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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 3550 - (show annotations) (download)
Thu Aug 1 09:56:27 2019 UTC (4 years, 7 months ago) by schoenebeck
File size: 79019 byte(s)
- Just minor cleanup of ChangeLog.

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

  ViewVC Help
Powered by ViewVC