ViewVC logotype

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 2183 - (show annotations) (download)
Sat Jun 11 17:53:32 2011 UTC (9 years, 3 months ago) by persson
File size: 39179 byte(s)
* Mac OS X fixes: support the new dir for Core Audio SDK, fixed name
  collision of enum in EffectControl, fixed building outside source
  directory, fixed wrong name of destructor in
* made sure all source files for hostplugins are included when doing
  "make dist"
* removed empty directories left from the cvs to svn migration

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

  ViewVC Help
Powered by ViewVC