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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 3487 - (show annotations) (download)
Wed Feb 27 14:50:36 2019 UTC (17 months, 2 weeks ago) by schoenebeck
File size: 78107 byte(s)
* SFZ: Fixed memory leak when releasing samples
  (fixes bug #307, patch by Jacek Roszkowski)
* Bumped version (2.1.0.svn5).

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

  ViewVC Help
Powered by ViewVC