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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 2072 - (show annotations) (download)
Sat Mar 20 11:37:52 2010 UTC (14 years ago) by persson
File size: 33630 byte(s)
* sfz engine: added support for random, seq_position, seq_length and
  volume
* sfz parser: added v1 LFO opcodes (no support in engine yet)

1 Version CVS HEAD (?)
2
3 * General changes:
4 - Refactoring: moved the independent code from
5 the Gigasampler format engine to base classes
6 - command line option '--profile' is currently disabled, since the
7 respective profiling code is currently broken
8 - Introduced support for optional environment variable
9 "LINUXSAMPLER_PLUGIN_DIR" which allows to override the directory
10 where the sampler shall look for instrument editor plugins
11 (patch by Luis Garrido, slightly modified).
12
13 * Gigasampler format engine:
14 - implemented the "round robin keyboard" dimension
15 - fixed round robin and random dimensions for cases when number of
16 dimension zones is not a power of two
17 - made round robin use a counter for each region instead of each
18 key
19
20 * SFZ format engine:
21 - Initial implementation (not usable yet)
22 - added support for v2 multiple stage envelope generators
23 - added a fine-tuned v1 envelope generator instead of using the
24 one from the gig engine
25 - fixed memory leak and memory handling errors
26 - added support for trigger=first, trigger=legato and sw_previous
27 - allow non-numerical key values ("C#4" for example)
28 - "key" opcode now sets pitch_keycenter too
29 - fixed error when unloading instrument with same sample used by
30 multiple regions
31 - added some opcode aliases, like loopmode for loop_mode, to be
32 more compatible
33 - added support for trigger=release and rt_decay
34 - added support for off_mode=normal
35 - added support for random, seq_position, seq_length and volume
36 - added v1 LFO opcodes to parser (no support in engine yet)
37
38 * SoundFont format engine:
39 - Initial implementation (not usable yet)
40
41 * Host plugins:
42 - AU bugfix: failed to destroy its audio/MIDI devices
43 - Listen to all interfaces on Mac OS X (INADDR_ANY)
44 - VST bugfix: If the host called resume() before and after
45 changing sample rate or block size, the number of channels was
46 incorrectly set to two. This caused silence in Cubase 5.
47
48 * MIDI driver:
49 - ALSA MIDI driver supports now "NAME" device parameter, for overriding
50 the ALSA sequencer client name
51
52 * LSCP server:
53 - added support for sending MIDI CC messages via LSCP command
54 "SEND CHANNEL MIDI_DATA CC <sampler-chan> <ctrl> <val>"
55
56 * Bug fixes:
57 - Fixed crash which may occur when MIDI key + transpose is out of range
58
59 Version 1.0.0 (31 July 2009)
60
61 * packaging changes:
62 - autoconf bugfix: the PKG_CONFIG variable wasn't initialized properly,
63 causing e.g. the libgig test to fail when
64 "./configure --disable-jack-driver" was used
65 (patch by Alexis Ballier)
66 - fixed compilation with gcc 4.3
67 - fixes for building on OS X (thanks to Ebrahim Mayat for testing)
68 - fixed configure so it detects x86_64 (#107)
69 - fixes for building with newer MinGW versions
70 - fix for building with bison 2.4 (#111)
71 - fixed building with libgig installed in a non-standard directory
72 - minor fix in configure for mmsystem.h detection on MinGW
73 - Windows: look for editor plugins and Fantasia using base
74 directory of liblinuxsampler dll (look in the same directory and one
75 directory above)
76 - configure script fix: removed unconditional use of SSE
77 - fixed building with sqlite installed in a non-standard directory
78 - when cross-compiling, don't try to create instruments.db
79 - fix for new mingw-w64 version, which has usleep
80
81 * general changes:
82 - bugfix: on some POSIX systems instrument editor plugins refused to
83 load as we used a non-portable Linux specific struct field
84 (fixes bug #70, patch by Ronald Baljeu)
85 - fixed endless loop which occured when loading an instrument editor
86 plugin DLL on Windows systems failed
87 - fixed memory leaks that occurred when liblinuxsampler was unloaded
88 - optimized the SynchronizedConfig class so it doesn't wait
89 unnecessarily long after an update
90 - added support for notifying instrument editors on note-on / note-off
