ViewVC logotype

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 2175 - (show annotations) (download)
Mon Apr 25 08:12:36 2011 UTC (9 years, 5 months ago) by persson
File size: 38623 byte(s)
* sfz engine: implemeted filters. Filter types: lowpass, bandpass,
  bandreject and highpass. 1, 2, 4 and 6 pole filters. Opcodes:
  fil_type, cutoff, resonance, fil_veltrack, fil_keytrack,
  fil_keycenter, cutoff_cc, cutoff_chanaft.
* sfz engine: bugfix: zero ampeg_sustain didn't work
* gig engine: bugfix: pitch LFO controller "internal+aftertouch" was broken
* gig engine: bugfix: filter keyboard tracking was broken
* gig engine: filter performance fix (an unnecessary copy was made of
  the filter parameters in each sub fragment)
* ASIO driver: fixes for newer gcc versions (fix from PortAudio)

1 Version CVS HEAD (?)
3 * plugin changes:
4 - LV2 Persist + Files support (patch by David Robillard)
6 * packaging changes:
7 - fixed building with newer MinGW-w64
9 * general changes:
10 - Refactoring: moved the independent code from
11 the Gigasampler format engine to base classes
12 - command line option '--profile' is currently disabled, since the
13 respective profiling code is currently broken
14 - Introduced support for optional environment variable
15 "LINUXSAMPLER_PLUGIN_DIR" which allows to override the directory
16 where the sampler shall look for instrument editor plugins
17 (patch by Luis Garrido, slightly modified).
18 - implemented Roland GS NRPN 1ArrH which allows to set volume per note
19 - implemented Roland GS NRPN 1CrrH which allows to set pan per note
20 - implemented Roland GS NRPN 1DrrH which allows to set reverb send per
21 note (in this implementation of the sampler its simply hard routed to
22 the 1st effect send of the sampler channel, no matter what the actual
23 effect type is)
24 - implemented Roland GS NRPN 1ErrH which allows to set chorus send per
25 note (in this implementation of the sampler its simply hard routed to
26 the 2nd effect send of the sampler channel, no matter what the actual
27 effect type is)
28 - implemented support for internal LADSPA effects
29 - raised limit of program change queue from 100 to 512
30 (as suggested by Alex Stone)
32 * Gigasampler format engine:
33 - implemented the "round robin keyboard" dimension
34 - fixed round robin and random dimensions for cases when number of
35 dimension zones is not a power of two
36 - made round robin use a counter for each region instead of each
37 key
39 * SFZ format engine:
40 - Initial implementation (not usable yet)
41 - added support for v2 multiple stage envelope generators
42 - added a fine-tuned v1 envelope generator instead of using the
43 one from the gig engine
44 - fixed memory leak and memory handling errors
45 - added support for trigger=first, trigger=legato and sw_previous
46 - allow non-numerical key values ("C#4" for example)
47 - "key" opcode now sets pitch_keycenter too
48 - fixed error when unloading instrument with same sample used by
49 multiple regions
50 - added some opcode aliases, like loopmode for loop_mode, to be
51 more compatible
52 - added support for trigger=release and rt_decay
53 - added support for off_mode=normal
54 - added support for random, seq_position, seq_length and volume
55 - added v1 LFO opcodes to parser (no support in engine yet)
56 - added support for amp_veltrack and amp_velcurve_N
57 - fine-tuned the default velocity curve
58 - added support for transpose
59 - fixed crash when using sw_down/up
60 - improved logic for sw_lokey/hikey/up/down/last
61 - added more v1 aliases to parser
62 - reduced memory usage for sfz data
63 - RT-safeness: avoid malloc in audio thread
64 - fixed a bug that could cause voice stealing to fail
65 - optimized sample lookup
66 - improved support for exclusive groups (group, off_by and
67 off_mode)
68 - added support for controller triggered regions (on_locc/on_hicc)
69 - added support for loop_mode=one_shot
70 - fixed playback of 16 bit wav files on big endian CPUs
71 - added support for Ogg Vorbis sample files
72 - use loop markers from sample file if loop_start and loop_end are
73 not set in sfz file
74 - implemeted filters. Filter types: lowpass, bandpass, bandreject
75 and highpass. 1, 2, 4 and 6 pole filters. Opcodes: fil_type,
76 cutoff, resonance, fil_veltrack, fil_keytrack, fil_keycenter,
77 cutoff_cc, cutoff_chanaft.
78 - bugfix: zero ampeg_sustain didn't work
80 * SoundFont format engine:
81 - Initial implementation (not usable yet)
82 - RT-safeness: avoid malloc in audio thread
83 - fixed a bug that could cause voice stealing to fail
85 * Gigasampler format engine:
86 - bugfix: pitch LFO controller "internal+aftertouch" was broken
87 - bugfix: filter keyboard tracking was broken
88 - filter performance fix (an unnecessary copy was made of the
89 filter parameters in each sub fragment)
91 * Host plugins:
92 - AU bugfix: failed to destroy its audio/MIDI devices
93 - Listen to all interfaces on Mac OS X (INADDR_ANY)
94 - VST bugfix: If the host called resume() before and after
95 changing sample rate or block size, the number of channels was
96 incorrectly set to two. This caused silence in Cubase 5.
