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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 2837 - (show annotations) (download)
Sun Aug 23 06:14:00 2015 UTC (4 years, 3 months ago) by persson
File size: 61065 byte(s)
* fixed printf type errors (mostly in debug messages)


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

  ViewVC Help
Powered by ViewVC