ViewVC logotype

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 3381 - (show annotations) (download)
Tue Nov 28 15:54:49 2017 UTC (3 years ago) by schoenebeck
File size: 77558 byte(s)
* NKSP: Fixed behavior of built-in functions change_sustain(),
  change_cutoff_attack(), change_cutoff_decay(), change_cutoff_sustain()
  and change_cutoff_release().
* Bumped version (2.1.0.svn1).

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

  ViewVC Help
Powered by ViewVC