ViewVC logotype

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 3308 - (show annotations) (download)
Sat Jul 15 01:14:20 2017 UTC (3 years ago) by schoenebeck
File size: 75777 byte(s)
* NKSP script editor API: Fixed app termination due to a lexer start
  condition stack underrun.
* NKSP preprocessor: Fixed wrong behavior on nested USE_CODE_IF() and
  USE_CODE_IF_NOT() preprocessor statements.
* Bumped version (2.0.0.svn70).

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

  ViewVC Help
Powered by ViewVC