matter what the actual
419 effect type is)
420 - implemented Roland GS NRPN 1ErrH which allows to set chorus send per
421 note (in this implementation of the sampler its simply hard routed to
422 the 2nd effect send of the sampler channel, no matter what the actual
423 effect type is)
424 - implemented support for internal LADSPA effects
425 - raised limit of program change queue from 100 to 512
426 (as suggested by Alex Stone)
427 - implemented sine LFO, pulse LFO and saw LFO
428 - experimental support for per voice equalization
429 - added command line option --exec-after-init
430 - Introduced new C++ API method:
431 EngineChannel::InstrumentFileName(int index)
432 allowing to retrieve the whole list of files used for the loaded
433 instrument on an engine channel (a.k.a. part). Some GigaStudio
434 instruments for example are splitted over several files like
435 "Foo.gig", "Foo.gx01", "Foo.gx02", ...
436 - Added new C++ API method Sampler::GetGlobalMaxVoices().
437 - Added new C++ API method Sampler::GetGlobalMaxStreams().
438 - Added new C++ API method Sampler::SetGlobalMaxVoices().
439 - Added new C++ API method Sampler::SetGlobalMaxStreams().
440 - Various "const" and "restrict" optimizations.
441 - all engines: add pan CC value to instrument pan parameter before
442 applying panning, instead of using two separate pan functions in
443 series (#182)
444 - added a lock guard class for exception safe mutex handling and
445 used it everywhere appropriate
446 - Immediately apply scale tuning changes to active voices.
447 - Exposed scale tuning to C++ API (along to the already existing standard
448 SysEx way).
449 - lsatomic.h fixes: seq_cst load and store were suboptimal for x86
450 and broken for ppc64. (Seq_cst loads and stores are actually not
451 used in LS, so the bug wasn't noticable.)
452 - lsatomic.h: added ARMv7 support
453 - Added support for multiple MIDI input ports per sampler channel (and
454 added various new C++ methods for this new feature / design change, old
455 C++ API methods for managing SamplerChannel's MIDI inputs are now marked
456 as deprecated but are still there and should provide full behavior
457 backward compatibility).
458 - AbstractEngine::GSChecksum(): don't allocate memory on the stack (was
459 unsafe and caused compilation error with clang 2.x).
460 - Bugfix: only process the latest MIDI program change event.
461 - Introducing the LSCP shell, which provides convenient control of the
462 sampler from the command line by providing LSCP aware features.
463 - VirtualMidiDevice: Added support for program change.
464 - VirtualMidiDevice: Added support for bank select (MSB & LSB).
465 - VirtualMidiDevice: Added support for pitch bend.
466 - Aftertouch: extended API to explicitly handle channel pressure and
467 polyphonic key pressure events (so far polyphonic pressure was not
468 supported at all, and channel pressure was rerouted as CC128 but not
469 used so far).
470 - Added initial support for real-time instrument scripts. The script VM
471 code is shared by all sampler engine implemementations, however only the
472 gig file format currently provides support for storing instrument scripts
473 (as LinuxSampler extension to the original GigaStudio 4 format).
474
475 * Gigasampler format engine:
476 - implemented the "round robin keyboard" dimension
477 - fixed round robin and random dimensions for cases when number of
478 dimension zones is not a power of two
479 - made round robin use a counter for each region instead of each
480 key
481 - bugfix: pitch LFO controller "internal+aftertouch" was broken
482 - bugfix: filter keyboard tracking was broken
483 - filter performance fix (an unnecessary copy was made of the
484 filter parameters in each sub fragment)
485 - handle special case when pan parameter in gig file has max or
486 min value
487 - Exclusive Groups: don't ever stop voices of the same note,
488 doesn't sound naturally with a drumkit
489 - fixed EG1 modulation when attack or release is zero
490 - Fixed support for 'aftertouch' attenuation controller.
491 - Fixed crash that happened with velocity split sounds under certain
492 conditions (see also previous commit on libgig).
493 - fixed behaviour of filter LFO
494
495 * SFZ format engine:
496 - Initial implementation (not usable yet)
497 - added support for v2 multiple stage envelope generators
498 - added a fine-tuned v1 envelope generator instead of using the
499 one from the gig engine
500 - fixed memory leak and memory handling errors
501 - added support for trigger=first, trigger=legato and sw_previous
502 - allow non-numerical key values ("C#4" for example)
503 - "key" opcode now sets pitch_keycenter too
504 - fixed error when unloading instrument with same sample used by
505 multiple regions
506 - added some opcode aliases, like loopmode for loop_mode, to be
507 more compatible
508 - added support for trigger=release and rt_decay
509 - added support for off_mode=normal
510 - added support for random, seq_position, seq_length and volume
511 - added v1 LFO opcodes to parser (no support in engine yet)
512 - added support for amp_veltrack and amp_velcurve_N
513 - fine-tuned the default velocity curve
514 - added support for transpose
515 - fixed crash when using sw_down/up
516 - improved logic for sw_lokey/hikey/up/down/last
517 - added more v1 aliases to parser
518 - reduced memory usage for sfz data
519 - RT-safeness: avoid malloc in audio thread
520 - fixed a bug that could cause voice stealing to fail
521 - optimized sample lookup
522 - improved support for exclusive groups (group, off_by and
523 off_mode)
524 - added support for controller triggered regions (on_locc/on_hicc)
525 - added support for loop_mode=one_shot
526 - fixed playback of 16 bit wav files on big endian CPUs
527 - added support for Ogg Vorbis sample files
528 - use loop markers from sample file if loop_start and loop_end are
529 not set in sfz file
530 - implemeted filters. Filter types: lowpass, bandpass, bandreject
531 and highpass. 1, 2, 4 and 6 pole filters. Opcodes: fil_type,
532 cutoff, resonance, fil_veltrack, fil_keytrack, fil_keycenter,
533 cutoff_cc, cutoff_chanaft.
