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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 2506 - (show annotations) (download)
Sun Jan 12 11:27:05 2014 UTC (10 years, 2 months ago) by schoenebeck
File size: 54085 byte(s)
* Bugfix: only process the latest MIDI program change event.

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

  ViewVC Help
Powered by ViewVC