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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 3073 - (show annotations) (download)
Thu Jan 5 16:04:00 2017 UTC (7 years, 2 months ago) by schoenebeck
File size: 69091 byte(s)
* NKSP: Implemented built-in script array variable "%ALL_EVENTS".
* Bumped version (2.0.0.svn35).

1 Version SVN trunk (?)
2
3 * SFZ format engine:
4 - added support for <global>, <master> and #define (patch by Alby M)
5
6 * Gigasampler format engine:
7 - Fixed clicks and pumping noise with Lowpass Turbo filter on very low
8 cutoff settings.
9
10 * general changes:
11 - fixed printf type errors (mostly in debug messages)
12 - use unique_ptr instead of auto_ptr when building with C++11
13 - Added RTAVLTree class which is a real-time safe ordered multi-map, thus
14 allowing to sort data efficiently in real-time safe manner.
15 - RTList class: added methods for moving/inserting elements to arbitrary
16 position within a list.
17 - RTList class: added method fromPtr() for creating an Iterator object from
18 a raw element pointer.
19 - Attempt to partly fix resetting engine channels vs. resetting engine, an
20 overall cleanup of the Reset*(), ConnectAudioDevice(),
21 DisconnectAudioDevice() API methods would still be desirable though,
22 because the current situation is still inconsistent and error prone.
23 - Active voices are now internally grouped to "Note" objects, instead of
24 being directly assigned to a keyboard key. This allows more fine graded
25 processing of voices, which is i.e. required for certain instrument
26 script features.
27 - Fix: Release trigger voices were not spawned on sustain pedal up
28 (CC #64) events.
29 - Fix: Release trigger voices can now also distinguish correctly between
30 note-off and sustain pedal up events.
31 - All engines: Increased ramp speed of volume smoother and pan smoother
32 (while slow rate of crossfade smoother is preserved) to allow quick
33 volume and pan changes by instrument scripts for instance.
34 - gig/sf2/sfz: Fixed aftertouch channel pressure events being processed
35 properly and thus were ignored for certain uses.
36 - Added new C++ API method
37 VirtualMidiDevice::SendChannelPressureToSampler().
38 - windows, 32-bit: fixed potential crashes by making sure the stack in
39 sub threads is 16-byte aligned
40 - fixed numerous compiler warnings
41
42 * packaging changes:
43 - removed unnecessary dependency to libuuid
44 (originated by libgig's usage of it)
45 - Automake: set environment variable GCC_COLORS=auto to allow GCC to
46 auto detect whether it (sh/c)ould output its messages in color.
47
48 * Real-time instrument scripts:
49 - Implemented scheduler for delayed MIDI events and for suspended scripts.
50 - Built-in script function "wait()": implemented support for function's
51 "duration-us" argument, thus scripts using this function are now
52 correctly resumed after the requested amount of microseconds.
53 - Built-in script function "play_note()": implemented support for
54 function's "duration-us" argument, thus notes triggered with this
55 argument are now correctly released after the requested amount of
56 microseconds.
57 - Fix: script events were not cleared when engine channel was reset,
58 potentially causing undefined behavior.
59 - Fixed crash which happened when trying to reference an undeclared
60 variable.
61 - Built-in script function "play_note()": Added support for passing
62 special value -1 for "duration-us" argument, which will cause the
63 triggered note to be released once the original note was released.
64 - Instrument script classes now exported with the liblinuxsampler C++ API.
65 - Added new API method ScriptVM::syntaxHighlighting() which provides
66 a convenient syntax highlighting backend for external instrument
67 script editor applications.
68 - Added new C++ API class "ScriptVMFactory".
69 - Extended parser issues to provide not only first line and first
70 column, but also last line and last column of issue (thus marking
71 the precise span of the issue within the source code).
72 - Reload script automatically after being modified by an instrument
73 editor.
74 - NKSP language grammar correction: allow empty event handler bodies
75 like "on note end on".
76 - Implemented built-in script function "change_vol()".
77 - Implemented built-in script function "change_tune()".
78 - Implemented built-in script function "change_pan()".
79 - Implemented built-in script function "change_cutoff()".
80 - Implemented built-in script function "change_reso()".
81 - Implemented built-in script function "event_status()".
82 - Added built-in script constants "$EVENT_STATUS_INACTIVE" and
83 "$EVENT_STATUS_NOTE_QUEUE" both for being used as flags for
84 "event_status()" function.
85 - NKSP language: Added support for bitwise operators ".or.", ".and."
86 and ".not.".
87 - NKSP language scanner: Fixed IDs matching to require at least one
88 character (i.e. when matching function names or variable names).
89 - NKSP language scanner: disabled unusued rules.
90 - Fixed behavior of play_note() and note_off() functions which must
91 be distinguished engine internally from "real" MIDI note on/off
92 events in order to avoid misbehaviors like hanging notes.
93 - Implemented built-in script variable "$KSP_TIMER".
94 - Implemented built-in script variable "$NKSP_REAL_TIMER".
95 - Implemented built-in script variable "$NKSP_PERF_TIMER".
96 - Implemented built-in script variable "$ENGINE_UPTIME".
97 - Implemented built-in script function "inc()".
98 - Implemented built-in script function "dec()".
99 - NKSP language fix: division expressions were evaluated too often.
100 - NKSP language fix: string concatenation operator was right
101 associative instead of left (to right).
102 - NKSP language correction: allow empty statements in entire language
103 (i.e. "if end if").
104 - Implemented built-in script function "stop_wait()".
105 - Implemented built-in script variable "$NI_CALLBACK_ID".
106 - Implemented built-in script variable "$NI_CALLBACK_TYPE".
107 - Implemented built-in script variable "$NKSP_IGNORE_WAIT".
108 - Added support for read-only built-in variables (respectively
109 handled by the script parser).
110 - Added built-in script constant "$NI_CB_TYPE_INIT".
111 - Added built-in script constant "$NI_CB_TYPE_NOTE".
112 - Added built-in script constant "$NI_CB_TYPE_RELEASE".
113 - Added built-in script constant "$NI_CB_TYPE_CONTROLLER".
114 - NKSP Language: Added support for user defined script functions.
115 - Implemented built-in script function "change_attack()".
116 - Implemented built-in script function "change_decay()".
117 - Implemented built-in script function "change_release()".
118 - Fixed all change_*() built-in script functions to apply their
119 synthesis parameter changes immediately in case the respective note
120 was triggered at the same time, instead of scheduling the parameter
121 change, especially because it would cause some parameter types's
122 changes either to be ramped (i.e. change_vol()) or other types even
123 to have not effect at all (i.e. change_attack()).
124 - Implemented built-in script function "sh_left()".
125 - Implemented built-in script function "sh_right()".
126 - Implemented built-in script function "min()".
127 - Implemented built-in script function "max()".
128 - NKSP Fix: "init" event handler was not always executed when a script
129 was loaded.
130 - NKSP built-in wait() script function: abort script execution if
131 a negative or zero wait time was passed as argument (since this is
132 a common indication of a bug either of the script or even of the
133 engine, which could lead to RT instability or even worse).
134 - ScriptVM: Implemented automatic suspension of RT safety
135 threatening scripts.
136 - Provide more user friendly error messages on syntax errors.
137 - Fixed NKSP parser warning "Not a statement" when assigning an
138 initializer list to an array variable.
139 - Implemented built-in script array variable "%ALL_EVENTS".
140
141 * Instruments DB:
142 - Fixed memory access bug of general DB access code which lead to
143 undefined behavior.
144
145 Version 2.0.0 (15 July 2015)
146
147 * packaging changes:
148 - fixed building with newer MinGW-w64
149 - Mac OS X: support the new dir for Core Audio SDK
150 - Mac OS X: fixed building outside source directory
151 - made sure all source files for hostplugins are included when
152 doing "make dist"
153 - removed empty directories left from the cvs to svn migration
154 - fixed compilation with gcc 4.6.1
155 - another "make dist" fix, for the LV2 plugin
156 - made --enable-pthread-testcancel default on Mac OS X
157 - minor Makefile fix for building lscpparser when build and source
158 directories are separate
159 - Mac OS X: made it possible to specify plugin installation dir to
160 configure
161 - Mac OS X: Makefile fix for the install-strip target
162 - fixed compilation with gcc 4.7
163 - fixed configure script error with old autoconf versions
164 - lsatomic.h: use gcc provided atomic functions if building with
165 gcc 4.7 and C++11
166 - modernized configure script
167 - fixed linkage error when building with
168 LDFLAGS="-Wl,--no-undefined" (#190)
169 - fixed compilation with Clang 3.2
170 - removed usage of deprecated Automake variable INCLUDES
171 - fixed building with C++11
172 - build fix: ChangeFlagRelaxed.h was missing in makefile
173 - build fix: libsndfile compiler flags were missing in some
174 makefiles
175 - fix for building with bison 3.0 (#202)
176 - Mac OS X: added temporary hack allowing to spawn gigedit as callback
177 on the process's main thread
178 - fixed build error on newer MinGW
179 - support building with older jack versions
180 - support building with spaces in vst sdk path
181 - enabled automake 'subdir-objects' option and moved external
182 source references (vst, au, asio) from makefiles to cpp files,
183 in order to get rid of warnings from automake 1.14
184
185 * general changes:
186 - Refactoring: moved the independent code from
187 the Gigasampler format engine to base classes
188 - command line option '--profile' is currently disabled, since the
189 respective profiling code is currently broken
190 - Introduced support for optional environment variable
191 "LINUXSAMPLER_PLUGIN_DIR" which allows to override the directory
192 where the sampler shall look for instrument editor plugins
193 (patch by Luis Garrido, slightly modified).
