ViewVC logotype

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 2244 - (show annotations) (download)
Thu Aug 18 11:32:33 2011 UTC (9 years, 3 months ago) by iliev
File size: 43264 byte(s)
* sfz engine: use common pool of CC objects to minimize RAM usage

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

  ViewVC Help
Powered by ViewVC