91 events (e.g. to highlight the pressed keys on the virtual keyboard
92 of gigedit)
93 - added support for triggering notes by instrument editors (see above)
94 - be verbose on DLL load errors (on Linux)
95 - fixed di-harmonic triangle LFO implementation (this LFO implementation
96 is only used by very few systems, most pick the int math
97 implementation, which is usually faster)
98 - fixes for audio drivers with varying buffer sizes
99 - experimental support for running LinuxSampler as a DSSI, LV2 and
100 VST plugin
101 - notification events for stream/voice count statistics are now sent
102 only when there are actual changes
103 - added memory ordering constraints to improve stability on
104 multi-core and multi-cpu systems
105 - maximum voices and disk streams can now be altered at runtime
106 - fixed CPU feature detection on x86_64 (maybe fixes #108)
107 - automatic stacktrace mechanism is now turned off by default and can
108 be switched on by command line option "--stacktrace" (the automatic
109 stacktrace mechanism seems to be broken on most systems at the moment)
110 - C++ API method InstrumentManager::LaunchInstrumentEditor() now returns
111 a pointer to the launched InstrumentEditor object
112 - added optional 3rd party user data parameter for following
113 liblinuxsampler C++ API methods: InstrumentEditor::Main(),
114 InstrumentEditor::Launch(),
115 InstrumentManager::LaunchInstrumentEditor()
116 - theoretical fix: made SynchronizedConfig follow C++0x memory
117 model more strictly
118 - fixes for using large audio device buffers
119 - Windows: add the installation directory to the DLL search path
120 when loading an editor plugin (solves problems with VST and
121 gigedit on systems with other GTK versions installed)
122
123 * audio driver:
124 - removed the nonsense audio channel constraint (which was hard coded to
125 max. 100 audio channels) for most audio drivers
126 - JACK audio driver did not offer a device parameter "SAMPLERATE" as
127 opposed to the LSCP specs
128 - bugfix: the SAMPLERATE parameter of some drivers (e.g. JACK)
129 reflected the wrong value
130 - fixed a memory management bug in ASIO driver
131 - Makefile fix: JACK_CFLAGS wasn't used
132 - JACK: use jack_client_open instead of the deprecated
133 jack_client_new
134 - added (experimental) CoreAudio driver
135 - applied old fixes to the ASIO driver that were included in the
136 previous binary release but accidentally never committed to CVS
137 (fixes #117)
138 - fixes for ASIO on mingw-w64 (iasio wrapper is not needed on
139 win64)
140 - VST: added support for sample rate and buffer size changes
141 - VST: close editor (Fantasia) when the VST is removed
142 - VST: avoid opening Fantasia more than once for each VST instance
143 - VST: export main function as "main" on Linux too (fix for
144 energyXT)
145 - VST: changed number of output channels from one stereo to 16
146 stereo
147 - added channel routing, fxsends and midi maps to the settings
148 stored in the plugin state
149 - performance optimization of AudioChannel::MixTo() and
150 AudioChannel::CopyTo() methods using GCC vector exensions
151 (if available)
152 - ASIO fixes: avoid initializing the device twice, avoid throwing
153 exception when getting parameters from a disconnected device
154
155 * MIDI driver:
156 - added JACK MIDI driver
157 - dispatch bank select as ordinary CC as well, the user might seriously
158 want to (mis)use it for some purpose ("fixed" in all current MIDI
159 input drivers: ALSA, CoreMIDI, JACK, MidiShare, MME)
160 - bugfix: pitch bend wasn't working with JackMidi, VST, LV2, MME,
161 CoreMidi or AU
162 - fixed mingw-w64 compilation error in MME driver
163 - made program change handling in MIDI thread real-time safe by
164 moving the logic to a non-RT thread
165 - fixed minor memory leak in ALSA driver
166
167 * instruments database:
168 - avoid time consuming samples scanning when adding instruments
169 to the instruments database
170 - added support for handling lost files in the instruments database
171 - Implemented option for adding instruments in separate directories
172 in the instruments database
173 (patch by Chris Cherrett & Andrew Williams, a bit adjusted)
174 - work-around for missing fnmatch function on Windows to make
175 instrument database compilable
176 - added instrument database support on Windows
177 works with both standalone mode and VST plugin.
