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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 2840 - (show annotations) (download)
Sun Aug 30 10:00:44 2015 UTC (8 years, 7 months ago) by persson
File size: 61131 byte(s)
* use unique_ptr instead of auto_ptr when building with C++11

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

  ViewVC Help
Powered by ViewVC