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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 3015 - (show annotations) (download)
Tue Oct 18 20:13:08 2016 UTC (7 years, 5 months ago) by schoenebeck
File size: 68404 byte(s)
* gig/sf2/sfz: Fixed aftertouch channel pressure events being processed
  properly and thus were ignored for certain uses.
* Bumped version (2.0.0.svn28).

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

  ViewVC Help
Powered by ViewVC