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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 3514 - (show annotations) (download)
Sat May 25 17:22:08 2019 UTC (4 years, 10 months ago) by schoenebeck
File size: 78628 byte(s)
* "optional" class: Fixed comparison operators.
* Bumped version (2.1.0.svn9).

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

  ViewVC Help
Powered by ViewVC