178 the instruments DB file is located in
179 %USERPROFILE%\.linuxsampler\instruments.db which allows different
180 databases for each windows user
181 if no DB is present it automatically creates the .linuxsampler subdir
182 and then creates an empty DB
183 - fixed recursive import, which was broken on Windows
184
185 * Gigasampler format engine:
186 - fixed a memory leak that could happen when a channel was deleted
187 while notes were playing
188 - made it possible to load an instrument even if the audio thread
189 isn't running
190 - added partial support for the "Controller Triggered" MIDI rule,
191 enough for piano gigs with pedal noise samples
192 - minor fix: only mark FX sends as being modified if really the
193 respective FX send MIDI controller was used
194 - added support for GM global device volume SysEx message
195 (can be selected at compile time to either apply globally to the
196 whole sampler [default] or only to the sampler channels that are
197 connected to the respective MIDI input port the SysEx message arrived
198 on)
199 - bugfix: notes triggered at position 0 in the audio buffer were
200 sometimes wrongly killed in the same buffer, causing no sound to
201 be played
202 - added support for chromatic / drumkit mode Roland GS Sysex message
203 which is usally used to switch a part between chromatic sounds and
204 drumkit sounds (as opposed to their common default setting of MIDI
205 part 10 being a drumkit part and all other ones chromatic parts), in
206 LS however one can switch between the first 16 MIDI instrument maps
207 defined for the sampler
208 - made it possible to create multiple sample channels even if the
209 audio thread isn't running
210 - fixed crash when removing channel with active voices (#116)
211 - bugfix: on sample reference changes (instrument editor), only
212 un-cache the respective sample if it's really not used by any
213 sampler engine anymore
214 - re-cache samples in case they were changed by an instrument editor,
215 e.g. when a sample was added while playing (#82)
216 - fixed hanging notes which occured when note-off event had the exact
217 same time stamp as the voice's note-on event and occured both in the
218 same audio fragment cycle (fixes bug #112)
219 - added support for the "fine tune" and "pitch bend range"
220 instrument-level gig parameters
221 - fixed minor artifacts in pitch bend handling
222 - added support for GS Reset SysEx message
223 - allow gig files to use unlimited downward pitch shifting
224 - added a limit check for upward pitch shifting
225 - bugfix: sometimes, when playing a note twice fast, the second
226 note was silent
227 - fixed crash happening when a pitch bend event arrived at the
228 same time a new instrument was loading
229
230 * LSCP server:
231 - added new LSCP event "CHANNEL_MIDI" which can be used by frontends to
232 react on MIDI data arriving on certain sampler channels (so far only
233 Note-On and Note-Off events are sent via this LSCP event)
234 - added new LSCP event "DEVICE_MIDI" which can be used by frontends to
235 react on MIDI data arriving on certain MIDI input devices (so far only
236 Note-On and Note-Off events are sent via this LSCP event)
237 - added new LSCP commands: FIND LOST DB_INSTRUMENT_FILES and
238 SET DB_INSTRUMENT FILE_PATH
239 - added new LSCP command: SEND CHANNEL MIDI_DATA which can be used by
240 frontends to send MIDI messages to specific sampler channel
241 - added two additional fields to GET FILE INSTRUMENT INFO command -
242 KEY_BINDINGS and KEYSWITCH_BINDINGS
243 - bugfix: the bank number provided by MIDI_INSTRUMENT_INFO notifications
244 was incorrect
245 - Optimized the retrieval of the MIDI instrument mappings
246 - added new LSCP command "SET VOICES" to globally alter the maximum
247 amount of voices, added new LSCP event "GLOBAL_INFO:VOICES" which will
248 be triggered respectively
249 - added new LSCP command "SET STREAMS" to globally alter the maximum
250 amount of disk streams, added new LSCP event "GLOBAL_INFO:STREAMS"
251 which will be triggered respectively
