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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 2927 - (show annotations) (download)
Thu Jun 30 16:44:46 2016 UTC (7 years, 8 months ago) by schoenebeck
File size: 64287 byte(s)
* Fix: Release trigger voices were not spawned on sustain pedal up
  (CC #64) events.
* Fix: Release trigger voices can now also distinguish correctly between
  note-off and sustain pedal up events.
* Bumped version (2.0.0.svn10).

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

  ViewVC Help
Powered by ViewVC