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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 3228 - (show annotations) (download)
Sun May 28 14:46:14 2017 UTC (6 years, 10 months ago) by schoenebeck
File size: 72262 byte(s)
* Implemented built-in script function "same_region()" (currently only
  available for gig format engine).
* Bumped version (2.0.0.svn53).

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

  ViewVC Help
Powered by ViewVC