252 - bugfix: retry if "select" returns EINTR (this fixes a crash when
253 a gigedit file dialog is opened)
254 - close all connections when LSCPServer is deleted
255 - hard close of all sockets on exit
256 - bugfix: SET CHANNEL MIDI_INPUT_TYPE didn't work with the MME
257 driver
258
259 * Bug fixes:
260 - fixed a crash which occurs when removing a sampler channel waiting
261 to start instrument loading after another channel
262 - fixed a crash which occurs when removing a sampler channel with
263 instrument loading in progress (bug #113)
264 - fixed termination caused by uncaught exception when adding MIDI
265 instrument with PERSISTENT load mode
266 - fixed possible iterator invalidations when resetting the sampler
267 - fixed memory leaks when issuing the following LSCP commands:
268 GET AUDIO_OUTPUT_DRIVER INFO
269 GET AUDIO_OUTPUT_DRIVER_PARAMETER INFO
270 GET MIDI_INPUT_DRIVER INFO
271 GET MIDI_INPUT_DRIVER_PARAMETER INFO
272 - fixed possible compilation error when sqlite is not present
273 - fixed orphaned pointers when setting maximum voices limit (bug #118)
274 - fixed crash when changing the audio output device of a sampler
275 channel with loaded instrument and start playing notes
276 - fixed endless loop in Engine::SuspendAll() (bug #120)
277 - fixed a low-level atomic load function that was broken on 64-bit
278 PowerPC, which probably could cause crashes on that platform
279 - fixed a memory management error which could cause a crash when a
280 plugin was unloaded
281 - bugfix: two private structs had the same name, which could cause
282 problems if the linker chose the wrong constructor
283 - fixed low-level ConditionServer usage bug that caused lockups on
284 Windows
285
286
287 Version 0.5.1 (6 December 2007)
288
289 * packaging changes:
290 - added autoconf checks for pthread library
291 - added autoconf check for pthread bug found on certain NPTL-enabled
292 glibc versions (see Gentoo bug report #194076)
293 - added autoconf checks for MS Windows
294
295 * general changes:
296 - bugfix: the thread used by an editor plugin didn't die when the
297 editor closed
298 - bugfix: Ringbuffer.h: fill_write_space_with_null() did not zero
299 out all the space. operator--() did not apply size_mask after
300 decrementing the read_ptr. DEFAULT_WRAP_ELEMENTS set to 0 in
301 order to avoid problems with the _NonVolatileReader functions.
302 - bugfix: Stream.h: added a missing divide by BytesPerSample in
303 GetWriteSpace(). Since this function is currently only used in
304 the stream's qsort() compare function, it didn't trigger any
305 bugs.
306 - Resampler.h, Synthesizer.h: cubic interpolation now works in
307 24bit mode too. Faster method to read 24bit words on little
308 endian machines (x86): replaced 3 byte reads + shifts with a 1
309 unaligned 32bit read and shift
310 - experimental support for MS Windows (MIDI input via MME, AUDIO
311 output via ASIO)
312 - made handling of SIGINT signal (Ctrl-C) a bit more robust
313 - support for monitoring the total number of active disk streams
314 (new LSCP commands: GET TOTAL_STREAM_COUNT,
315 SUBSCRIBE TOTAL_STREAM_COUNT, UNSUBSCRIBE TOTAL_STREAM_COUNT)
316
317 * AUDIO driver:
318 - added Windows ASIO low latency audio driver
319
320 * MIDI driver:
321 - added MME Windows MIDI driver
322
323 * LSCP server:
324 - added support for Windows style path / filenames, however with
325 forward slash path separators instead of backslash
326 (i.e. "C:/foo/bar.gig")
327 - allow naughty liblscp to send non-string device parameters within
328 apostrophes as well
329 - added new LSCP commands: "GET FILE INSTRUMENTS <file>",
330 "LIST FILE INSTRUMENTS <file>" and
331 "GET FILE INSTRUMENT INFO <file> <index>" for retrieving informations
332 about an arbitrary instrument file on the system where the sampler is
333 running on
334
335 Version 0.5.0 (15 October 2007)
336
337 * packaging changes:
338 - config.h is not going to be installed along with liblinuxsampler's
339 API header files anymore