194 - implemented Roland GS NRPN 1ArrH which allows to set volume per note
195 - implemented Roland GS NRPN 1CrrH which allows to set pan per note
196 - implemented Roland GS NRPN 1DrrH which allows to set reverb send per
197 note (in this implementation of the sampler its simply hard routed to
198 the 1st effect send of the sampler channel, no matter what the actual
199 effect type is)
200 - implemented Roland GS NRPN 1ErrH which allows to set chorus send per
201 note (in this implementation of the sampler its simply hard routed to
202 the 2nd effect send of the sampler channel, no matter what the actual
203 effect type is)
204 - implemented support for internal LADSPA effects
205 - raised limit of program change queue from 100 to 512
206 (as suggested by Alex Stone)
207 - implemented sine LFO, pulse LFO and saw LFO
208 - experimental support for per voice equalization
209 - added command line option --exec-after-init
210 - Introduced new C++ API method:
211 EngineChannel::InstrumentFileName(int index)
212 allowing to retrieve the whole list of files used for the loaded
213 instrument on an engine channel (a.k.a. part). Some GigaStudio
214 instruments for example are splitted over several files like
215 "Foo.gig", "Foo.gx01", "Foo.gx02", ...
216 - Added new C++ API method Sampler::GetGlobalMaxVoices().
217 - Added new C++ API method Sampler::GetGlobalMaxStreams().
218 - Added new C++ API method Sampler::SetGlobalMaxVoices().
219 - Added new C++ API method Sampler::SetGlobalMaxStreams().
220 - Various "const" and "restrict" optimizations.
221 - all engines: add pan CC value to instrument pan parameter before
222 applying panning, instead of using two separate pan functions in
223 series (#182)
224 - added a lock guard class for exception safe mutex handling and
225 used it everywhere appropriate
226 - Immediately apply scale tuning changes to active voices.
227 - Exposed scale tuning to C++ API (along to the already existing standard
228 SysEx way).
229 - lsatomic.h fixes: seq_cst load and store were suboptimal for x86
230 and broken for ppc64. (Seq_cst loads and stores are actually not
231 used in LS, so the bug wasn't noticable.)
232 - lsatomic.h: added ARMv7 support
233 - Added support for multiple MIDI input ports per sampler channel (and
234 added various new C++ methods for this new feature / design change, old
235 C++ API methods for managing SamplerChannel's MIDI inputs are now marked
236 as deprecated but are still there and should provide full behavior
237 backward compatibility).
238 - AbstractEngine::GSChecksum(): don't allocate memory on the stack (was
239 unsafe and caused compilation error with clang 2.x).
240 - Bugfix: only process the latest MIDI program change event.
241 - Introducing the LSCP shell, which provides convenient control of the
242 sampler from the command line by providing LSCP aware features.
243 - VirtualMidiDevice: Added support for program change.
244 - VirtualMidiDevice: Added support for bank select (MSB & LSB).
245 - VirtualMidiDevice: Added support for pitch bend.
246 - Aftertouch: extended API to explicitly handle channel pressure and
247 polyphonic key pressure events (so far polyphonic pressure was not
248 supported at all, and channel pressure was rerouted as CC128 but not
249 used so far).
250 - Added initial support for real-time instrument scripts. The script VM
251 code is shared by all sampler engine implemementations, however only the
252 gig file format currently provides support for storing instrument scripts
253 (as LinuxSampler extension to the original GigaStudio 4 format).
254
255 * Gigasampler format engine:
256 - implemented the "round robin keyboard" dimension
257 - fixed round robin and random dimensions for cases when number of
258 dimension zones is not a power of two
259 - made round robin use a counter for each region instead of each
260 key
261 - bugfix: pitch LFO controller "internal+aftertouch" was broken
262 - bugfix: filter keyboard tracking was broken
263 - filter performance fix (an unnecessary copy was made of the
264 filter parameters in each sub fragment)
265 - handle special case when pan parameter in gig file has max or
266 min value
267 - Exclusive Groups: don't ever stop voices of the same note,
268 doesn't sound naturally with a drumkit
269 - fixed EG1 modulation when attack or release is zero
270 - Fixed support for 'aftertouch' attenuation controller.
271 - Fixed crash that happened with velocity split sounds under certain
272 conditions (see also previous commit on libgig).
273 - fixed behaviour of filter LFO
274
275 * SFZ format engine:
276 - Initial implementation (not usable yet)
277 - added support for v2 multiple stage envelope generators
278 - added a fine-tuned v1 envelope generator instead of using the
279 one from the gig engine
280 - fixed memory leak and memory handling errors
281 - added support for trigger=first, trigger=legato and sw_previous
282 - allow non-numerical key values ("C#4" for example)
283 - "key" opcode now sets pitch_keycenter too
284 - fixed error when unloading instrument with same sample used by
285 multiple regions
286 - added some opcode aliases, like loopmode for loop_mode, to be
287 more compatible
288 - added support for trigger=release and rt_decay
289 - added support for off_mode=normal
290 - added support for random, seq_position, seq_length and volume
291 - added v1 LFO opcodes to parser (no support in engine yet)
292 - added support for amp_veltrack and amp_velcurve_N
293 - fine-tuned the default velocity curve
294 - added support for transpose
295 - fixed crash when using sw_down/up
296 - improved logic for sw_lokey/hikey/up/down/last
297 - added more v1 aliases to parser
298 - reduced memory usage for sfz data
299 - RT-safeness: avoid malloc in audio thread
300 - fixed a bug that could cause voice stealing to fail
301 - optimized sample lookup
302 - improved support for exclusive groups (group, off_by and
303 off_mode)
304 - added support for controller triggered regions (on_locc/on_hicc)
305 - added support for loop_mode=one_shot
306 - fixed playback of 16 bit wav files on big endian CPUs
307 - added support for Ogg Vorbis sample files
308 - use loop markers from sample file if loop_start and loop_end are
309 not set in sfz file
310 - implemeted filters. Filter types: lowpass, bandpass, bandreject
311 and highpass. 1, 2, 4 and 6 pole filters. Opcodes: fil_type,
312 cutoff, resonance, fil_veltrack, fil_keytrack, fil_keycenter,
313 cutoff_cc, cutoff_chanaft.