534 - bugfix: zero ampeg_sustain didn't work
535 - added support for velocity effect on amplifier envelope time
536 (ampeg_vel2attack, ampeg_vel2decay, ampeg_vel2sustain and
537 ampeg_vel2release)
538 - added support for EG hold (ampeg_hold)
539 - added support for sample offset (offset)
540 - use the newly introduced signal units model
541 - implemented opcodes lfoN_delay,
542 lfoN_freq, lfoN_pan, lfoN_cutoff, lfoN_resonance
543 - implemented opcodes ampeg_delay, ampeg_vel2delay
544 pitcheg_delay, pitcheg_start, pitcheg_attack, pitcheg_hold,
545 pitcheg_decay, pitcheg_sustain, pitcheg_release, pitcheg_vel2delay,
546 pitcheg_vel2attack, pitcheg_vel2hold, pitcheg_vel2decay,
547 pitcheg_vel2sustain, pitcheg_vel2release, pitcheg_depth
548 - implemented opcodes pitchlfo_delay, pitchlfo_freq,
549 pitchlfo_depth, fillfo_delay, fillfo_freq, fillfo_depth,
550 amplfo_delay, amplfo_freq, amplfo_depth
551 - implemented opcodes fileg_delay, fileg_start, fileg_attack,
552 fileg_hold, fileg_decay, fileg_sustain, fileg_release, fileg_vel2delay,
553 fileg_vel2attack, fileg_vel2hold, fileg_vel2decay, fileg_vel2sustain,
554 fileg_vel2release
555 - implemented opcode lfoN_wave
556 - implemented opcode pitchlfo_depthccN
557 - implemented opcodes lfoN_phase, lfoN_phase_onccX,
558 lfoN_pitch, lfoN_pitch_onccX
559 - implemented opcodes pitchlfo_fade,
560 fillfo_fade, amplfo_fade, lfoN_fade, lfoN_fade_onccX
561 - implemented opcodes fillfo_freqccN,
562 pitchlfo_freqccN, amplfo_freqccN, lfoN_freq_onccX
563 - implemented opcodes ampeg_delayccN, ampeg_startccN,
564 ampeg_attackccN, ampeg_holdccN, ampeg_decayccN, ampeg_sustainccN,
565 ampeg_releaseccN, egN_timeX_onccY, egN_levelX_onccY
566 - lfoN_* and egN_* opcodes defined in group sections
567 are now taken into account
568 - implemented curves
569 - implemented opcodes volume_onccN, volume_curveccN
570 - implemented opcode volume_smoothccN
571 - implemented opcodes fillfo_depthccN, amplfo_depthccN,
572 lfoN_volume, lfoN_volume_onccX, lfoN_volume_smoothccX,
573 lfoN_freq_smoothccX, lfoN_pitch_smoothccX, lfoN_pan_onccX,
574 lfoN_pan_smoothccX, lfoN_cutoff_onccX, lfoN_cutoff_smoothccX,
575 lfoN_resonance_onccX, lfoN_resonance_smoothccX, lfoN_delay_onccX
576 - implemented opcode end
577 - implemented opcodes egN_amplitude_onccX,
578 egN_volume, egN_volume_onccX, egN_cutoff, egN_cutoff_onccX,
579 egN_pitch, egN_pitch_onccX, egN_resonance, egN_resonance_onccX
580 - implemented opcodes xfin_lokey, xfin_hikey,
581 xfout_lokey, xfout_hikey, xf_keycurve, xfin_lovel, xfin_hivel,
582 xfout_lovel, xfout_hivel, xf_velcurve, xfin_loccN, xfin_hiccN,
583 xfout_loccN, xfout_hiccN, xf_cccurve
584 - implemented opcodes pan_onccN, pan_smoothccN,
585 pan_curveccN, egN_pan, egN_pan_curve, egN_pan_onccX,
586 egN_pan_curveccX, pitch_veltrack
587 - when failed to parse a sfz file
588 print the line number on which the error occurs
589 - use common pool of CC objects to minimize RAM usage
590 - implemented opcodes amplfo_delay_onccN,
591 amplfo_fade_onccN, fillfo_delay_onccN, fillfo_fade_onccN,
592 pitchlfo_delay_onccN, pitchlfo_fade_onccN
593 - implemented opcodes fileg_delay_onccN,
594 fileg_start_onccN, fileg_attack_onccN, fileg_hold_onccN,
595 fileg_decay_onccN, fileg_sustain_onccN, fileg_release_onccN,
596 fileg_depth_onccN, pitcheg_delay_onccN, pitcheg_start_onccN,
597 pitcheg_attack_onccN, pitcheg_hold_onccN, pitcheg_decay_onccN,
598 pitcheg_sustain_onccN, pitcheg_release_onccN, pitcheg_depth_onccN
599 - implemented automatic aliasing *ccN <-> *_onccN
600 - *lfo_freqccN wasn't working when the respective *lfo_freq
601 was not set or was set to zero
602 - lfoN_freq_onccX wasn't working when lfoN_freq
603 was not set or was set to zero
604 - implemented opcodes resonance_onccN, resonance_smoothccN,
605 resonance_curveccN, cutoff_smoothccN, cutoff_curveccN
606 - implemented opcodes fillfo_depthchanaft,
607 fillfo_freqchanaft, amplfo_depthchanaft, amplfo_freqchanaft,
608 pitchlfo_depthchanaft, pitchlfo_freqchanaft
609 - implemented opcodes pitch_onccN,
610 pitch_curveccN, pitch_smoothccN, pitch_stepccN
611 - implemented opcodes volume_stepccN, pan_stepccN,
612 cutoff_stepccN, resonance_stepccN, lfoN_freq_stepccX,
613 lfoN_volume_stepccX, lfoN_pitch_stepccX, lfoN_pan_stepccX,
614 lfoN_cutoff_stepccX, lfoN_resonance_stepccX
615 - implemented opcodes eq1_freq, eq2_freq, eq3_freq,
616 eq1_freqccN, eq2_freqccN, eq3_freqccN, eq1_bw, eq2_bw, eq3_bw,
617 eq1_bwccN, eq2_bwccN, eq3_bwccN, eq1_gain, eq2_gain, eq3_gain,
618 eq1_gainccN, eq2_gainccN, eq3_gainccN
619 - implemented opcodes delay, delay_onccN, delay_random,
620 delay_samples, delay_samples_onccN
621 - implemented opcodes egN_eq1gain, egN_eq2gain, egN_eq3gain,
622 egN_eq1gain_onccX, egN_eq2gain_onccX, egN_eq3gain_onccX, egN_eq1freq,
623 egN_eq2freq, egN_eq3freq, egN_eq1freq_onccX, egN_eq2freq_onccX,
624 egN_eq3freq_onccX, egN_eq1bw, egN_eq2bw, egN_eq3bw, egN_eq1bw_onccX,
625 egN_eq2bw_onccX, egN_eq3bw_onccX, lfoN_eq1gain, lfoN_eq2gain,
626 lfoN_eq3gain, lfoN_eq1gain_onccX, lfoN_eq2gain_onccX, lfoN_eq3gain_onccX,
627 lfoN_eq1gain_smoothccX, lfoN_eq2gain_smoothccX, lfoN_eq3gain_smoothccX,
628 lfoN_eq1gain_stepccX, lfoN_eq2gain_stepccX, lfoN_eq3gain_stepccX,
629 lfoN_eq1freq, lfoN_eq2freq, lfoN_eq3freq, lfoN_eq1freq_onccX,
630 lfoN_eq2freq_onccX, lfoN_eq3freq_onccX, lfoN_eq1freq_smoothccX,
631 lfoN_eq2freq_smoothccX, lfoN_eq3freq_smoothccX, lfoN_eq1freq_stepccX,
632 lfoN_eq2freq_stepccX, lfoN_eq3freq_stepccX, lfoN_eq1bw, lfoN_eq2bw,
633 lfoN_eq3bw, lfoN_eq1bw_onccX, lfoN_eq2bw_onccX, lfoN_eq3bw_onccX,
634 lfoN_eq1bw_smoothccX, lfoN_eq2bw_smoothccX, lfoN_eq3bw_smoothccX,
635 lfoN_eq1bw_stepccX, lfoN_eq2bw_stepccX, lfoN_eq3bw_stepccX
636 - implemented opcodes eq1_vel2freq, eq2_vel2freq,
637 eq3_vel2freq, eq1_vel2gain, eq2_vel2gain, eq3_vel2gain
638 - sfz parser: allow double spaces in sample filenames
639 - sfz parser: allow absolute paths for sample filenames
640 - use linear decay and release for filter and pitch EG
641 - bugfix: only the first amp_veltrack definition in a file was
642 used
643 - bugfix: looping was disabled if loop_start was set to 0
644 - allow regions with end=-1 to turn off other regions using the
645 group and off_by opcodes (#168)
646 - made end=0 play the whole sample
647 - fixed support for lochan and hichan opcodes (#155)
648 - fixed crash when using lochan/hichan opcodes (#187)
649 - sfz parser: allow -200 to 200 for pan_oncc opcode (#182)
650 - added FLAC support (#191)
651 - sfz parser bugfix: lines starting with whitespace were ignored
652 - added amplitude opcode
653 - added support for "#include" instruction
654 (modified patch which was originally posted by Sergey on LS mailing list)
655 - bugfix: generation of velocity curves etc should not be done
656 after each "#include", only after the main file is parsed
657 - bugfix: line numbers in error messages were wrong after "#include"
658 - added support for float and 32 bit sample files
659
660 * SoundFont format engine:
661 - Initial implementation (not usable yet)
662 - RT-safeness: avoid malloc in audio thread
663 - fixed a bug that could cause voice stealing to fail
664 - fine-tuned amplitude EG (by switching from gig to sfz EG)
665 - initial implementation of Vibrato LFO and Modulation LFO
666 - initial implementation of cutoff filter
667 - use linear decay and release for filter and pitch EG
668
669 * Host plugins (VST, AU, LV2, DSSI):
670 - AU bugfix: failed to destroy its audio/MIDI devices
671 - Listen to all interfaces on Mac OS X (INADDR_ANY)
672 - VST bugfix: If the host called resume() before and after
673 changing sample rate or block size, the number of channels was
674 incorrectly set to two. This caused silence in Cubase 5.
