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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 2463 - (show annotations) (download)
Wed Sep 4 20:40:12 2013 UTC (10 years, 6 months ago) by schoenebeck
File size: 53301 byte(s)
* CoreAudio: fixed minor error handling bug.
* Bumped version to 1.0.0.svn22.

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

  ViewVC Help
Powered by ViewVC