340 - only the API relevant header (and source) files will be exported to
341 the Doxygen API documentation (explicitly listed in Doxyfile.in)
342 - added completely new XCode project files for Mac OSX which is now
343 capable to execute our autoconf environment, thus no need anymore
344 to maintain the compile time configuration file (osx/version.h) for
345 OSX manually (patch by Toshi Nagata)
346 - fixed buggy boolean --enable-foo / --disable-foo configure script
347 parameters
348 - global.h now only covers global definitions that are needed for the
349 C++ API header files, all implementation internal global definitions
350 are now in global_private.h
351 - atomic.h is not exposed to the C++ API anymore
352 - no need to include config.h anymore for using LS's API header files
353 - fixed warnings in API doc generation
354
355 * general changes:
356 - replaced the old, confusing MIDI program change mechanism by a
357 flexible MIDI instrument mapper which allows to map arbitrary
358 (MIDI bank, MIDI program) pairs with arbitrary
359 (engine type, instrument file, file index) triplets which will be
360 loaded on the respective channel when such MIDI program change
361 messages arrive, beside that, each entry allows to define a life-time
362 strategy for the instrument, so the user can define whether the
363 instrument should i.e. be permanently loaded (to be able to switch
364 quickly among sounds) or i.e. loaded just on demand when the
365 respective program change arrives (to safe RAM space), as well as a
366 global volume factor for each entry, so the user can adjust the volume
367 dynamic of his mapped instrument collection without having to modify
368 the instrument files, also one can manage arbitrary amount of such
369 MIDI instrument maps and assign each sampler channel individually
370 a certain map, so that i.e. separation of normal instruments and
371 drumkits is possible
372 - new notification events for tracking changes to audio/MIDI devices,
373 MIDI instrument maps, MIDI instruments, FX sends, global volume.
374 - sampler was limited to load max. 200 instruments in the background
375 due to a constant size RingBuffer FIFO which is now replaced by a
376 dynamic (unlimited) size std::list FIFO
377 - added FX sends, these allow to route audio signals to arbitrary audio
378 output channels for being processed by external effect processors
379 (i.e. jack-rack), the send levels are controllable via arbitrary MIDI
380 controllers
381 - global (sampler wide) volume can now be controlled at runtime
382 - Implemented new, improved notification system
383 - fixed compilation errors regarding OSX
384 (patch by Toshi Nagata)
385 - implemented instruments database
386 - added support for escape sequences to the instruments database
387 - added highly experimental support for on-the-fly instrument editing
388 within the sampler's process (by using instrument editor plugins),
389 you'll notice the new "Registered instrument editors:" message on
390 startup, added a new LSCP command:
391 "EDIT CHANNEL INSTRUMENT <sampler-channel>"
392 to spawn a matching instrument editor for the instrument on the
393 given sampler channel, the plugin path can be overridden at compile
394 time with ./configure --enable-plugin-dir=/some/dir
395 - added experimental code for synchronizing instrument editors hosted
396 in the sampler's process to safely edit instruments while playing
397 without a crash (hopefully) by either suspending single regions
398 wherever possible or if unavoidable whole engine(s)
399 - fixed several issues in fundamental "Thread" class: set scheduling
400 policy and priority on thread level, set a minimum stack size for
401 thread (TODO: a reasonable value yet to be tested), bugfix: non-RT
402 threads simply inherited properties of starting thread instead of
403 setting their own policy and priority
404 - minor fix in our automatic stack trace mechanism on crashes, the
405 main process did not wait for the stack trace process to finish
406 its output
407 - fixed some minor memory leaks
