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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 2879 - (show annotations) (download)
Tue Apr 19 14:07:53 2016 UTC (7 years, 11 months ago) by schoenebeck
File size: 63232 byte(s)
* All engines: Active voices are now internally grouped to "Note" objects,
  instead of being directly assigned to a keyboard key. This allows more
  fine graded processing of voices, which is i.e. required for certain
  instrument script features.
* Built-in script function "play_note()": Added support for passing
  special value -1 for "duration-us" argument, which will cause the
  triggered note to be released once the original note was released.
* Bumped version (2.0.0.svn3).

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

  ViewVC Help
Powered by ViewVC