314 - bugfix: zero ampeg_sustain didn't work
315 - added support for velocity effect on amplifier envelope time
316 (ampeg_vel2attack, ampeg_vel2decay, ampeg_vel2sustain and
317 ampeg_vel2release)
318 - added support for EG hold (ampeg_hold)
319 - added support for sample offset (offset)
320 - use the newly introduced signal units model
321 - implemented opcodes lfoN_delay,
322 lfoN_freq, lfoN_pan, lfoN_cutoff, lfoN_resonance
323 - implemented opcodes ampeg_delay, ampeg_vel2delay
324 pitcheg_delay, pitcheg_start, pitcheg_attack, pitcheg_hold,
325 pitcheg_decay, pitcheg_sustain, pitcheg_release, pitcheg_vel2delay,
326 pitcheg_vel2attack, pitcheg_vel2hold, pitcheg_vel2decay,
327 pitcheg_vel2sustain, pitcheg_vel2release, pitcheg_depth
328 - implemented opcodes pitchlfo_delay, pitchlfo_freq,
329 pitchlfo_depth, fillfo_delay, fillfo_freq, fillfo_depth,
330 amplfo_delay, amplfo_freq, amplfo_depth
331 - implemented opcodes fileg_delay, fileg_start, fileg_attack,
332 fileg_hold, fileg_decay, fileg_sustain, fileg_release, fileg_vel2delay,
333 fileg_vel2attack, fileg_vel2hold, fileg_vel2decay, fileg_vel2sustain,
334 fileg_vel2release
335 - implemented opcode lfoN_wave
336 - implemented opcode pitchlfo_depthccN
337 - implemented opcodes lfoN_phase, lfoN_phase_onccX,
338 lfoN_pitch, lfoN_pitch_onccX
339 - implemented opcodes pitchlfo_fade,
340 fillfo_fade, amplfo_fade, lfoN_fade, lfoN_fade_onccX
341 - implemented opcodes fillfo_freqccN,
342 pitchlfo_freqccN, amplfo_freqccN, lfoN_freq_onccX
343 - implemented opcodes ampeg_delayccN, ampeg_startccN,
344 ampeg_attackccN, ampeg_holdccN, ampeg_decayccN, ampeg_sustainccN,
345 ampeg_releaseccN, egN_timeX_onccY, egN_levelX_onccY
346 - lfoN_* and egN_* opcodes defined in group sections
347 are now taken into account
348 - implemented curves
349 - implemented opcodes volume_onccN, volume_curveccN
350 - implemented opcode volume_smoothccN
351 - implemented opcodes fillfo_depthccN, amplfo_depthccN,
352 lfoN_volume, lfoN_volume_onccX, lfoN_volume_smoothccX,
353 lfoN_freq_smoothccX, lfoN_pitch_smoothccX, lfoN_pan_onccX,
354 lfoN_pan_smoothccX, lfoN_cutoff_onccX, lfoN_cutoff_smoothccX,
355 lfoN_resonance_onccX, lfoN_resonance_smoothccX, lfoN_delay_onccX
356 - implemented opcode end
357 - implemented opcodes egN_amplitude_onccX,
358 egN_volume, egN_volume_onccX, egN_cutoff, egN_cutoff_onccX,
359 egN_pitch, egN_pitch_onccX, egN_resonance, egN_resonance_onccX
360 - implemented opcodes xfin_lokey, xfin_hikey,
361 xfout_lokey, xfout_hikey, xf_keycurve, xfin_lovel, xfin_hivel,
362 xfout_lovel, xfout_hivel, xf_velcurve, xfin_loccN, xfin_hiccN,
363 xfout_loccN, xfout_hiccN, xf_cccurve
364 - implemented opcodes pan_onccN, pan_smoothccN,
365 pan_curveccN, egN_pan, egN_pan_curve, egN_pan_onccX,
366 egN_pan_curveccX, pitch_veltrack
367 - when failed to parse a sfz file
368 print the line number on which the error occurs
369 - use common pool of CC objects to minimize RAM usage
370 - implemented opcodes amplfo_delay_onccN,
371 amplfo_fade_onccN, fillfo_delay_onccN, fillfo_fade_onccN,
372 pitchlfo_delay_onccN, pitchlfo_fade_onccN
373 - implemented opcodes fileg_delay_onccN,
374 fileg_start_onccN, fileg_attack_onccN, fileg_hold_onccN,
375 fileg_decay_onccN, fileg_sustain_onccN, fileg_release_onccN,
376 fileg_depth_onccN, pitcheg_delay_onccN, pitcheg_start_onccN,
377 pitcheg_attack_onccN, pitcheg_hold_onccN, pitcheg_decay_onccN,
378 pitcheg_sustain_onccN, pitcheg_release_onccN, pitcheg_depth_onccN
379 - implemented automatic aliasing *ccN <-> *_onccN
380 - *lfo_freqccN wasn't working when the respective *lfo_freq
381 was not set or was set to zero
382 - lfoN_freq_onccX wasn't working when lfoN_freq
383 was not set or was set to zero
384 - implemented opcodes resonance_onccN, resonance_smoothccN,
385 resonance_curveccN, cutoff_smoothccN, cutoff_curveccN
386 - implemented opcodes fillfo_depthchanaft,
387 fillfo_freqchanaft, amplfo_depthchanaft, amplfo_freqchanaft,
388 pitchlfo_depthchanaft, pitchlfo_freqchanaft
389 - implemented opcodes pitch_onccN,
390 pitch_curveccN, pitch_smoothccN, pitch_stepccN
391 - implemented opcodes volume_stepccN, pan_stepccN,
392 cutoff_stepccN, resonance_stepccN, lfoN_freq_stepccX,
393 lfoN_volume_stepccX, lfoN_pitch_stepccX, lfoN_pan_stepccX,
394 lfoN_cutoff_stepccX, lfoN_resonance_stepccX
395 - implemented opcodes eq1_freq, eq2_freq, eq3_freq,
396 eq1_freqccN, eq2_freqccN, eq3_freqccN, eq1_bw, eq2_bw, eq3_bw,
397 eq1_bwccN, eq2_bwccN, eq3_bwccN, eq1_gain, eq2_gain, eq3_gain,
398 eq1_gainccN, eq2_gainccN, eq3_gainccN
399 - implemented opcodes delay, delay_onccN, delay_random,
400 delay_samples, delay_samples_onccN
401 - implemented opcodes egN_eq1gain, egN_eq2gain, egN_eq3gain,
402 egN_eq1gain_onccX, egN_eq2gain_onccX, egN_eq3gain_onccX, egN_eq1freq,
403 egN_eq2freq, egN_eq3freq, egN_eq1freq_onccX, egN_eq2freq_onccX,
404 egN_eq3freq_onccX, egN_eq1bw, egN_eq2bw, egN_eq3bw, egN_eq1bw_onccX,
405 egN_eq2bw_onccX, egN_eq3bw_onccX, lfoN_eq1gain, lfoN_eq2gain,
406 lfoN_eq3gain, lfoN_eq1gain_onccX, lfoN_eq2gain_onccX, lfoN_eq3gain_onccX,
407 lfoN_eq1gain_smoothccX, lfoN_eq2gain_smoothccX, lfoN_eq3gain_smoothccX,
408 lfoN_eq1gain_stepccX, lfoN_eq2gain_stepccX, lfoN_eq3gain_stepccX,
409 lfoN_eq1freq, lfoN_eq2freq, lfoN_eq3freq, lfoN_eq1freq_onccX,
410 lfoN_eq2freq_onccX, lfoN_eq3freq_onccX, lfoN_eq1freq_smoothccX,
411 lfoN_eq2freq_smoothccX, lfoN_eq3freq_smoothccX, lfoN_eq1freq_stepccX,
412 lfoN_eq2freq_stepccX, lfoN_eq3freq_stepccX, lfoN_eq1bw, lfoN_eq2bw,
413 lfoN_eq3bw, lfoN_eq1bw_onccX, lfoN_eq2bw_onccX, lfoN_eq3bw_onccX,
414 lfoN_eq1bw_smoothccX, lfoN_eq2bw_smoothccX, lfoN_eq3bw_smoothccX,
415 lfoN_eq1bw_stepccX, lfoN_eq2bw_stepccX, lfoN_eq3bw_stepccX
416 - implemented opcodes eq1_vel2freq, eq2_vel2freq,
417 eq3_vel2freq, eq1_vel2gain, eq2_vel2gain, eq3_vel2gain
418 - sfz parser: allow double spaces in sample filenames
419 - sfz parser: allow absolute paths for sample filenames
420 - use linear decay and release for filter and pitch EG
421 - bugfix: only the first amp_veltrack definition in a file was
422 used
423 - bugfix: looping was disabled if loop_start was set to 0
424 - allow regions with end=-1 to turn off other regions using the
425 group and off_by opcodes (#168)
426 - made end=0 play the whole sample
427 - fixed support for lochan and hichan opcodes (#155)
428 - fixed crash when using lochan/hichan opcodes (#187)
429 - sfz parser: allow -200 to 200 for pan_oncc opcode (#182)
430 - added FLAC support (#191)
431 - sfz parser bugfix: lines starting with whitespace were ignored
432 - added amplitude opcode
433 - added support for "#include" instruction
434 (modified patch which was originally posted by Sergey on LS mailing list)
435 - bugfix: generation of velocity curves etc should not be done
436 after each "#include", only after the main file is parsed
437 - bugfix: line numbers in error messages were wrong after "#include"
438 - added support for float and 32 bit sample files
439
440 * SoundFont format engine:
441 - Initial implementation (not usable yet)
442 - RT-safeness: avoid malloc in audio thread
443 - fixed a bug that could cause voice stealing to fail
444 - fine-tuned amplitude EG (by switching from gig to sfz EG)
445 - initial implementation of Vibrato LFO and Modulation LFO
446 - initial implementation of cutoff filter
447 - use linear decay and release for filter and pitch EG
448
449 * Host plugins (VST, AU, LV2, DSSI):
450 - AU bugfix: failed to destroy its audio/MIDI devices
451 - Listen to all interfaces on Mac OS X (INADDR_ANY)
452 - VST bugfix: If the host called resume() before and after
453 changing sample rate or block size, the number of channels was
454 incorrectly set to two. This caused silence in Cubase 5.