408 - reenabled assembly features support, at the moment only for
409 enabling a fast denormal FPU mode (x86 platforms supporting SSE2)
410 - minor assembly fix in x86 features detection (don't use the PIC
411 register, to avoid relocations in the text segment at runtime)
412 - POSIX callback functions of Thread.h are hidden
413
414 * MIDI driver:
415 - dispatch bank select (MSB and LSB) messages
416
417 * audio driver:
418 - the ALSA audio output driver parameters now reflect the correct
419 parameter value ranges for the respective selected sound card
420 (patch by Till Wimmer, a bit fixed and extended)
421
422 * Gigasampler format engine:
423 - if a filter is used and EG2 finishes before EG1, let the voice
424 die when EG2 has finished (this fixes a problem with clicks and
425 voice starvation for some gigs)
426 - playback is no longer disabled during instrument loading
427 - all notes playing on a channel that changes its instrument keep
428 playing with the old instrument until they get a note off
429 command
430 - EG fix: a release value of zero could cause noises or crash
431 - handle MIDI coarse tuning messages (MIDI RPN #0 MSB #2 LSB)
432 - EG fine tuning: when attack is zero the EG starts at a level
433 above max sustain level, which means that there is a decay phase
434 even if sustain is 100%
435 - more EG fixes: the level could sometimes go below zero and cause
436 noises or crashes
437 - minor fix of EGDecay (patch by Toshi Nagata)
438 - fixed compiler error when --enable-override-filter-type was
439 supplied to the configure script (fixes #46)
440 - disk thread: queue sizes are now proportional to CONFIG_MAX_STREAMS
441 instead of a fix value
442 - behavior fix: on MIDI CC# 65 (portamento on / off), 126 (mono mode),
443 127 (solo mode) only kill voices if the respective mode really
444 changed
445
446 * LSCP server:
447 - fixed compile time error for old Bison versions
448 (i.e. v1.28 found on MacOS 10.4, patch by Toshi Nagata)
449 - parser now supports extended ASCII character set
450 (up to ASCII code 255, i.e. includes now umlauts and accents)
451 - filename arguments in LSCP commands now allow to use escape
452 sequences, that is directly literal as one of: \', \", \\, \n, \r,
453 \f, \t, \v, or as octal ASCII code value like \132, or as hex ASCII
454 code value like \xf2) (fixes bug #24)
455 - the following LSCP commands now also support escape sequences for at
456 least one of their text-based parameters (i.e. name, description):
457 "ADD MIDI_INSTRUMENT_MAP", "MAP MIDI_INSTRUMENT",
458 "SET MIDI_INSTRUMENT_MAP NAME", "SET FX_SEND NAME", "CREATE FX_SEND",
459 "SET DB_INSTRUMENT_DIRECTORY NAME",
460 "SET DB_INSTRUMENT_DIRECTORY DESCRIPTION", "SET DB_INSTRUMENT NAME",
461 "SET DB_INSTRUMENT DESCRIPTION", "FIND DB_INSTRUMENTS",
462 "FIND DB_INSTRUMENT_DIRECTORIES"
463 - returns verbose syntax errors (line and column where syntax error
464 occured, the unexpected character and the actually expected, possible
465 character(s), the latter only if less than 5 possibilities)
466 - made sure that LSCP syntax is not affected by gigedit locale
467 settings
468 - bugfix regarding strings parameter lists: all comma separated lists
469 of strings were treated as being one string containing commas
470 (fixes #57)
471
472 * Bug fixes:
473 - fixed crash occurring on certain LSCP scripts (Bug 39)
474 - another thread safety fix for lscp "load engine" and "set
475 channel audio output device"
476 - fixed a crash which occurs when reassigning the same engine
477 on a sampler channel with connected MIDI device
478 - fixed a crash which occurs when changing the number of ports of a MIDI
479 device connected to a sampler channel to number less then or equal
480 to the index of the port to which the sampler channel is connected.
481 - The previous bindings were not been disconnected when altering
482 the ALSA_SEQ_BINDINGS parameter. Introduced a NONE keyword for
483 unsubscribing from all bindings (e.g. ALSA_SEQ_BINDINGS=NONE).
484 - The active stream/voice count statistic was incorrect.
