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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 3050 - (show annotations) (download)
Thu Dec 8 12:36:56 2016 UTC (7 years, 3 months ago) by schoenebeck
File size: 68721 byte(s)
* Instruments DB: Fixed memory access bug of general DB
  access code which lead to undefined behavior.

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

  ViewVC Help
Powered by ViewVC