455 - save engine type (gig, sfz or sf2) in plugin state
456 - VST: when opening Fantasia, look for both 32 and 64 bit Java on
457 64 bit Windows
458 - AU: changed number of output channels from one stereo to 16
459 stereo
460 - VST: made it possible to build the VST plugin for Mac
461 - AU: link AU plugin dynamically if --disable-shared isn't
462 specified
463 - LV2 "state" extension support (patch by David Robillard)
464 - VST bugfix: instrument loading hang and crashed the host when
465 the plugin was loaded a second time (#174)
466 - plugin bugfix: instrument loading hang when the plugin was
467 loaded a second time (this time it's for Linux and Mac, previous
468 similar fix was for Windows)
469 - thread safety fixes for the instrument loading thread
470 - LV2: use the new lv2 package if present
471 - VST: try to open Fantasia automatically on Linux and Mac too (on
472 Linux, the Fantasia jar should be placed in <prefix>/share/java)
473 - VST: fixed crashes on Linux Ardour and EnergyXT
474 - DSSI bugfix: it wasn't possible to change engine type. The MIDI
475 port and audio channel routing for DSSI plugins are now visible.
476 - LV2: use urid and atom extensions instead of deprecated uri-map
477 and event
478 - LV2: lv2 package 1.0 is now required to build the LV2 plugin
479 - LV2: changed number of output channels to 16 stereo, just like
480 the VST and AU plugins
481 - LV2: fixed save/restore of SFZ state (patch by David Robillard)
482 - LV2: made LV2 plugin buildable on Windows and Mac
483 - VST: implemented retrieval and switching of programs using the
484 sampler's internal MIDI instrument mapping system
485
486 * Instrument editor interface:
487 - Changed instrument editor plugin interface, providing additional
488 informations like the EngineChannel for which the instrument editor was
489 spawned for. This allows the instrument editors to interact more actively
490 with the sampler.
491
492 * MIDI driver:
493 - ALSA MIDI driver supports now "NAME" device parameter, for overriding
494 the ALSA sequencer client name
495 - removed limit of maximum amount of MIDI ports per MIDI device, since
496 there is no reason for this limit
497 - MME: fixed memory handling bug found with cppcheck
498 - MME: removed compiler warning
499 - CoreMIDI: implemented driver specific port parameter "CORE_MIDI_BINDINGS",
500 which allows to retrieve the list of CoreMIDI clients / ports and to connect
501 to them a la JACK, via the usual sampler APIs
502 - CoreMIDI: added driver specific port parameter "AUTO_BIND", if enabled
503 the driver will automatically connect to other CoreMIDI clients' ports
504 (e.g. external MIDI devices being attached to the Mac)
505 - added support for MIDI note on velocity filter
506 - CoreMIDI: fixed memory deallocation error
507 - Fixed variable underflow in VirtualMidiDevice, which caused graphical
508 virtual keyboards in frontends / instrument editors being stuck.
509 - Bugfix in VirtualMidiDevice: process note on with velocity 0 as note off.
510 - Implemented missing handling of MIDI "running status".
511 - CoreMIDI fix: a MIDIPacket can contain more than one event per packet.
512 - MME bugfix: driver wasn't closed properly
513 - CoreMIDI: automatically connect to all input sources by default (driver
514 parameter "AUTO_BIND").
515 - CoreMIDI: fixed auto bind feature to CoreMIDI ports that go online
516 - CoreMIDI: fixed minor memory leak
517
518 * audio driver:
519 - ASIO driver fixes for newer gcc versions (fix from PortAudio)
520 - JACK audio: react on sample rate changes.
521 - JACK audio: react on buffer size changes.
522 - JACK audio: jack_port_get_buffer() was cached and called outside
523 RT context.
524 - ASIO driver: removed compiler warnings
525 - CoreAudio: fixed minor error handling bug
526 - ASIO driver: be more verbose when no ASIO card could be found (fixes #203)
527 - JACK audio: return the JACK server's current sample rate as default value
528 for audio device parameter "SAMPLERATE" (fixes #166).
529
530 * LSCP server:
531 - added support for sending MIDI CC messages via LSCP command
532 "SEND CHANNEL MIDI_DATA CC <sampler-chan> <ctrl> <val>"
533 - added LSCP command "GET AVAILABLE_EFFECTS"
534 - added LSCP command "LIST AVAILABLE_EFFECTS"
535 - added LSCP command "GET EFFECT INFO <effect-index>"
536 - added LSCP command "CREATE EFFECT_INSTANCE <effect-index>"
537 - added LSCP command
538 "CREATE EFFECT_INSTANCE <effect-system> <module> <effect-name>"
539 - added LSCP command "DESTROY EFFECT_INSTANCE <effect-instance>"
540 - added LSCP command "GET EFFECT_INSTANCES"
541 - added LSCP command "LIST EFFECT_INSTANCES"
542 - added LSCP command "GET EFFECT_INSTANCE INFO <effect-instance>"
543 - added LSCP command
544 "GET EFFECT_INSTANCE_INPUT_CONTROL INFO <effect-instance> <input-control>"
545 - added LSCP command "SET EFFECT_INSTANCE_INPUT_CONTROL VALUE
546 <effect-instance> <input-control> <value>"
547 - added LSCP command "GET SEND_EFFECT_CHAINS <audio-device>"
548 - added LSCP command "LIST SEND_EFFECT_CHAINS <audio-device>"
549 - added LSCP command "ADD SEND_EFFECT_CHAIN <audio-device>"
550 - added LSCP command
551 "REMOVE SEND_EFFECT_CHAIN <audio-device> <effect-chain>"
552 - added LSCP command
553 "GET SEND_EFFECT_CHAIN INFO <audio-device> <effect-chain>"
554 - added LSCP command "APPEND SEND_EFFECT_CHAIN EFFECT <audio-device>
555 <effect-chain> <effect-instance>"
556 - added LSCP command "INSERT SEND_EFFECT_CHAIN EFFECT <audio-device>
557 <effect-chain> <effect-chain-pos> <effect-instance>"
558 - added LSCP command "REMOVE SEND_EFFECT_CHAIN EFFECT <audio-device>
559 <effect-chain> <chain-pos>"
560 - added LSCP command "SET FX_SEND EFFECT <sampler_channel>
561 <fx_send_id> <effect_chain> <chain_pos>"
562 - added LSCP command "REMOVE FX_SEND EFFECT <sampler_channel> <fx_send_id>"
563 - added LSCP commands "SUBSCRIBE EFFECT_INSTANCE_COUNT",
564 "SUBSCRIBE EFFECT_INSTANCE_INFO", "SUBSCRIBE SEND_EFFECT_CHAIN_COUNT",
565 "SUBSCRIBE SEND_EFFECT_CHAIN_INFO"
566 - provide comprehensive error messages on LSCP syntax errors
567 (suggesting expected next non-terminal symbols)
568 - Fixed client connection not being closed after network errors.
569
570 * LSCP shell:
571 - Added support for auto correction of obvious and trivial syntax mistakes.
572 - Added support for auto completion by tab key.
573 - Show currently available auto completion while typing.
574 - Added support for browsing command history with up / down keys.
575 - Show all possible next symbols immediately right to the current command
576 line while typing (no double tab required for this feature, as it would
577 be the case in other shells).
578 - Added support for moving cursor left/right with arrow keys.
579 - Added support for built-in LSCP reference documentation, which will
580 automatically show the relevant LSCP reference section on screen as soon
581 as one specific LSCP command was detected while typing on the command
582 line.
583
584 * Real-time instrument scripts:
585 - Implemented built-in script array variable %CC.
586 - Implemented built-in script int variable $CC_NUM.
587 - Implemented built-in script int variable $EVENT_NOTE.
588 - Implemented built-in script int variable $EVENT_VELOCITY.
589 - Implemented built-in script constant variable $VCC_MONO_AT.
590 - Implemented built-in script constant variable $VCC_PITCH_BEND.
591 - Implemented execution of script event handler "init".
592 - Implemented execution of script event handler "controller".
593 - Implemented execution of script event handler "note".
594 - Implemented execution of script event handler "release".
595 - Implemented built-in script function "play_note()" (only two of the
596 max. four function arguments are currently implemented yet though).
597 - Implemented built-in script int variable $EVENT_ID.
598 - Implemented built-in script function "ignore_event()"
599 - Implemented built-in script function "ignore_controller()" (may have one
600 or no argument).
601 - Implemented built-in script function "set_controller()".
602 - Added extended script VM for the Gigasampler/GigaStudio format sampler
603 engine, which extends the general instrument script VM with Giga format
604 specific variables and functions.