675 - save engine type (gig, sfz or sf2) in plugin state
676 - VST: when opening Fantasia, look for both 32 and 64 bit Java on
677 64 bit Windows
678 - AU: changed number of output channels from one stereo to 16
679 stereo
680 - VST: made it possible to build the VST plugin for Mac
681 - AU: link AU plugin dynamically if --disable-shared isn't
682 specified
683 - LV2 "state" extension support (patch by David Robillard)
684 - VST bugfix: instrument loading hang and crashed the host when
685 the plugin was loaded a second time (#174)
686 - plugin bugfix: instrument loading hang when the plugin was
687 loaded a second time (this time it's for Linux and Mac, previous
688 similar fix was for Windows)
689 - thread safety fixes for the instrument loading thread
690 - LV2: use the new lv2 package if present
691 - VST: try to open Fantasia automatically on Linux and Mac too (on
692 Linux, the Fantasia jar should be placed in <prefix>/share/java)
693 - VST: fixed crashes on Linux Ardour and EnergyXT
694 - DSSI bugfix: it wasn't possible to change engine type. The MIDI
695 port and audio channel routing for DSSI plugins are now visible.
696 - LV2: use urid and atom extensions instead of deprecated uri-map
697 and event
698 - LV2: lv2 package 1.0 is now required to build the LV2 plugin
699 - LV2: changed number of output channels to 16 stereo, just like
700 the VST and AU plugins
701 - LV2: fixed save/restore of SFZ state (patch by David Robillard)
702 - LV2: made LV2 plugin buildable on Windows and Mac
703 - VST: implemented retrieval and switching of programs using the
704 sampler's internal MIDI instrument mapping system
705
706 * Instrument editor interface:
707 - Changed instrument editor plugin interface, providing additional
708 informations like the EngineChannel for which the instrument editor was
709 spawned for. This allows the instrument editors to interact more actively
710 with the sampler.
711
712 * MIDI driver:
713 - ALSA MIDI driver supports now "NAME" device parameter, for overriding
714 the ALSA sequencer client name
715 - removed limit of maximum amount of MIDI ports per MIDI device, since
716 there is no reason for this limit
717 - MME: fixed memory handling bug found with cppcheck
718 - MME: removed compiler warning
719 - CoreMIDI: implemented driver specific port parameter "CORE_MIDI_BINDINGS",
720 which allows to retrieve the list of CoreMIDI clients / ports and to connect
721 to them a la JACK, via the usual sampler APIs
722 - CoreMIDI: added driver specific port parameter "AUTO_BIND", if enabled
723 the driver will automatically connect to other CoreMIDI clients' ports
724 (e.g. external MIDI devices being attached to the Mac)
725 - added support for MIDI note on velocity filter
726 - CoreMIDI: fixed memory deallocation error
727 - Fixed variable underflow in VirtualMidiDevice, which caused graphical
728 virtual keyboards in frontends / instrument editors being stuck.
729 - Bugfix in VirtualMidiDevice: process note on with velocity 0 as note off.
730 - Implemented missing handling of MIDI "running status".
731 - CoreMIDI fix: a MIDIPacket can contain more than one event per packet.
732 - MME bugfix: driver wasn't closed properly
733 - CoreMIDI: automatically connect to all input sources by default (driver
734 parameter "AUTO_BIND").
735 - CoreMIDI: fixed auto bind feature to CoreMIDI ports that go online
736 - CoreMIDI: fixed minor memory leak
737
738 * audio driver:
739 - ASIO driver fixes for newer gcc versions (fix from PortAudio)
740 - JACK audio: react on sample rate changes.
741 - JACK audio: react on buffer size changes.
742 - JACK audio: jack_port_get_buffer() was cached and called outside
743 RT context.
744 - ASIO driver: removed compiler warnings
745 - CoreAudio: fixed minor error handling bug
746 - ASIO driver: be more verbose when no ASIO card could be found (fixes #203)
747 - JACK audio: return the JACK server's current sample rate as default value
748 for audio device parameter "SAMPLERATE" (fixes #166).
749
750 * LSCP server:
751 - added support for sending MIDI CC messages via LSCP command
752 "SEND CHANNEL MIDI_DATA CC <sampler-chan> <ctrl> <val>"
753 - added LSCP command "GET AVAILABLE_EFFECTS"
754 - added LSCP command "LIST AVAILABLE_EFFECTS"
755 - added LSCP command "GET EFFECT INFO <effect-index>"
756 - added LSCP command "CREATE EFFECT_INSTANCE <effect-index>"
757 - added LSCP command
758 "CREATE EFFECT_INSTANCE <effect-system> <module> <effect-name>"
759 - added LSCP command "DESTROY EFFECT_INSTANCE <effect-instance>"
760 - added LSCP command "GET EFFECT_INSTANCES"
761 - added LSCP command "LIST EFFECT_INSTANCES"
762 - added LSCP command "GET EFFECT_INSTANCE INFO <effect-instance>"
763 - added LSCP command
764 "GET EFFECT_INSTANCE_INPUT_CONTROL INFO <effect-instance> <input-control>"
765 - added LSCP command "SET EFFECT_INSTANCE_INPUT_CONTROL VALUE
766 <effect-instance> <input-control> <value>"
767 - added LSCP command "GET SEND_EFFECT_CHAINS <audio-device>"
768 - added LSCP command "LIST SEND_EFFECT_CHAINS <audio-device>"
769 - added LSCP command "ADD SEND_EFFECT_CHAIN <audio-device>"
770 - added LSCP command
771 "REMOVE SEND_EFFECT_CHAIN <audio-device> <effect-chain>"
772 - added LSCP command
773 "GET SEND_EFFECT_CHAIN INFO <audio-device> <effect-chain>"
774 - added LSCP command "APPEND SEND_EFFECT_CHAIN EFFECT <audio-device>
775 <effect-chain> <effect-instance>"
776 - added LSCP command "INSERT SEND_EFFECT_CHAIN EFFECT <audio-device>
777 <effect-chain> <effect-chain-pos> <effect-instance>"
778 - added LSCP command "REMOVE SEND_EFFECT_CHAIN EFFECT <audio-device>
779 <effect-chain> <chain-pos>"
780 - added LSCP command "SET FX_SEND EFFECT <sampler_channel>
781 <fx_send_id> <effect_chain> <chain_pos>"
782 - added LSCP command "REMOVE FX_SEND EFFECT <sampler_channel> <fx_send_id>"
783 - added LSCP commands "SUBSCRIBE EFFECT_INSTANCE_COUNT",
784 "SUBSCRIBE EFFECT_INSTANCE_INFO", "SUBSCRIBE SEND_EFFECT_CHAIN_COUNT",
785 "SUBSCRIBE SEND_EFFECT_CHAIN_INFO"
786 - provide comprehensive error messages on LSCP syntax errors
787 (suggesting expected next non-terminal symbols)
788 - Fixed client connection not being closed after network errors.
