ViewVC logotype

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 2123 - (show annotations) (download)
Wed Sep 15 10:04:07 2010 UTC (10 years, 10 months ago) by schoenebeck
File size: 35331 byte(s)
* fixed crash which occured when changing an already deployed sampler
  channel to a different engine type (e.g. from gig to sf2)
* bumped version to 1.0.0cvs5

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

  ViewVC Help
Powered by ViewVC