605 - Giga format scripts: added built-in script int constant variables
606 $GIG_DIM_CHANNEL, $GIG_DIM_LAYER, $GIG_DIM_VELOCITY, $GIG_DIM_AFTERTOUCH,
607 $GIG_DIM_RELEASE, $GIG_DIM_KEYBOARD, $GIG_DIM_ROUNDROBIN, $GIG_DIM_RANDOM,
608 $GIG_DIM_SMARTMIDI, $GIG_DIM_ROUNDROBINKEY, $GIG_DIM_MODWHEEL,
609 $GIG_DIM_BREATH, $GIG_DIM_FOOT, $GIG_DIM_PORTAMENTOTIME, $GIG_DIM_EFFECT1,
610 $GIG_DIM_EFFECT2, $GIG_DIM_GENPURPOSE1, $GIG_DIM_GENPURPOSE2,
611 $GIG_DIM_GENPURPOSE3, $GIG_DIM_GENPURPOSE4, $GIG_DIM_SUSTAIN,
612 $GIG_DIM_PORTAMENTO, $GIG_DIM_SOSTENUTO, $GIG_DIM_SOFT,
613 $GIG_DIM_GENPURPOSE5, $GIG_DIM_GENPURPOSE6, $GIG_DIM_GENPURPOSE7,
614 $GIG_DIM_GENPURPOSE8, $GIG_DIM_EFFECT1DEPTH, $GIG_DIM_EFFECT2DEPTH,
615 $GIG_DIM_EFFECT3DEPTH, $GIG_DIM_EFFECT4DEPTH, $GIG_DIM_EFFECT5DEPTH.
616 - Giga format scripts: Implemented built-in script function
617 "gig_set_dim_zone(event_id, dimension, zone)".
618 - Implemented built-in script int array variable $KEY_DOWN.
619 - Implemented built-in script function "abs()".
620 - Implemented built-in script function "random()".
621 - Implemented built-in script function "num_elements()".
622 - Implemented built-in script function "note_off()".
623 - Implemented built-in script function "set_event_mark()".
624 - Implemented built-in script function "delete_event_mark()".
625 - Implemented built-in script function "by_marks()".
626 - Added built-in script int const variables $MARK_1 to $MARK_28.
627 - Built-in script functions "ignore_event()", "note_off()" and
628 "gig_set_dim_zone()" now also accept an array of event IDs as argument
629 (i.e. return value of new script function "by_marks()").
630 - Pass/preserve polyphonic variable data from respective "note" event
631 handler to "release" event handler.
632 - Fixed crash when using built-in script function "by_marks()".
633
634 * Bug fixes:
635 - Fixed crash which may occur when MIDI key + transpose is out of range
636 - minor valgrind fixes
637 - fixed crash which occurred when changing an already deployed sampler
638 channel to a different engine type
639 - fixed crash when deleting a sampler channel or changing engine
640 type while an instrument load was in progress
641 - bugfix: playing a note while changing the instrument could cause
642 a crash, or give "Handing back unknown region" error messages
643 - bugfix: calling SET VOICES (which Fantasia does on start and
644 refresh) could cause instruments to be unloaded
645 - Mac OS X: fixed name collision of enum in EffectControl and
646 wrong name of destructor in AudioOutputDeviceCoreAudio.cpp
647 - Mac OS X: fixed hanging threads
648 - Windows: disabled the previous thread fix on non-Mac systems, as
649 it caused hanging threads on Windows
650 - Fixed possible crashes due to corrupted MIDI/audio device list
651 after MIDI/audio device creation failure
652 - When creating MIDI instrument map entries with "PERSISTENT" type, the
653 instruments were uselessly precached with zero samples, however it still
654 took the full preloading time and on 1st program change the respective
655 instrument was completely reloaded again.
656 - fixed handling of rapid bank select and program change messages
657 sent to the same sampler channel (patch from the Open Octave
658 project, slightly adjusted)
659 - fixed crash when trying to create an effect instance with controls
660 which min and/or max values depend on the sample rate
661 - fixed bug #162
662 - bugfix: LADSPA_PATH was not evaluated correctly when containing
663 multiple paths (#165)
664 - thread safety fixes for the instrument loading thread
665 - bugfix: instrument loading crashed for sfz and sf2 in Ardour
666 (#176)
667 - more thread safety fixes for the instrument loading thread
668 - sfz/sf2 engine: fixed crash when using small audio fragment size
669 - Mac OS X: fixed crash when unloading plugin on 10.7 and later
670 - Mac OS X: fixed process hang when unloading 32-bit plugin (bug
671 introduced in previous fix)
672 - fixed crash when a channel received a program change while
673 playing a note in a key group
674 - fixed erroneous error message piping in VoiceBase.h
675 (labelled "Disk stream not available in time")
676 - Update effects on sample rate & period size changes (to avoid
677 crashes and noise on such transitions).
678 - ignore missing LADSPA paths without ignoring valid LADSPA paths
679 (fixes #208)
680 - Fixed bug in LSCP grammar definition which caused a statement like
681 "GET SERVER INFOasdf\n" to be accepted as valid statement (was so far
682 practically irrelevant, however it caused problems with the new LSCP
683 shell's auto completion feature).
684 - Fixed MIDI program change messages being ignored if quickly executed
685 after each other (fixes #231).
686
687 Version 1.0.0 (31 July 2009)
688
689 * packaging changes:
690 - autoconf bugfix: the PKG_CONFIG variable wasn't initialized properly,
691 causing e.g. the libgig test to fail when
692 "./configure --disable-jack-driver" was used
693 (patch by Alexis Ballier)
694 - fixed compilation with gcc 4.3
695 - fixes for building on OS X (thanks to Ebrahim Mayat for testing)
696 - fixed configure so it detects x86_64 (#107)
697 - fixes for building with newer MinGW versions
698 - fix for building with bison 2.4 (#111)
699 - fixed building with libgig installed in a non-standard directory
700 - minor fix in configure for mmsystem.h detection on MinGW
701 - Windows: look for editor plugins and Fantasia using base
702 directory of liblinuxsampler dll (look in the same directory and one
703 directory above)
704 - configure script fix: removed unconditional use of SSE
705 - fixed building with sqlite installed in a non-standard directory
706 - when cross-compiling, don't try to create instruments.db
707 - fix for new mingw-w64 version, which has usleep
708
709 * general changes:
710 - bugfix: on some POSIX systems instrument editor plugins refused to
711 load as we used a non-portable Linux specific struct field
712 (fixes bug #70, patch by Ronald Baljeu)
713 - fixed endless loop which occured when loading an instrument editor
714 plugin DLL on Windows systems failed
715 - fixed memory leaks that occurred when liblinuxsampler was unloaded
716 - optimized the SynchronizedConfig class so it doesn't wait
717 unnecessarily long after an update
718 - added support for notifying instrument editors on note-on / note-off
719 events (e.g. to highlight the pressed keys on the virtual keyboard
720 of gigedit)
721 - added support for triggering notes by instrument editors (see above)
722 - be verbose on DLL load errors (on Linux)
723 - fixed di-harmonic triangle LFO implementation (this LFO implementation
724 is only used by very few systems, most pick the int math
725 implementation, which is usually faster)
726 - fixes for audio drivers with varying buffer sizes
727 - experimental support for running LinuxSampler as a DSSI, LV2 and
728 VST plugin
729 - notification events for stream/voice count statistics are now sent
730 only when there are actual changes
731 - added memory ordering constraints to improve stability on
732 multi-core and multi-cpu systems
733 - maximum voices and disk streams can now be altered at runtime
734 - fixed CPU feature detection on x86_64 (maybe fixes #108)
735 - automatic stacktrace mechanism is now turned off by default and can
736 be switched on by command line option "--stacktrace" (the automatic
737 stacktrace mechanism seems to be broken on most systems at the moment)
738 - C++ API method InstrumentManager::LaunchInstrumentEditor() now returns
739 a pointer to the launched InstrumentEditor object
740 - added optional 3rd party user data parameter for following
741 liblinuxsampler C++ API methods: InstrumentEditor::Main(),
742 InstrumentEditor::Launch(),
743 InstrumentManager::LaunchInstrumentEditor()
744 - theoretical fix: made SynchronizedConfig follow C++0x memory
745 model more strictly
746 - fixes for using large audio device buffers
747 - Windows: add the installation directory to the DLL search path
748 when loading an editor plugin (solves problems with VST and
749 gigedit on systems with other GTK versions installed)
750 - updated linuxsampler man page
751
752 * audio driver:
753 - removed the nonsense audio channel constraint (which was hard coded to
754 max. 100 audio channels) for most audio drivers
755 - JACK audio driver did not offer a device parameter "SAMPLERATE" as
756 opposed to the LSCP specs
757 - bugfix: the SAMPLERATE parameter of some drivers (e.g. JACK)
758 reflected the wrong value
759 - fixed a memory management bug in ASIO driver
760 - Makefile fix: JACK_CFLAGS wasn't used
761 - JACK: use jack_client_open instead of the deprecated
762 jack_client_new
763 - added (experimental) CoreAudio driver
764 - applied old fixes to the ASIO driver that were included in the
765 previous binary release but accidentally never committed to CVS
766 (fixes #117)
767 - fixes for ASIO on mingw-w64 (iasio wrapper is not needed on
768 win64)
769 - VST: added support for sample rate and buffer size changes
770 - VST: close editor (Fantasia) when the VST is removed
771 - VST: avoid opening Fantasia more than once for each VST instance
772 - VST: export main function as "main" on Linux too (fix for
773 energyXT)
774 - VST: changed number of output channels from one stereo to 16
775 stereo
776 - added channel routing, fxsends and midi maps to the settings
777 stored in the plugin state
778 - performance optimization of AudioChannel::MixTo() and
779 AudioChannel::CopyTo() methods using GCC vector exensions
780 (if available)
781 - ASIO fixes: avoid initializing the device twice, avoid throwing
782 exception when getting parameters from a disconnected device
783
784 * MIDI driver:
785 - added JACK MIDI driver
786 - dispatch bank select as ordinary CC as well, the user might seriously
787 want to (mis)use it for some purpose ("fixed" in all current MIDI
788 input drivers: ALSA, CoreMIDI, JACK, MidiShare, MME)
789 - bugfix: pitch bend wasn't working with JackMidi, VST, LV2, MME,
790 CoreMidi or AU
791 - fixed mingw-w64 compilation error in MME driver
792 - made program change handling in MIDI thread real-time safe by
793 moving the logic to a non-RT thread
794 - fixed minor memory leak in ALSA driver
795
796 * instruments database:
797 - avoid time consuming samples scanning when adding instruments
798 to the instruments database
799 - added support for handling lost files in the instruments database
800 - Implemented option for adding instruments in separate directories
801 in the instruments database
802 (patch by Chris Cherrett & Andrew Williams, a bit adjusted)
803 - work-around for missing fnmatch function on Windows to make
804 instrument database compilable
805 - added instrument database support on Windows
806 works with both standalone mode and VST plugin.