789
790 * LSCP shell:
791 - Added support for auto correction of obvious and trivial syntax mistakes.
792 - Added support for auto completion by tab key.
793 - Show currently available auto completion while typing.
794 - Added support for browsing command history with up / down keys.
795 - Show all possible next symbols immediately right to the current command
796 line while typing (no double tab required for this feature, as it would
797 be the case in other shells).
798 - Added support for moving cursor left/right with arrow keys.
799 - Added support for built-in LSCP reference documentation, which will
800 automatically show the relevant LSCP reference section on screen as soon
801 as one specific LSCP command was detected while typing on the command
802 line.
803
804 * Real-time instrument scripts:
805 - Implemented built-in script array variable %CC.
806 - Implemented built-in script int variable $CC_NUM.
807 - Implemented built-in script int variable $EVENT_NOTE.
808 - Implemented built-in script int variable $EVENT_VELOCITY.
809 - Implemented built-in script constant variable $VCC_MONO_AT.
810 - Implemented built-in script constant variable $VCC_PITCH_BEND.
811 - Implemented execution of script event handler "init".
812 - Implemented execution of script event handler "controller".
813 - Implemented execution of script event handler "note".
814 - Implemented execution of script event handler "release".
815 - Implemented built-in script function "play_note()" (only two of the
816 max. four function arguments are currently implemented yet though).
817 - Implemented built-in script int variable $EVENT_ID.
818 - Implemented built-in script function "ignore_event()"
819 - Implemented built-in script function "ignore_controller()" (may have one
820 or no argument).
821 - Implemented built-in script function "set_controller()".
822 - Added extended script VM for the Gigasampler/GigaStudio format sampler
823 engine, which extends the general instrument script VM with Giga format
824 specific variables and functions.
825 - Giga format scripts: added built-in script int constant variables
826 $GIG_DIM_CHANNEL, $GIG_DIM_LAYER, $GIG_DIM_VELOCITY, $GIG_DIM_AFTERTOUCH,
827 $GIG_DIM_RELEASE, $GIG_DIM_KEYBOARD, $GIG_DIM_ROUNDROBIN, $GIG_DIM_RANDOM,
828 $GIG_DIM_SMARTMIDI, $GIG_DIM_ROUNDROBINKEY, $GIG_DIM_MODWHEEL,
829 $GIG_DIM_BREATH, $GIG_DIM_FOOT, $GIG_DIM_PORTAMENTOTIME, $GIG_DIM_EFFECT1,
830 $GIG_DIM_EFFECT2, $GIG_DIM_GENPURPOSE1, $GIG_DIM_GENPURPOSE2,
831 $GIG_DIM_GENPURPOSE3, $GIG_DIM_GENPURPOSE4, $GIG_DIM_SUSTAIN,
832 $GIG_DIM_PORTAMENTO, $GIG_DIM_SOSTENUTO, $GIG_DIM_SOFT,
833 $GIG_DIM_GENPURPOSE5, $GIG_DIM_GENPURPOSE6, $GIG_DIM_GENPURPOSE7,
834 $GIG_DIM_GENPURPOSE8, $GIG_DIM_EFFECT1DEPTH, $GIG_DIM_EFFECT2DEPTH,
835 $GIG_DIM_EFFECT3DEPTH, $GIG_DIM_EFFECT4DEPTH, $GIG_DIM_EFFECT5DEPTH.
836 - Giga format scripts: Implemented built-in script function
837 "gig_set_dim_zone(event_id, dimension, zone)".
838 - Implemented built-in script int array variable %KEY_DOWN.
839 - Implemented built-in script function "abs()".
840 - Implemented built-in script function "random()".
841 - Implemented built-in script function "num_elements()".
842 - Implemented built-in script function "note_off()".
843 - Implemented built-in script function "set_event_mark()".
844 - Implemented built-in script function "delete_event_mark()".
845 - Implemented built-in script function "by_marks()".
846 - Added built-in script int const variables $MARK_1 to $MARK_28.
847 - Built-in script functions "ignore_event()", "note_off()" and
848 "gig_set_dim_zone()" now also accept an array of event IDs as argument
849 (i.e. return value of new script function "by_marks()").
850 - Pass/preserve polyphonic variable data from respective "note" event
851 handler to "release" event handler.
852 - Fixed crash when using built-in script function "by_marks()".
853
854 * Bug fixes:
855 - Fixed crash which may occur when MIDI key + transpose is out of range
856 - minor valgrind fixes
857 - fixed crash which occurred when changing an already deployed sampler
858 channel to a different engine type
859 - fixed crash when deleting a sampler channel or changing engine
860 type while an instrument load was in progress
861 - bugfix: playing a note while changing the instrument could cause
862 a crash, or give "Handing back unknown region" error messages
863 - bugfix: calling SET VOICES (which Fantasia does on start and
864 refresh) could cause instruments to be unloaded
865 - Mac OS X: fixed name collision of enum in EffectControl and
866 wrong name of destructor in AudioOutputDeviceCoreAudio.cpp
867 - Mac OS X: fixed hanging threads
868 - Windows: disabled the previous thread fix on non-Mac systems, as
869 it caused hanging threads on Windows
870 - Fixed possible crashes due to corrupted MIDI/audio device list
871 after MIDI/audio device creation failure
872 - When creating MIDI instrument map entries with "PERSISTENT" type, the
873 instruments were uselessly precached with zero samples, however it still
874 took the full preloading time and on 1st program change the respective
875 instrument was completely reloaded again.
876 - fixed handling of rapid bank select and program change messages
877 sent to the same sampler channel (patch from the Open Octave
878 project, slightly adjusted)
879 - fixed crash when trying to create an effect instance with controls
880 which min and/or max values depend on the sample rate
881 - fixed bug #162
882 - bugfix: LADSPA_PATH was not evaluated correctly when containing
883 multiple paths (#165)
884 - thread safety fixes for the instrument loading thread
885 - bugfix: instrument loading crashed for sfz and sf2 in Ardour
886 (#176)
887 - more thread safety fixes for the instrument loading thread
888 - sfz/sf2 engine: fixed crash when using small audio fragment size
889 - Mac OS X: fixed crash when unloading plugin on 10.7 and later
890 - Mac OS X: fixed process hang when unloading 32-bit plugin (bug
891 introduced in previous fix)
892 - fixed crash when a channel received a program change while
893 playing a note in a key group
894 - fixed erroneous error message piping in VoiceBase.h
895 (labelled "Disk stream not available in time")
896 - Update effects on sample rate & period size changes (to avoid
897 crashes and noise on such transitions).
898 - ignore missing LADSPA paths without ignoring valid LADSPA paths
899 (fixes #208)
900 - Fixed bug in LSCP grammar definition which caused a statement like
901 "GET SERVER INFOasdf\n" to be accepted as valid statement (was so far
902 practically irrelevant, however it caused problems with the new LSCP
903 shell's auto completion feature).
904 - Fixed MIDI program change messages being ignored if quickly executed
905 after each other (fixes #231).