485 - notification events were not been sent for some sampler
486 channel changes
487 - added default min and max values to restrict the number of allowed
488 audio output channels and MIDI input ports
489 - the connection to the PCM interface is now closed when destroying
490 an audio output device
491 - files with slash in their path or filename could not be loaded
492
493 * test cases:
494 - updated and fixed (haven't been touched in a while)
495
496 Version 0.4.0 (24 November 2006)
497
498 * packaging changes:
499 - changed deprecated copyright attribute to license; added ldconfig
500 to post-(un)install steps; added devel package for liblinuxsampler;
501 to linuxsampler.spec (RPM)
502 - install necessary development header files for allowing 3rd party
503 applications to link against liblinuxsampler
504 - liblinuxsampler's API documentation can be generated with 'make docs'
505 (Doxygen required)
506 - added benchmark to automatically detect the best triangle LFO
507 implementation (currently either an integer math solution or a
508 di-harmonic approximation), automatic detection can be overriden
509 with --enable-signed-triang-algo=x and --enable-unsigned-triang-algo=x
510 configure script argument though (mandatory for cross-compilation)
511 - do not automatically pick optimized gcc flags if the user already
512 provided some on his own (as CXXFLAGS)
513 - added compile time option to disable processing of All-Notes-Off MIDI
514 messages
515 - added compile time options to allow disabling the various audio and
516 MIDI drivers
517 - fixed automatic GCC CPU switch detection on PPC
518 (patch by Ebrahim Mayat)
519
520 * Gigasampler format engine:
521 - extensive synthesis optimization
522 (reimplementation of EGs and LFO(s), removed synthesis parameter
523 prerendering and the synthesis parameter matrix in general, splitting
524 each audio fragment into subfragments now where each subfragment uses
525 constant pitch and filter coefficients. The volume coefficient is
526 linearly interpolated inside a subfragment, unless
527 --disable-interpolate-volume is set.)
528 - fine tuning of the EG modulation parameters
529 - improved filter cutoff calculation by adding support for the
530 following gig parameters: Cutoff freq (used when no cutoff
531 controller is defined), Control invert, Minimum cutoff, Velocity
532 curve and Velocity range. The keyboard tracking now scales
533 cutoff frequency instead of resonance.
534 - added support for gig parameter Resonance.
535 - fixed bug in sysex handling (patch by Juan Linietsky)
536 - added global volume attenuation of -9 dB (0.35f) to prevent clipping
537 which can be overridden with --enable-global-attenuation
538 - EG fixes: made the length of "attack hold" stage more
539 accurate. Release stage can now start before attack stage
540 ends. Cancel release didn't work when sustain was zero. Attack
541 time now has a minimal value to prevent clicks.
542 - fixed pitch changes larger than one octave
543 - fixed EG3 (pitch envelope) synthesis which was neutral all the time
544 - implemented portamento mode and solo mode (a.k.a 'mono mode'):
545 all modes can be altered via standard GM messages, that is CC5 for
546 altering portamento time, CC65 for enabling / disabling portamento
547 mode, CC126 for enabling solo mode and CC127 for disabling solo mode
548 - fine tuning of the curves for volume (CC7), pan (CC10 and gig
549 parameter) and crossfade
550 - added support for the "attenuation controller threshold" gig
551 parameter
552 - added smoothing of volume changes caused by control change
553 messages
554 - sample loop parameters are now taken from the DimensionRegion
555 instead of the wave chunk
556 - fixed keyswitching for v3 gigs with a number of keyswitch splits
557 not equal to a power of two
558 - reimplementation of the filter algorithm. The new filter is more
559 accurate and supports all gig filter types, including bandreject
560 and lowpass turbo.
