ViewVC logotype

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 2351 - (show annotations) (download)
Sat Jun 23 11:02:48 2012 UTC (8 years, 3 months ago) by persson
File size: 49242 byte(s)
* Mac OS X: fixed process hang when unloading 32-bit plugin (bug
  introduced in previous fix)

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

  ViewVC Help
Powered by ViewVC