906
907 Version 1.0.0 (31 July 2009)
908
909 * packaging changes:
910 - autoconf bugfix: the PKG_CONFIG variable wasn't initialized properly,
911 causing e.g. the libgig test to fail when
912 "./configure --disable-jack-driver" was used
913 (patch by Alexis Ballier)
914 - fixed compilation with gcc 4.3
915 - fixes for building on OS X (thanks to Ebrahim Mayat for testing)
916 - fixed configure so it detects x86_64 (#107)
917 - fixes for building with newer MinGW versions
918 - fix for building with bison 2.4 (#111)
919 - fixed building with libgig installed in a non-standard directory
920 - minor fix in configure for mmsystem.h detection on MinGW
921 - Windows: look for editor plugins and Fantasia using base
922 directory of liblinuxsampler dll (look in the same directory and one
923 directory above)
924 - configure script fix: removed unconditional use of SSE
925 - fixed building with sqlite installed in a non-standard directory
926 - when cross-compiling, don't try to create instruments.db
927 - fix for new mingw-w64 version, which has usleep
928
929 * general changes:
930 - bugfix: on some POSIX systems instrument editor plugins refused to
931 load as we used a non-portable Linux specific struct field
932 (fixes bug #70, patch by Ronald Baljeu)
933 - fixed endless loop which occured when loading an instrument editor
934 plugin DLL on Windows systems failed
935 - fixed memory leaks that occurred when liblinuxsampler was unloaded
936 - optimized the SynchronizedConfig class so it doesn't wait
937 unnecessarily long after an update
938 - added support for notifying instrument editors on note-on / note-off
939 events (e.g. to highlight the pressed keys on the virtual keyboard
940 of gigedit)
941 - added support for triggering notes by instrument editors (see above)
942 - be verbose on DLL load errors (on Linux)
943 - fixed di-harmonic triangle LFO implementation (this LFO implementation
944 is only used by very few systems, most pick the int math
945 implementation, which is usually faster)
946 - fixes for audio drivers with varying buffer sizes
947 - experimental support for running LinuxSampler as a DSSI, LV2 and
948 VST plugin
949 - notification events for stream/voice count statistics are now sent
950 only when there are actual changes
951 - added memory ordering constraints to improve stability on
952 multi-core and multi-cpu systems
953 - maximum voices and disk streams can now be altered at runtime
954 - fixed CPU feature detection on x86_64 (maybe fixes #108)
955 - automatic stacktrace mechanism is now turned off by default and can
956 be switched on by command line option "--stacktrace" (the automatic
957 stacktrace mechanism seems to be broken on most systems at the moment)
958 - C++ API method InstrumentManager::LaunchInstrumentEditor() now returns
959 a pointer to the launched InstrumentEditor object
960 - added optional 3rd party user data parameter for following
961 liblinuxsampler C++ API methods: InstrumentEditor::Main(),
962 InstrumentEditor::Launch(),
963 InstrumentManager::LaunchInstrumentEditor()
964 - theoretical fix: made SynchronizedConfig follow C++0x memory
965 model more strictly
966 - fixes for using large audio device buffers
967 - Windows: add the installation directory to the DLL search path
968 when loading an editor plugin (solves problems with VST and
969 gigedit on systems with other GTK versions installed)
970 - updated linuxsampler man page
971
972 * audio driver:
973 - removed the nonsense audio channel constraint (which was hard coded to
974 max. 100 audio channels) for most audio drivers
975 - JACK audio driver did not offer a device parameter "SAMPLERATE" as
976 opposed to the LSCP specs
977 - bugfix: the SAMPLERATE parameter of some drivers (e.g. JACK)
978 reflected the wrong value
979 - fixed a memory management bug in ASIO driver
980 - Makefile fix: JACK_CFLAGS wasn't used
981 - JACK: use jack_client_open instead of the deprecated
982 jack_client_new
983 - added (experimental) CoreAudio driver
984 - applied old fixes to the ASIO driver that were included in the
985 previous binary release but accidentally never committed to CVS
986 (fixes #117)
987 - fixes for ASIO on mingw-w64 (iasio wrapper is not needed on
988 win64)
989 - VST: added support for sample rate and buffer size changes
990 - VST: close editor (Fantasia) when the VST is removed
991 - VST: avoid opening Fantasia more than once for each VST instance
992 - VST: export main function as "main" on Linux too (fix for
993 energyXT)
994 - VST: changed number of output channels from one stereo to 16
995 stereo
996 - added channel routing, fxsends and midi maps to the settings
997 stored in the plugin state
998 - performance optimization of AudioChannel::MixTo() and
999 AudioChannel::CopyTo() methods using GCC vector exensions
1000 (if available)
1001 - ASIO fixes: avoid initializing the device twice, avoid throwing
1002 exception when getting parameters from a disconnected device
1003
1004 * MIDI driver:
1005 - added JACK MIDI driver
1006 - dispatch bank select as ordinary CC as well, the user might seriously
1007 want to (mis)use it for some purpose ("fixed" in all current MIDI
1008 input drivers: ALSA, CoreMIDI, JACK, MidiShare, MME)
1009 - bugfix: pitch bend wasn't working with JackMidi, VST, LV2, MME,
1010 CoreMidi or AU
1011 - fixed mingw-w64 compilation error in MME driver
1012 - made program change handling in MIDI thread real-time safe by
1013 moving the logic to a non-RT thread
1014 - fixed minor memory leak in ALSA driver
1015
1016 * instruments database:
1017 - avoid time consuming samples scanning when adding instruments
1018 to the instruments database
1019 - added support for handling lost files in the instruments database
1020 - Implemented option for adding instruments in separate directories
1021 in the instruments database
1022 (patch by Chris Cherrett & Andrew Williams, a bit adjusted)
1023 - work-around for missing fnmatch function on Windows to make
1024 instrument database compilable
1025 - added instrument database support on Windows
1026 works with both standalone mode and VST plugin.
