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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 3054 - (show annotations) (download)
Thu Dec 15 12:47:45 2016 UTC (7 years, 3 months ago) by schoenebeck
File size: 68760 byte(s)
* Fixed numerous compiler warnings.
* Bumped version (2.0.0.svn32).

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

  ViewVC Help
Powered by ViewVC