807 the instruments DB file is located in
808 %USERPROFILE%\.linuxsampler\instruments.db which allows different
809 databases for each windows user
810 if no DB is present it automatically creates the .linuxsampler subdir
811 and then creates an empty DB
812 - fixed recursive import, which was broken on Windows
813
814 * Gigasampler format engine:
815 - fixed a memory leak that could happen when a channel was deleted
816 while notes were playing
817 - made it possible to load an instrument even if the audio thread
818 isn't running
819 - added partial support for the "Controller Triggered" MIDI rule,
820 enough for piano gigs with pedal noise samples
821 - minor fix: only mark FX sends as being modified if really the
822 respective FX send MIDI controller was used
823 - added support for GM global device volume SysEx message
824 (can be selected at compile time to either apply globally to the
825 whole sampler [default] or only to the sampler channels that are
826 connected to the respective MIDI input port the SysEx message arrived
827 on)
828 - bugfix: notes triggered at position 0 in the audio buffer were
829 sometimes wrongly killed in the same buffer, causing no sound to
830 be played
831 - added support for chromatic / drumkit mode Roland GS Sysex message
832 which is usally used to switch a part between chromatic sounds and
833 drumkit sounds (as opposed to their common default setting of MIDI
834 part 10 being a drumkit part and all other ones chromatic parts), in
835 LS however one can switch between the first 16 MIDI instrument maps
836 defined for the sampler
837 - made it possible to create multiple sample channels even if the
838 audio thread isn't running
839 - fixed crash when removing channel with active voices (#116)
840 - bugfix: on sample reference changes (instrument editor), only
841 un-cache the respective sample if it's really not used by any
842 sampler engine anymore
843 - re-cache samples in case they were changed by an instrument editor,
844 e.g. when a sample was added while playing (#82)
845 - fixed hanging notes which occured when note-off event had the exact
846 same time stamp as the voice's note-on event and occured both in the
847 same audio fragment cycle (fixes bug #112)
848 - added support for the "fine tune" and "pitch bend range"
849 instrument-level gig parameters
850 - fixed minor artifacts in pitch bend handling
851 - added support for GS Reset SysEx message
852 - allow gig files to use unlimited downward pitch shifting
853 - added a limit check for upward pitch shifting
854 - bugfix: sometimes, when playing a note twice fast, the second
855 note was silent
856 - fixed crash happening when a pitch bend event arrived at the
857 same time a new instrument was loading
858
859 * LSCP server:
860 - added new LSCP event "CHANNEL_MIDI" which can be used by frontends to
861 react on MIDI data arriving on certain sampler channels (so far only
862 Note-On and Note-Off events are sent via this LSCP event)
863 - added new LSCP event "DEVICE_MIDI" which can be used by frontends to
864 react on MIDI data arriving on certain MIDI input devices (so far only
865 Note-On and Note-Off events are sent via this LSCP event)
866 - added new LSCP commands: FIND LOST DB_INSTRUMENT_FILES and
867 SET DB_INSTRUMENT FILE_PATH
868 - added new LSCP command: SEND CHANNEL MIDI_DATA which can be used by
869 frontends to send MIDI messages to specific sampler channel
870 - added two additional fields to GET FILE INSTRUMENT INFO command -
871 KEY_BINDINGS and KEYSWITCH_BINDINGS
872 - bugfix: the bank number provided by MIDI_INSTRUMENT_INFO notifications
873 was incorrect
874 - Optimized the retrieval of the MIDI instrument mappings
875 - added new LSCP command "SET VOICES" to globally alter the maximum
876 amount of voices, added new LSCP event "GLOBAL_INFO:VOICES" which will
877 be triggered respectively
878 - added new LSCP command "SET STREAMS" to globally alter the maximum
879 amount of disk streams, added new LSCP event "GLOBAL_INFO:STREAMS"
880 which will be triggered respectively
881 - bugfix: retry if "select" returns EINTR (this fixes a crash when
882 a gigedit file dialog is opened)
883 - close all connections when LSCPServer is deleted
884 - hard close of all sockets on exit
885 - bugfix: SET CHANNEL MIDI_INPUT_TYPE didn't work with the MME
886 driver
887
888 * Bug fixes:
889 - fixed a crash which occurs when removing a sampler channel waiting
890 to start instrument loading after another channel
891 - fixed a crash which occurs when removing a sampler channel with
892 instrument loading in progress (bug #113)
893 - fixed termination caused by uncaught exception when adding MIDI
894 instrument with PERSISTENT load mode
895 - fixed possible iterator invalidations when resetting the sampler
896 - fixed memory leaks when issuing the following LSCP commands:
897 GET AUDIO_OUTPUT_DRIVER INFO
898 GET AUDIO_OUTPUT_DRIVER_PARAMETER INFO
899 GET MIDI_INPUT_DRIVER INFO
900 GET MIDI_INPUT_DRIVER_PARAMETER INFO
901 - fixed possible compilation error when sqlite is not present
902 - fixed orphaned pointers when setting maximum voices limit (bug #118)
903 - fixed crash when changing the audio output device of a sampler
904 channel with loaded instrument and start playing notes
905 - fixed endless loop in Engine::SuspendAll() (bug #120)
906 - fixed a low-level atomic load function that was broken on 64-bit
907 PowerPC, which probably could cause crashes on that platform
908 - fixed a memory management error which could cause a crash when a
909 plugin was unloaded
910 - bugfix: two private structs had the same name, which could cause
911 problems if the linker chose the wrong constructor
912 - fixed low-level ConditionServer usage bug that caused lockups on
913 Windows
914
915
916 Version 0.5.1 (6 December 2007)
917
918 * packaging changes:
919 - added autoconf checks for pthread library
920 - added autoconf check for pthread bug found on certain NPTL-enabled
921 glibc versions (see Gentoo bug report #194076)
922 - added autoconf checks for MS Windows
923
924 * general changes:
925 - bugfix: the thread used by an editor plugin didn't die when the
926 editor closed
927 - bugfix: Ringbuffer.h: fill_write_space_with_null() did not zero
928 out all the space. operator--() did not apply size_mask after
929 decrementing the read_ptr. DEFAULT_WRAP_ELEMENTS set to 0 in
930 order to avoid problems with the _NonVolatileReader functions.
931 - bugfix: Stream.h: added a missing divide by BytesPerSample in
932 GetWriteSpace(). Since this function is currently only used in
933 the stream's qsort() compare function, it didn't trigger any
934 bugs.
