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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1994 - (show annotations) (download)
Sun Aug 30 13:30:01 2009 UTC (14 years, 7 months ago) by iliev
File size: 31484 byte(s)
* bugfix: AU plugin hangs on some hosts if bound to loopback

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

  ViewVC Help
Powered by ViewVC