ViewVC logotype

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 2053 - (show annotations) (download)
Wed Jan 27 20:29:36 2010 UTC (10 years, 8 months ago) by schoenebeck
File size: 32837 byte(s)
* Introduced support for optional environment variable
  "LINUXSAMPLER_PLUGIN_DIR" which allows to override the directory
  where the sampler shall look for instrument editor plugins
  (patch by Luis Garrido, slightly modified).
* bumped version to 1.0.0.cvs3

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

  ViewVC Help
Powered by ViewVC