935 - Resampler.h, Synthesizer.h: cubic interpolation now works in
936 24bit mode too. Faster method to read 24bit words on little
937 endian machines (x86): replaced 3 byte reads + shifts with a 1
938 unaligned 32bit read and shift
939 - experimental support for MS Windows (MIDI input via MME, AUDIO
940 output via ASIO)
941 - made handling of SIGINT signal (Ctrl-C) a bit more robust
942 - support for monitoring the total number of active disk streams
943 (new LSCP commands: GET TOTAL_STREAM_COUNT,
944 SUBSCRIBE TOTAL_STREAM_COUNT, UNSUBSCRIBE TOTAL_STREAM_COUNT)
945
946 * AUDIO driver:
947 - added Windows ASIO low latency audio driver
948
949 * MIDI driver:
950 - added MME Windows MIDI driver
951
952 * LSCP server:
953 - added support for Windows style path / filenames, however with
954 forward slash path separators instead of backslash
955 (i.e. "C:/foo/bar.gig")
956 - allow naughty liblscp to send non-string device parameters within
957 apostrophes as well
958 - added new LSCP commands: "GET FILE INSTRUMENTS <file>",
959 "LIST FILE INSTRUMENTS <file>" and
960 "GET FILE INSTRUMENT INFO <file> <index>" for retrieving informations
961 about an arbitrary instrument file on the system where the sampler is
962 running on
963
964 Version 0.5.0 (15 October 2007)
965
966 * packaging changes:
967 - config.h is not going to be installed along with liblinuxsampler's
968 API header files anymore
969 - only the API relevant header (and source) files will be exported to
970 the Doxygen API documentation (explicitly listed in Doxyfile.in)
971 - added completely new XCode project files for Mac OSX which is now
972 capable to execute our autoconf environment, thus no need anymore
973 to maintain the compile time configuration file (osx/version.h) for
974 OSX manually (patch by Toshi Nagata)
975 - fixed buggy boolean --enable-foo / --disable-foo configure script
976 parameters
977 - global.h now only covers global definitions that are needed for the
978 C++ API header files, all implementation internal global definitions
979 are now in global_private.h
980 - atomic.h is not exposed to the C++ API anymore
981 - no need to include config.h anymore for using LS's API header files
982 - fixed warnings in API doc generation
983
984 * general changes:
985 - replaced the old, confusing MIDI program change mechanism by a
986 flexible MIDI instrument mapper which allows to map arbitrary
987 (MIDI bank, MIDI program) pairs with arbitrary
988 (engine type, instrument file, file index) triplets which will be
989 loaded on the respective channel when such MIDI program change
990 messages arrive, beside that, each entry allows to define a life-time
991 strategy for the instrument, so the user can define whether the
992 instrument should i.e. be permanently loaded (to be able to switch
993 quickly among sounds) or i.e. loaded just on demand when the
994 respective program change arrives (to safe RAM space), as well as a
995 global volume factor for each entry, so the user can adjust the volume
996 dynamic of his mapped instrument collection without having to modify
997 the instrument files, also one can manage arbitrary amount of such
998 MIDI instrument maps and assign each sampler channel individually
999 a certain map, so that i.e. separation of normal instruments and
1000 drumkits is possible
1001 - new notification events for tracking changes to audio/MIDI devices,
1002 MIDI instrument maps, MIDI instruments, FX sends, global volume.
1003 - sampler was limited to load max. 200 instruments in the background
1004 due to a constant size RingBuffer FIFO which is now replaced by a
1005 dynamic (unlimited) size std::list FIFO
1006 - added FX sends, these allow to route audio signals to arbitrary audio
1007 output channels for being processed by external effect processors
1008 (i.e. jack-rack), the send levels are controllable via arbitrary MIDI
1009 controllers
1010 - global (sampler wide) volume can now be controlled at runtime
1011 - Implemented new, improved notification system
1012 - fixed compilation errors regarding OSX
1013 (patch by Toshi Nagata)
1014 - implemented instruments database
1015 - added support for escape sequences to the instruments database
1016 - added highly experimental support for on-the-fly instrument editing
1017 within the sampler's process (by using instrument editor plugins),
1018 you'll notice the new "Registered instrument editors:" message on
1019 startup, added a new LSCP command:
1020 "EDIT CHANNEL INSTRUMENT <sampler-channel>"
1021 to spawn a matching instrument editor for the instrument on the
1022 given sampler channel, the plugin path can be overridden at compile
1023 time with ./configure --enable-plugin-dir=/some/dir
1024 - added experimental code for synchronizing instrument editors hosted
1025 in the sampler's process to safely edit instruments while playing
1026 without a crash (hopefully) by either suspending single regions
1027 wherever possible or if unavoidable whole engine(s)
1028 - fixed several issues in fundamental "Thread" class: set scheduling
1029 policy and priority on thread level, set a minimum stack size for
1030 thread (TODO: a reasonable value yet to be tested), bugfix: non-RT
1031 threads simply inherited properties of starting thread instead of
1032 setting their own policy and priority
1033 - minor fix in our automatic stack trace mechanism on crashes, the
1034 main process did not wait for the stack trace process to finish
1035 its output
1036 - fixed some minor memory leaks
1037 - reenabled assembly features support, at the moment only for
1038 enabling a fast denormal FPU mode (x86 platforms supporting SSE2)
1039 - minor assembly fix in x86 features detection (don't use the PIC
1040 register, to avoid relocations in the text segment at runtime)
1041 - POSIX callback functions of Thread.h are hidden
1042
1043 * MIDI driver:
1044 - dispatch bank select (MSB and LSB) messages
1045
1046 * audio driver:
1047 - the ALSA audio output driver parameters now reflect the correct
1048 parameter value ranges for the respective selected sound card
1049 (patch by Till Wimmer, a bit fixed and extended)
1050
1051 * Gigasampler format engine:
1052 - if a filter is used and EG2 finishes before EG1, let the voice
1053 die when EG2 has finished (this fixes a problem with clicks and
1054 voice starvation for some gigs)
1055 - playback is no longer disabled during instrument loading
1056 - all notes playing on a channel that changes its instrument keep
1057 playing with the old instrument until they get a note off
1058 command
1059 - EG fix: a release value of zero could cause noises or crash
1060 - handle MIDI coarse tuning messages (MIDI RPN #0 MSB #2 LSB)
1061 - EG fine tuning: when attack is zero the EG starts at a level
1062 above max sustain level, which means that there is a decay phase
1063 even if sustain is 100%
1064 - more EG fixes: the level could sometimes go below zero and cause
1065 noises or crashes
1066 - minor fix of EGDecay (patch by Toshi Nagata)
1067 - fixed compiler error when --enable-override-filter-type was
1068 supplied to the configure script (fixes #46)
1069 - disk thread: queue sizes are now proportional to CONFIG_MAX_STREAMS
1070 instead of a fix value
1071 - behavior fix: on MIDI CC# 65 (portamento on / off), 126 (mono mode),
1072 127 (solo mode) only kill voices if the respective mode really
1073 changed
1074
1075 * LSCP server:
1076 - fixed compile time error for old Bison versions
1077 (i.e. v1.28 found on MacOS 10.4, patch by Toshi Nagata)
1078 - parser now supports extended ASCII character set
1079 (up to ASCII code 255, i.e. includes now umlauts and accents)
1080 - filename arguments in LSCP commands now allow to use escape
1081 sequences, that is directly literal as one of: \', \", \\, \n, \r,
1082 \f, \t, \v, or as octal ASCII code value like \132, or as hex ASCII
1083 code value like \xf2) (fixes bug #24)
1084 - the following LSCP commands now also support escape sequences for at
1085 least one of their text-based parameters (i.e. name, description):
1086 "ADD MIDI_INSTRUMENT_MAP", "MAP MIDI_INSTRUMENT",
1087 "SET MIDI_INSTRUMENT_MAP NAME", "SET FX_SEND NAME", "CREATE FX_SEND",
1088 "SET DB_INSTRUMENT_DIRECTORY NAME",
1089 "SET DB_INSTRUMENT_DIRECTORY DESCRIPTION", "SET DB_INSTRUMENT NAME",
1090 "SET DB_INSTRUMENT DESCRIPTION", "FIND DB_INSTRUMENTS",
1091 "FIND DB_INSTRUMENT_DIRECTORIES"
1092 - returns verbose syntax errors (line and column where syntax error
1093 occured, the unexpected character and the actually expected, possible
1094 character(s), the latter only if less than 5 possibilities)
1095 - made sure that LSCP syntax is not affected by gigedit locale
1096 settings
1097 - bugfix regarding strings parameter lists: all comma separated lists
1098 of strings were treated as being one string containing commas
1099 (fixes #57)
1100
1101 * Bug fixes:
1102 - fixed crash occurring on certain LSCP scripts (Bug 39)
1103 - another thread safety fix for lscp "load engine" and "set
1104 channel audio output device"
1105 - fixed a crash which occurs when reassigning the same engine
1106 on a sampler channel with connected MIDI device
1107 - fixed a crash which occurs when changing the number of ports of a MIDI
1108 device connected to a sampler channel to number less then or equal
1109 to the index of the port to which the sampler channel is connected.
1110 - The previous bindings were not been disconnected when altering
1111 the ALSA_SEQ_BINDINGS parameter. Introduced a NONE keyword for
1112 unsubscribing from all bindings (e.g. ALSA_SEQ_BINDINGS=NONE).