561 - real support for 24 bit samples - samples are not truncated to
562 16 bits anymore
563 - support for aftertouch (channel pressure, not polyphonic
564 aftertouch)
565 - LFO1 behaviour fixed (dampening from max volume instead of
566 amplifying from 0)
567
568 * LSCP server:
569 - fixed application exit on broken pipe error (fixes bug #20)
570 - fixed the notification messages delay due to lack of
571 network activity (fixes bug #26)
572 - fixed parser bug which occured on space(s) within device parameters
573 (e.g. "SET AUDIO_OUTPUT_CHANNEL_PARAMETER 0 0 SOMEPARAM='foo bar'")
574
575 * audio driver:
576 - added aRts audio output driver (by no means RT safe)
577
578 * MIDI driver:
579 - fixed legacy sysex code which caused dispatching of MIDI SysEx
580 messages several times instead of once
581 - API extension for MIDI drivers which already supply exact time stamps
582 for events (i.e. for offline rendering based MIDI drivers)
583 - added checks for bad MIDI data
584
585 * linuxsampler application:
586 - show available sampler engine types on startup
587
588 * general changes:
589 - support for muting sampler channels and solo mode of the same, two new
590 LSCP commands ("SET CHANNEL MUTE" and "SET CHANNEL SOLO") and two new
591 fields ("MUTE" and "SOLO") for command "GET CHANNEL INFO" were
592 introduced for this, the behavior is the same like on a mixer console
593 (patch by Grigor Iliev, a bit adjusted). Also added configure option
594 --enable-process-muted-channels which can be used to enable the
595 processing of muted channels.
596 - support for sostenuto pedal
597 - support for monitoring the total number of active voices
598 (new LSCP commands: GET TOTAL_VOICE_COUNT, GET TOTAL_VOICE_COUNT_MAX,
599 SUBSCRIBE TOTAL_VOICE_COUNT, UNSUBSCRIBE TOTAL_VOICE_COUNT)
600 - fixed some memory management errors
601 - fixed some concurrency problems that could lead to crashes when
602 LSCP commands were executed
603 - fixed crash when instrument loading failed previously on the same
604 sampler channel (fixes bug #36)
605
606 Version 0.3.3 (15 July 2005)
607
608 * packaging changes:
609 - fixed compilation with gcc 4.0
610
611 * Gigasampler format engine:
612 - LFO preprocessing was performed even though the respective LFO was
613 disabled by the instrument patch (minor efficiency fix)
614 - if period time of chosen audio device is too small (< MIN_RELEASE_TIME)
615 for volume ramp downs in the same fragment (needed for current voice
616 stealing implementation) then simply inform the user with a warning
617 about possible click sounds and reduce the volume ramp down
618 appropriately instead of cancelling the audio device connection
619
620 Version 0.3.2 (24 June 2005)
621
622 * packaging changes:
623 - updated autotools build files to compile on OS X
624 (conditional compilation of CoreMIDI and MidiShare drivers)
625 - hand-crafted assembly optimization code can be disabled with
626 './configure --disable-asm' (definitely not recommended)
627 - fixed 'make dist' rule to include all necessary files
628 - require automake (>= 1.5) for 'make -f Makefile.cvs'
629 (due to 'dist-bzip2' automake option)
630
631 * Gigasampler format engine:
632 - support for the gig parameters for "release velocity response" curves
633 - fine tuning of the EGADSR envelope
634 - volume of release triggered samples now depends on note-on velocity,
635 note length and gig parameter "release trigger decay" instead of
636 note-off velocity.
637 - revised voice stealing
638 (fixes crash and endless loop caused by voice stealing)
639 - don't reset scale tuning on instrument or audio output device change
640 - handle key group conflicts right at the beginning of each fragment
641 instead of when the respective voice is actually launched
642 (fixes undefined behavior if stolen voices belonged to a key group -
643 this case was followed by a "killed voice survived" error message)
644 - fixed minor issue with null/silence samples
645 (those stole voices even though they don't need a voice at all which
646 resulted in "voice stealing didn't work out" messages)
647 - don't reset volume, pan, pitch and MIDI controller values on
648 instrument or audio output device change
649
650 * LSCP server:
651 - fixed some crashes (patch by Grigor Iliev, fixes #19)
652 - fixed LSCP event "CHANNEL_INFO" notification
653 (e.g. did not notify on volume changes or MIDI program change events)
654
655 * linuxsampler application:
656 - added command line parameters --lscp-addr and --lscp-port to override
657 default IP address and port of LSCP server
658
659 Version 0.3.1 (24 May 2005)
660
661 * initial release

  ViewVC Help
Powered by ViewVC