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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 2529 - (show annotations) (download)
Tue Mar 4 20:41:47 2014 UTC (10 years ago) by schoenebeck
File size: 55740 byte(s)
* SFZ format: Added support for "#include" instruction (modified patch
  which was originally posted by Sergey on LS mailing list).
* Bumped version (1.0.0.svn34).

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

  ViewVC Help
Powered by ViewVC