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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1993 - (show annotations) (download)
Sun Aug 30 11:27:35 2009 UTC (14 years, 7 months ago) by schoenebeck
File size: 31428 byte(s)
* ALSA MIDI driver supports now "NAME" device parameter, for overriding
  the ALSA sequencer client name

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

  ViewVC Help
Powered by ViewVC