1113 - The active stream/voice count statistic was incorrect.
1114 - notification events were not been sent for some sampler
1115 channel changes
1116 - added default min and max values to restrict the number of allowed
1117 audio output channels and MIDI input ports
1118 - the connection to the PCM interface is now closed when destroying
1119 an audio output device
1120 - files with slash in their path or filename could not be loaded
1121
1122 * test cases:
1123 - updated and fixed (haven't been touched in a while)
1124
1125 Version 0.4.0 (24 November 2006)
1126
1127 * packaging changes:
1128 - changed deprecated copyright attribute to license; added ldconfig
1129 to post-(un)install steps; added devel package for liblinuxsampler;
1130 to linuxsampler.spec (RPM)
1131 - install necessary development header files for allowing 3rd party
1132 applications to link against liblinuxsampler
1133 - liblinuxsampler's API documentation can be generated with 'make docs'
1134 (Doxygen required)
1135 - added benchmark to automatically detect the best triangle LFO
1136 implementation (currently either an integer math solution or a
1137 di-harmonic approximation), automatic detection can be overriden
1138 with --enable-signed-triang-algo=x and --enable-unsigned-triang-algo=x
1139 configure script argument though (mandatory for cross-compilation)
1140 - do not automatically pick optimized gcc flags if the user already
1141 provided some on his own (as CXXFLAGS)
1142 - added compile time option to disable processing of All-Notes-Off MIDI
1143 messages
1144 - added compile time options to allow disabling the various audio and
1145 MIDI drivers
1146 - fixed automatic GCC CPU switch detection on PPC
1147 (patch by Ebrahim Mayat)
1148
1149 * Gigasampler format engine:
1150 - extensive synthesis optimization
1151 (reimplementation of EGs and LFO(s), removed synthesis parameter
1152 prerendering and the synthesis parameter matrix in general, splitting
1153 each audio fragment into subfragments now where each subfragment uses
1154 constant pitch and filter coefficients. The volume coefficient is
1155 linearly interpolated inside a subfragment, unless
1156 --disable-interpolate-volume is set.)
1157 - fine tuning of the EG modulation parameters
1158 - improved filter cutoff calculation by adding support for the
1159 following gig parameters: Cutoff freq (used when no cutoff
1160 controller is defined), Control invert, Minimum cutoff, Velocity
1161 curve and Velocity range. The keyboard tracking now scales
1162 cutoff frequency instead of resonance.
1163 - added support for gig parameter Resonance.
1164 - fixed bug in sysex handling (patch by Juan Linietsky)
1165 - added global volume attenuation of -9 dB (0.35f) to prevent clipping
1166 which can be overridden with --enable-global-attenuation
1167 - EG fixes: made the length of "attack hold" stage more
1168 accurate. Release stage can now start before attack stage
1169 ends. Cancel release didn't work when sustain was zero. Attack
1170 time now has a minimal value to prevent clicks.
1171 - fixed pitch changes larger than one octave
1172 - fixed EG3 (pitch envelope) synthesis which was neutral all the time
1173 - implemented portamento mode and solo mode (a.k.a 'mono mode'):
1174 all modes can be altered via standard GM messages, that is CC5 for
1175 altering portamento time, CC65 for enabling / disabling portamento
1176 mode, CC126 for enabling solo mode and CC127 for disabling solo mode
1177 - fine tuning of the curves for volume (CC7), pan (CC10 and gig
1178 parameter) and crossfade
1179 - added support for the "attenuation controller threshold" gig
1180 parameter
1181 - added smoothing of volume changes caused by control change
1182 messages
1183 - sample loop parameters are now taken from the DimensionRegion
1184 instead of the wave chunk
1185 - fixed keyswitching for v3 gigs with a number of keyswitch splits
1186 not equal to a power of two
1187 - reimplementation of the filter algorithm. The new filter is more
1188 accurate and supports all gig filter types, including bandreject
1189 and lowpass turbo.
1190 - real support for 24 bit samples - samples are not truncated to
1191 16 bits anymore
1192 - support for aftertouch (channel pressure, not polyphonic
1193 aftertouch)
1194 - LFO1 behaviour fixed (dampening from max volume instead of
1195 amplifying from 0)
1196
1197 * LSCP server:
1198 - fixed application exit on broken pipe error (fixes bug #20)
1199 - fixed the notification messages delay due to lack of
1200 network activity (fixes bug #26)
1201 - fixed parser bug which occured on space(s) within device parameters
1202 (e.g. "SET AUDIO_OUTPUT_CHANNEL_PARAMETER 0 0 SOMEPARAM='foo bar'")
1203
1204 * audio driver:
1205 - added aRts audio output driver (by no means RT safe)
1206
1207 * MIDI driver:
1208 - fixed legacy sysex code which caused dispatching of MIDI SysEx
1209 messages several times instead of once
1210 - API extension for MIDI drivers which already supply exact time stamps
1211 for events (i.e. for offline rendering based MIDI drivers)
1212 - added checks for bad MIDI data
1213
1214 * linuxsampler application:
1215 - show available sampler engine types on startup
1216
1217 * general changes:
1218 - support for muting sampler channels and solo mode of the same, two new
1219 LSCP commands ("SET CHANNEL MUTE" and "SET CHANNEL SOLO") and two new
1220 fields ("MUTE" and "SOLO") for command "GET CHANNEL INFO" were
1221 introduced for this, the behavior is the same like on a mixer console
1222 (patch by Grigor Iliev, a bit adjusted). Also added configure option
1223 --enable-process-muted-channels which can be used to enable the
1224 processing of muted channels.
1225 - support for sostenuto pedal
1226 - support for monitoring the total number of active voices
1227 (new LSCP commands: GET TOTAL_VOICE_COUNT, GET TOTAL_VOICE_COUNT_MAX,
1228 SUBSCRIBE TOTAL_VOICE_COUNT, UNSUBSCRIBE TOTAL_VOICE_COUNT)
1229 - fixed some memory management errors
1230 - fixed some concurrency problems that could lead to crashes when
1231 LSCP commands were executed
1232 - fixed crash when instrument loading failed previously on the same
1233 sampler channel (fixes bug #36)
1234
1235 Version 0.3.3 (15 July 2005)
1236
1237 * packaging changes:
1238 - fixed compilation with gcc 4.0
1239
1240 * Gigasampler format engine:
1241 - LFO preprocessing was performed even though the respective LFO was
1242 disabled by the instrument patch (minor efficiency fix)
1243 - if period time of chosen audio device is too small (< MIN_RELEASE_TIME)
1244 for volume ramp downs in the same fragment (needed for current voice
1245 stealing implementation) then simply inform the user with a warning
1246 about possible click sounds and reduce the volume ramp down
1247 appropriately instead of cancelling the audio device connection
1248
1249 Version 0.3.2 (24 June 2005)
1250
1251 * packaging changes:
1252 - updated autotools build files to compile on OS X
1253 (conditional compilation of CoreMIDI and MidiShare drivers)
1254 - hand-crafted assembly optimization code can be disabled with
1255 './configure --disable-asm' (definitely not recommended)
1256 - fixed 'make dist' rule to include all necessary files
1257 - require automake (>= 1.5) for 'make -f Makefile.cvs'
1258 (due to 'dist-bzip2' automake option)
1259
1260 * Gigasampler format engine:
1261 - support for the gig parameters for "release velocity response" curves
1262 - fine tuning of the EGADSR envelope
1263 - volume of release triggered samples now depends on note-on velocity,
1264 note length and gig parameter "release trigger decay" instead of
1265 note-off velocity.
1266 - revised voice stealing
1267 (fixes crash and endless loop caused by voice stealing)
1268 - don't reset scale tuning on instrument or audio output device change
1269 - handle key group conflicts right at the beginning of each fragment
1270 instead of when the respective voice is actually launched
1271 (fixes undefined behavior if stolen voices belonged to a key group -
1272 this case was followed by a "killed voice survived" error message)
1273 - fixed minor issue with null/silence samples
1274 (those stole voices even though they don't need a voice at all which
1275 resulted in "voice stealing didn't work out" messages)
1276 - don't reset volume, pan, pitch and MIDI controller values on
1277 instrument or audio output device change
1278
1279 * LSCP server:
1280 - fixed some crashes (patch by Grigor Iliev, fixes #19)
1281 - fixed LSCP event "CHANNEL_INFO" notification
1282 (e.g. did not notify on volume changes or MIDI program change events)
1283
1284 * linuxsampler application:
1285 - added command line parameters --lscp-addr and --lscp-port to override
1286 default IP address and port of LSCP server
1287
1288 Version 0.3.1 (24 May 2005)
1289
1290 * initial release

  ViewVC Help
Powered by ViewVC