1027 the instruments DB file is located in
1028 %USERPROFILE%\.linuxsampler\instruments.db which allows different
1029 databases for each windows user
1030 if no DB is present it automatically creates the .linuxsampler subdir
1031 and then creates an empty DB
1032 - fixed recursive import, which was broken on Windows
1033
1034 * Gigasampler format engine:
1035 - fixed a memory leak that could happen when a channel was deleted
1036 while notes were playing
1037 - made it possible to load an instrument even if the audio thread
1038 isn't running
1039 - added partial support for the "Controller Triggered" MIDI rule,
1040 enough for piano gigs with pedal noise samples
1041 - minor fix: only mark FX sends as being modified if really the
1042 respective FX send MIDI controller was used
1043 - added support for GM global device volume SysEx message
1044 (can be selected at compile time to either apply globally to the
1045 whole sampler [default] or only to the sampler channels that are
1046 connected to the respective MIDI input port the SysEx message arrived
1047 on)
1048 - bugfix: notes triggered at position 0 in the audio buffer were
1049 sometimes wrongly killed in the same buffer, causing no sound to
1050 be played
1051 - added support for chromatic / drumkit mode Roland GS Sysex message
1052 which is usally used to switch a part between chromatic sounds and
1053 drumkit sounds (as opposed to their common default setting of MIDI
1054 part 10 being a drumkit part and all other ones chromatic parts), in
1055 LS however one can switch between the first 16 MIDI instrument maps
1056 defined for the sampler
1057 - made it possible to create multiple sample channels even if the
1058 audio thread isn't running
1059 - fixed crash when removing channel with active voices (#116)
1060 - bugfix: on sample reference changes (instrument editor), only
1061 un-cache the respective sample if it's really not used by any
1062 sampler engine anymore
1063 - re-cache samples in case they were changed by an instrument editor,
1064 e.g. when a sample was added while playing (#82)
1065 - fixed hanging notes which occured when note-off event had the exact
1066 same time stamp as the voice's note-on event and occured both in the
1067 same audio fragment cycle (fixes bug #112)
1068 - added support for the "fine tune" and "pitch bend range"
1069 instrument-level gig parameters
1070 - fixed minor artifacts in pitch bend handling
1071 - added support for GS Reset SysEx message
1072 - allow gig files to use unlimited downward pitch shifting
1073 - added a limit check for upward pitch shifting
1074 - bugfix: sometimes, when playing a note twice fast, the second
1075 note was silent
1076 - fixed crash happening when a pitch bend event arrived at the
1077 same time a new instrument was loading
1078
1079 * LSCP server:
1080 - added new LSCP event "CHANNEL_MIDI" which can be used by frontends to
1081 react on MIDI data arriving on certain sampler channels (so far only
1082 Note-On and Note-Off events are sent via this LSCP event)
1083 - added new LSCP event "DEVICE_MIDI" which can be used by frontends to
1084 react on MIDI data arriving on certain MIDI input devices (so far only
1085 Note-On and Note-Off events are sent via this LSCP event)
1086 - added new LSCP commands: FIND LOST DB_INSTRUMENT_FILES and
1087 SET DB_INSTRUMENT FILE_PATH
1088 - added new LSCP command: SEND CHANNEL MIDI_DATA which can be used by
1089 frontends to send MIDI messages to specific sampler channel
1090 - added two additional fields to GET FILE INSTRUMENT INFO command -
1091 KEY_BINDINGS and KEYSWITCH_BINDINGS
1092 - bugfix: the bank number provided by MIDI_INSTRUMENT_INFO notifications
1093 was incorrect
1094 - Optimized the retrieval of the MIDI instrument mappings
1095 - added new LSCP command "SET VOICES" to globally alter the maximum
1096 amount of voices, added new LSCP event "GLOBAL_INFO:VOICES" which will
1097 be triggered respectively
1098 - added new LSCP command "SET STREAMS" to globally alter the maximum
1099 amount of disk streams, added new LSCP event "GLOBAL_INFO:STREAMS"
1100 which will be triggered respectively
1101 - bugfix: retry if "select" returns EINTR (this fixes a crash when
1102 a gigedit file dialog is opened)
1103 - close all connections when LSCPServer is deleted
1104 - hard close of all sockets on exit
1105 - bugfix: SET CHANNEL MIDI_INPUT_TYPE didn't work with the MME
1106 driver
1107
1108 * Bug fixes:
1109 - fixed a crash which occurs when removing a sampler channel waiting
1110 to start instrument loading after another channel
1111 - fixed a crash which occurs when removing a sampler channel with
1112 instrument loading in progress (bug #113)
1113 - fixed termination caused by uncaught exception when adding MIDI
1114 instrument with PERSISTENT load mode
1115 - fixed possible iterator invalidations when resetting the sampler
1116 - fixed memory leaks when issuing the following LSCP commands:
1117 GET AUDIO_OUTPUT_DRIVER INFO
1118 GET AUDIO_OUTPUT_DRIVER_PARAMETER INFO
1119 GET MIDI_INPUT_DRIVER INFO
1120 GET MIDI_INPUT_DRIVER_PARAMETER INFO
1121 - fixed possible compilation error when sqlite is not present
1122 - fixed orphaned pointers when setting maximum voices limit (bug #118)
1123 - fixed crash when changing the audio output device of a sampler
1124 channel with loaded instrument and start playing notes
1125 - fixed endless loop in Engine::SuspendAll() (bug #120)
1126 - fixed a low-level atomic load function that was broken on 64-bit
1127 PowerPC, which probably could cause crashes on that platform
1128 - fixed a memory management error which could cause a crash when a
1129 plugin was unloaded
1130 - bugfix: two private structs had the same name, which could cause
1131 problems if the linker chose the wrong constructor
1132 - fixed low-level ConditionServer usage bug that caused lockups on
1133 Windows
1134
1135
1136 Version 0.5.1 (6 December 2007)
1137
1138 * packaging changes:
1139 - added autoconf checks for pthread library
1140 - added autoconf check for pthread bug found on certain NPTL-enabled
1141 glibc versions (see Gentoo bug report #194076)
1142 - added autoconf checks for MS Windows
1143
1144 * general changes:
1145 - bugfix: the thread used by an editor plugin didn't die when the
1146 editor closed
1147 - bugfix: Ringbuffer.h: fill_write_space_with_null() did not zero
1148 out all the space. operator--() did not apply size_mask after
1149 decrementing the read_ptr. DEFAULT_WRAP_ELEMENTS set to 0 in
1150 order to avoid problems with the _NonVolatileReader functions.
1151 - bugfix: Stream.h: added a missing divide by BytesPerSample in
1152 GetWriteSpace(). Since this function is currently only used in
1153 the stream's qsort() compare function, it didn't trigger any
1154 bugs.
1155 - Resampler.h, Synthesizer.h: cubic interpolation now works in
1156 24bit mode too. Faster method to read 24bit words on little
1157 endian machines (x86): replaced 3 byte reads + shifts with a 1
1158 unaligned 32bit read and shift
1159 - experimental support for MS Windows (MIDI input via MME, AUDIO
1160 output via ASIO)
1161 - made handling of SIGINT signal (Ctrl-C) a bit more robust
1162 - support for monitoring the total number of active disk streams
1163 (new LSCP commands: GET TOTAL_STREAM_COUNT,
1164 SUBSCRIBE TOTAL_STREAM_COUNT, UNSUBSCRIBE TOTAL_STREAM_COUNT)
1165
1166 * AUDIO driver:
1167 - added Windows ASIO low latency audio driver
1168
1169 * MIDI driver:
1170 - added MME Windows MIDI driver
1171
1172 * LSCP server:
1173 - added support for Windows style path / filenames, however with
1174 forward slash path separators instead of backslash
1175 (i.e. "C:/foo/bar.gig")
1176 - allow naughty liblscp to send non-string device parameters within
1177 apostrophes as well
1178 - added new LSCP commands: "GET FILE INSTRUMENTS <file>",
1179 "LIST FILE INSTRUMENTS <file>" and
1180 "GET FILE INSTRUMENT INFO <file> <index>" for retrieving informations
1181 about an arbitrary instrument file on the system where the sampler is
1182 running on
1183
1184 Version 0.5.0 (15 October 2007)
1185
1186 * packaging changes:
1187 - config.h is not going to be installed along with liblinuxsampler's
1188 API header files anymore
1189 - only the API relevant header (and source) files will be exported to
1190 the Doxygen API documentation (explicitly listed in Doxyfile.in)
1191 - added completely new XCode project files for Mac OSX which is now
1192 capable to execute our autoconf environment, thus no need anymore
1193 to maintain the compile time configuration file (osx/version.h) for
1194 OSX manually (patch by Toshi Nagata)
1195 - fixed buggy boolean --enable-foo / --disable-foo configure script
1196 parameters
1197 - global.h now only covers global definitions that are needed for the
1198 C++ API header files, all implementation internal global definitions
1199 are now in global_private.h
1200 - atomic.h is not exposed to the C++ API anymore
1201 - no need to include config.h anymore for using LS's API header files
1202 - fixed warnings in API doc generation
1203
1204 * general changes:
1205 - replaced the old, confusing MIDI program change mechanism by a
1206 flexible MIDI instrument mapper which allows to map arbitrary
1207 (MIDI bank, MIDI program) pairs with arbitrary
1208 (engine type, instrument file, file index) triplets which will be
1209 loaded on the respective channel when such MIDI program change
1210 messages arrive, beside that, each entry allows to define a life-time
1211 strategy for the instrument, so the user can define whether the
1212 instrument should i.e. be permanently loaded (to be able to switch
1213 quickly among sounds) or i.e. loaded just on demand when the
1214 respective program change arrives (to safe RAM space), as well as a
1215 global volume factor for each entry, so the user can adjust the volume
1216 dynamic of his mapped instrument collection without having to modify
1217 the instrument files, also one can manage arbitrary amount of such
1218 MIDI instrument maps and assign each sampler channel individually
1219 a certain map, so that i.e. separation of normal instruments and
1220 drumkits is possible
1221 - new notification events for tracking changes to audio/MIDI devices,
1222 MIDI instrument maps, MIDI instruments, FX sends, global volume.
1223 - sampler was limited to load max. 200 instruments in the background
1224 due to a constant size RingBuffer FIFO which is now replaced by a
1225 dynamic (unlimited) size std::list FIFO
1226 - added FX sends, these allow to route audio signals to arbitrary audio
1227 output channels for being processed by external effect processors
1228 (i.e. jack-rack), the send levels are controllable via arbitrary MIDI
1229 controllers
1230 - global (sampler wide) volume can now be controlled at runtime
1231 - Implemented new, improved notification system
1232 - fixed compilation errors regarding OSX
1233 (patch by Toshi Nagata)
1234 - implemented instruments database
1235 - added support for escape sequences to the instruments database
1236 - added highly experimental support for on-the-fly instrument editing
1237 within the sampler's process (by using instrument editor plugins),
1238 you'll notice the new "Registered instrument editors:" message on
1239 startup, added a new LSCP command:
1240 "EDIT CHANNEL INSTRUMENT <sampler-channel>"
1241 to spawn a matching instrument editor for the instrument on the
1242 given sampler channel, the plugin path can be overridden at compile
1243 time with ./configure --enable-plugin-dir=/some/dir
1244 - added experimental code for synchronizing instrument editors hosted
1245 in the sampler's process to safely edit instruments while playing
1246 without a crash (hopefully) by either suspending single regions
1247 wherever possible or if unavoidable whole engine(s)
1248 - fixed several issues in fundamental "Thread" class: set scheduling
1249 policy and priority on thread level, set a minimum stack size for
1250 thread (TODO: a reasonable value yet to be tested), bugfix: non-RT
1251 threads simply inherited properties of starting thread instead of
1252 setting their own policy and priority
1253 - minor fix in our automatic stack trace mechanism on crashes, the
1254 main process did not wait for the stack trace process to finish
1255 its output
1256 - fixed some minor memory leaks
1257 - reenabled assembly features support, at the moment only for
1258 enabling a fast denormal FPU mode (x86 platforms supporting SSE2)
1259 - minor assembly fix in x86 features detection (don't use the PIC
1260 register, to avoid relocations in the text segment at runtime)
1261 - POSIX callback functions of Thread.h are hidden
1262
1263 * MIDI driver:
1264 - dispatch bank select (MSB and LSB) messages
1265
1266 * audio driver:
1267 - the ALSA audio output driver parameters now reflect the correct
1268 parameter value ranges for the respective selected sound card
1269 (patch by Till Wimmer, a bit fixed and extended)
1270
1271 * Gigasampler format engine:
1272 - if a filter is used and EG2 finishes before EG1, let the voice
1273 die when EG2 has finished (this fixes a problem with clicks and
1274 voice starvation for some gigs)
1275 - playback is no longer disabled during instrument loading
1276 - all notes playing on a channel that changes its instrument keep
1277 playing with the old instrument until they get a note off
1278 command
1279 - EG fix: a release value of zero could cause noises or crash
1280 - handle MIDI coarse tuning messages (MIDI RPN #0 MSB #2 LSB)
1281 - EG fine tuning: when attack is zero the EG starts at a level
1282 above max sustain level, which means that there is a decay phase
1283 even if sustain is 100%
1284 - more EG fixes: the level could sometimes go below zero and cause
1285 noises or crashes
1286 - minor fix of EGDecay (patch by Toshi Nagata)
1287 - fixed compiler error when --enable-override-filter-type was
1288 supplied to the configure script (fixes #46)
1289 - disk thread: queue sizes are now proportional to CONFIG_MAX_STREAMS
1290 instead of a fix value
1291 - behavior fix: on MIDI CC# 65 (portamento on / off), 126 (mono mode),
1292 127 (solo mode) only kill voices if the respective mode really
1293 changed
1294
1295 * LSCP server:
1296 - fixed compile time error for old Bison versions
1297 (i.e. v1.28 found on MacOS 10.4, patch by Toshi Nagata)
1298 - parser now supports extended ASCII character set
1299 (up to ASCII code 255, i.e. includes now umlauts and accents)
1300 - filename arguments in LSCP commands now allow to use escape
1301 sequences, that is directly literal as one of: \', \", \\, \n, \r,
1302 \f, \t, \v, or as octal ASCII code value like \132, or as hex ASCII
1303 code value like \xf2) (fixes bug #24)
1304 - the following LSCP commands now also support escape sequences for at
1305 least one of their text-based parameters (i.e. name, description):
1306 "ADD MIDI_INSTRUMENT_MAP", "MAP MIDI_INSTRUMENT",
1307 "SET MIDI_INSTRUMENT_MAP NAME", "SET FX_SEND NAME", "CREATE FX_SEND",
1308 "SET DB_INSTRUMENT_DIRECTORY NAME",
1309 "SET DB_INSTRUMENT_DIRECTORY DESCRIPTION", "SET DB_INSTRUMENT NAME",
1310 "SET DB_INSTRUMENT DESCRIPTION", "FIND DB_INSTRUMENTS",
1311 "FIND DB_INSTRUMENT_DIRECTORIES"
1312 - returns verbose syntax errors (line and column where syntax error
1313 occured, the unexpected character and the actually expected, possible
1314 character(s), the latter only if less than 5 possibilities)
1315 - made sure that LSCP syntax is not affected by gigedit locale
1316 settings
1317 - bugfix regarding strings parameter lists: all comma separated lists
1318 of strings were treated as being one string containing commas
1319 (fixes #57)
1320
1321 * Bug fixes:
1322 - fixed crash occurring on certain LSCP scripts (Bug 39)
1323 - another thread safety fix for lscp "load engine" and "set
1324 channel audio output device"
1325 - fixed a crash which occurs when reassigning the same engine
1326 on a sampler channel with connected MIDI device
1327 - fixed a crash which occurs when changing the number of ports of a MIDI
1328 device connected to a sampler channel to number less then or equal
1329 to the index of the port to which the sampler channel is connected.
1330 - The previous bindings were not been disconnected when altering
1331 the ALSA_SEQ_BINDINGS parameter. Introduced a NONE keyword for
1332 unsubscribing from all bindings (e.g. ALSA_SEQ_BINDINGS=NONE).
1333 - The active stream/voice count statistic was incorrect.
1334 - notification events were not been sent for some sampler
1335 channel changes
1336 - added default min and max values to restrict the number of allowed
1337 audio output channels and MIDI input ports
1338 - the connection to the PCM interface is now closed when destroying
1339 an audio output device
1340 - files with slash in their path or filename could not be loaded
1341
1342 * test cases:
1343 - updated and fixed (haven't been touched in a while)
1344
1345 Version 0.4.0 (24 November 2006)
1346
1347 * packaging changes:
1348 - changed deprecated copyright attribute to license; added ldconfig
1349 to post-(un)install steps; added devel package for liblinuxsampler;
1350 to linuxsampler.spec (RPM)
1351 - install necessary development header files for allowing 3rd party
1352 applications to link against liblinuxsampler
1353 - liblinuxsampler's API documentation can be generated with 'make docs'
1354 (Doxygen required)
1355 - added benchmark to automatically detect the best triangle LFO
1356 implementation (currently either an integer math solution or a
1357 di-harmonic approximation), automatic detection can be overriden
1358 with --enable-signed-triang-algo=x and --enable-unsigned-triang-algo=x
1359 configure script argument though (mandatory for cross-compilation)
1360 - do not automatically pick optimized gcc flags if the user already
1361 provided some on his own (as CXXFLAGS)
1362 - added compile time option to disable processing of All-Notes-Off MIDI
1363 messages
1364 - added compile time options to allow disabling the various audio and
1365 MIDI drivers
1366 - fixed automatic GCC CPU switch detection on PPC
1367 (patch by Ebrahim Mayat)
1368
1369 * Gigasampler format engine:
1370 - extensive synthesis optimization
1371 (reimplementation of EGs and LFO(s), removed synthesis parameter
1372 prerendering and the synthesis parameter matrix in general, splitting
1373 each audio fragment into subfragments now where each subfragment uses
1374 constant pitch and filter coefficients. The volume coefficient is
1375 linearly interpolated inside a subfragment, unless
1376 --disable-interpolate-volume is set.)
1377 - fine tuning of the EG modulation parameters
1378 - improved filter cutoff calculation by adding support for the
1379 following gig parameters: Cutoff freq (used when no cutoff
1380 controller is defined), Control invert, Minimum cutoff, Velocity
1381 curve and Velocity range. The keyboard tracking now scales
1382 cutoff frequency instead of resonance.
1383 - added support for gig parameter Resonance.
1384 - fixed bug in sysex handling (patch by Juan Linietsky)
1385 - added global volume attenuation of -9 dB (0.35f) to prevent clipping
1386 which can be overridden with --enable-global-attenuation
1387 - EG fixes: made the length of "attack hold" stage more
1388 accurate. Release stage can now start before attack stage
1389 ends. Cancel release didn't work when sustain was zero. Attack
1390 time now has a minimal value to prevent clicks.
1391 - fixed pitch changes larger than one octave
1392 - fixed EG3 (pitch envelope) synthesis which was neutral all the time
1393 - implemented portamento mode and solo mode (a.k.a 'mono mode'):
1394 all modes can be altered via standard GM messages, that is CC5 for
1395 altering portamento time, CC65 for enabling / disabling portamento
1396 mode, CC126 for enabling solo mode and CC127 for disabling solo mode
1397 - fine tuning of the curves for volume (CC7), pan (CC10 and gig
1398 parameter) and crossfade
1399 - added support for the "attenuation controller threshold" gig
1400 parameter
1401 - added smoothing of volume changes caused by control change
1402 messages
1403 - sample loop parameters are now taken from the DimensionRegion
1404 instead of the wave chunk
1405 - fixed keyswitching for v3 gigs with a number of keyswitch splits
1406 not equal to a power of two
1407 - reimplementation of the filter algorithm. The new filter is more
1408 accurate and supports all gig filter types, including bandreject
1409 and lowpass turbo.
1410 - real support for 24 bit samples - samples are not truncated to
1411 16 bits anymore
1412 - support for aftertouch (channel pressure, not polyphonic
1413 aftertouch)
1414 - LFO1 behaviour fixed (dampening from max volume instead of
1415 amplifying from 0)
1416
1417 * LSCP server:
1418 - fixed application exit on broken pipe error (fixes bug #20)
1419 - fixed the notification messages delay due to lack of
1420 network activity (fixes bug #26)
1421 - fixed parser bug which occured on space(s) within device parameters
1422 (e.g. "SET AUDIO_OUTPUT_CHANNEL_PARAMETER 0 0 SOMEPARAM='foo bar'")
1423
1424 * audio driver:
1425 - added aRts audio output driver (by no means RT safe)
1426
1427 * MIDI driver:
1428 - fixed legacy sysex code which caused dispatching of MIDI SysEx
1429 messages several times instead of once
1430 - API extension for MIDI drivers which already supply exact time stamps
1431 for events (i.e. for offline rendering based MIDI drivers)
1432 - added checks for bad MIDI data
1433
1434 * linuxsampler application:
1435 - show available sampler engine types on startup
1436
1437 * general changes:
1438 - support for muting sampler channels and solo mode of the same, two new
1439 LSCP commands ("SET CHANNEL MUTE" and "SET CHANNEL SOLO") and two new
1440 fields ("MUTE" and "SOLO") for command "GET CHANNEL INFO" were
1441 introduced for this, the behavior is the same like on a mixer console
1442 (patch by Grigor Iliev, a bit adjusted). Also added configure option
1443 --enable-process-muted-channels which can be used to enable the
1444 processing of muted channels.
1445 - support for sostenuto pedal
1446 - support for monitoring the total number of active voices
1447 (new LSCP commands: GET TOTAL_VOICE_COUNT, GET TOTAL_VOICE_COUNT_MAX,
1448 SUBSCRIBE TOTAL_VOICE_COUNT, UNSUBSCRIBE TOTAL_VOICE_COUNT)
1449 - fixed some memory management errors
1450 - fixed some concurrency problems that could lead to crashes when
1451 LSCP commands were executed
1452 - fixed crash when instrument loading failed previously on the same
1453 sampler channel (fixes bug #36)
1454
1455 Version 0.3.3 (15 July 2005)
1456
1457 * packaging changes:
1458 - fixed compilation with gcc 4.0
1459
1460 * Gigasampler format engine:
1461 - LFO preprocessing was performed even though the respective LFO was
1462 disabled by the instrument patch (minor efficiency fix)
1463 - if period time of chosen audio device is too small (< MIN_RELEASE_TIME)
1464 for volume ramp downs in the same fragment (needed for current voice
1465 stealing implementation) then simply inform the user with a warning
1466 about possible click sounds and reduce the volume ramp down
1467 appropriately instead of cancelling the audio device connection
1468
1469 Version 0.3.2 (24 June 2005)
1470
1471 * packaging changes:
1472 - updated autotools build files to compile on OS X
1473 (conditional compilation of CoreMIDI and MidiShare drivers)
1474 - hand-crafted assembly optimization code can be disabled with
1475 './configure --disable-asm' (definitely not recommended)
1476 - fixed 'make dist' rule to include all necessary files
1477 - require automake (>= 1.5) for 'make -f Makefile.cvs'
1478 (due to 'dist-bzip2' automake option)
1479
1480 * Gigasampler format engine:
1481 - support for the gig parameters for "release velocity response" curves
1482 - fine tuning of the EGADSR envelope
1483 - volume of release triggered samples now depends on note-on velocity,
1484 note length and gig parameter "release trigger decay" instead of
1485 note-off velocity.
1486 - revised voice stealing
1487 (fixes crash and endless loop caused by voice stealing)
1488 - don't reset scale tuning on instrument or audio output device change
1489 - handle key group conflicts right at the beginning of each fragment
1490 instead of when the respective voice is actually launched
1491 (fixes undefined behavior if stolen voices belonged to a key group -
1492 this case was followed by a "killed voice survived" error message)
1493 - fixed minor issue with null/silence samples
1494 (those stole voices even though they don't need a voice at all which
1495 resulted in "voice stealing didn't work out" messages)
1496 - don't reset volume, pan, pitch and MIDI controller values on
1497 instrument or audio output device change
1498
1499 * LSCP server:
1500 - fixed some crashes (patch by Grigor Iliev, fixes #19)
1501 - fixed LSCP event "CHANNEL_INFO" notification
1502 (e.g. did not notify on volume changes or MIDI program change events)
1503
1504 * linuxsampler application:
1505 - added command line parameters --lscp-addr and --lscp-port to override
1506 default IP address and port of LSCP server
1507
1508 Version 0.3.1 (24 May 2005)
1509
1510 * initial release

  ViewVC Help
Powered by ViewVC