97 - save engine type (gig, sfz or sf2) in plugin state
98 - VST: when opening Fantasia, look for both 32 and 64 bit Java on
99 64 bit Windows
101 * MIDI driver:
102 - ALSA MIDI driver supports now "NAME" device parameter, for overriding
103 the ALSA sequencer client name
104 - removed limit of maximum amount of MIDI ports per MIDI device, since
105 there is no reason for this limit
107 * audio driver:
108 - ASIO driver fixes for newer gcc versions (fix from PortAudio)
110 * LSCP server:
111 - added support for sending MIDI CC messages via LSCP command
112 "SEND CHANNEL MIDI_DATA CC <sampler-chan> <ctrl> <val>"
113 - added LSCP command "GET AVAILABLE_EFFECTS"
114 - added LSCP command "LIST AVAILABLE_EFFECTS"
115 - added LSCP command "GET EFFECT INFO <effect-index>"
116 - added LSCP command "CREATE EFFECT_INSTANCE <effect-index>"
117 - added LSCP command
118 "CREATE EFFECT_INSTANCE <effect-system> <module> <effect-name>"
119 - added LSCP command "DESTROY EFFECT_INSTANCE <effect-instance>"
120 - added LSCP command "GET EFFECT_INSTANCES"
121 - added LSCP command "LIST EFFECT_INSTANCES"
122 - added LSCP command "GET EFFECT_INSTANCE INFO <effect-instance>"
123 - added LSCP command
124 "GET EFFECT_INSTANCE_INPUT_CONTROL INFO <effect-instance> <input-control>"
126 <effect-instance> <input-control> <value>"
127 - added LSCP command "GET SEND_EFFECT_CHAINS <audio-device>"
128 - added LSCP command "LIST SEND_EFFECT_CHAINS <audio-device>"
129 - added LSCP command "ADD SEND_EFFECT_CHAIN <audio-device>"
130 - added LSCP command
131 "REMOVE SEND_EFFECT_CHAIN <audio-device> <effect-chain>"
132 - added LSCP command
133 "GET SEND_EFFECT_CHAIN INFO <audio-device> <effect-chain>"
134 - added LSCP command "APPEND SEND_EFFECT_CHAIN EFFECT <audio-device>
135 <effect-chain> <effect-instance>"
136 - added LSCP command "INSERT SEND_EFFECT_CHAIN EFFECT <audio-device>
137 <effect-chain> <effect-chain-pos> <effect-instance>"
138 - added LSCP command "REMOVE SEND_EFFECT_CHAIN EFFECT <audio-device>
139 <effect-chain> <chain-pos>"
140 - added LSCP command "SET FX_SEND EFFECT <sampler_channel>
141 <fx_send_id> <effect_chain> <chain_pos>"
142 - added LSCP command "REMOVE FX_SEND EFFECT <sampler_channel> <fx_send_id>"
144 * Bug fixes:
145 - Fixed crash which may occur when MIDI key + transpose is out of range
146 - minor valgrind fixes
147 - fixed crash which occured when changing an already deployed sampler
148 channel to a different engine type
149 - fixed crash when deleting a sampler channel or changing engine
150 type while an instrument load was in progress
151 - bugfix: playing a note while changing the instrument could cause
152 a crash, or give "Handing back unknown region" error messages
153 - bugfix: calling SET VOICES (which Fantasia does on start and
154 refresh) could cause instruments to be unloaded
156 Version 1.0.0 (31 July 2009)
158 * packaging changes:
159 - autoconf bugfix: the PKG_CONFIG variable wasn't initialized properly,
160 causing e.g. the libgig test to fail when
161 "./configure --disable-jack-driver" was used
162 (patch by Alexis Ballier)
163 - fixed compilation with gcc 4.3
164 - fixes for building on OS X (thanks to Ebrahim Mayat for testing)
165 - fixed configure so it detects x86_64 (#107)
166 - fixes for building with newer MinGW versions
167 - fix for building with bison 2.4 (#111)
168 - fixed building with libgig installed in a non-standard directory
169 - minor fix in configure for mmsystem.h detection on MinGW
170 - Windows: look for editor plugins and Fantasia using base
171 directory of liblinuxsampler dll (look in the same directory and one
172 directory above)
173 - configure script fix: removed unconditional use of SSE
174 - fixed building with sqlite installed in a non-standard directory
175 - when cross-compiling, don't try to create instruments.db
176 - fix for new mingw-w64 version, which has usleep
178 * general changes:
179 - bugfix: on some POSIX systems instrument editor plugins refused to
180 load as we used a non-portable Linux specific struct field
181 (fixes bug #70, patch by Ronald Baljeu)
182 - fixed endless loop which occured when loading an instrument editor
183 plugin DLL on Windows systems failed
184 - fixed memory leaks that occurred when liblinuxsampler was unloaded
185 - optimized the SynchronizedConfig class so it doesn't wait
186 unnecessarily long after an update
187 - added support for notifying instrument editors on note-on / note-off
188 events (e.g. to highlight the pressed keys on the virtual keyboard
189 of gigedit)
190 - added support for triggering notes by instrument editors (see above)
191 - be verbose on DLL load errors (on Linux)
192 - fixed di-harmonic triangle LFO implementation (this LFO implementation
193 is only used by very few systems, most pick the int math
194 implementation, which is usually faster)
195 - fixes for audio drivers with varying buffer sizes
196 - experimental support for running LinuxSampler as a DSSI, LV2 and
197 VST plugin
198 - notification events for stream/voice count statistics are now sent
199 only when there are actual changes
200 - added memory ordering constraints to improve stability on
201 multi-core and multi-cpu systems
202 - maximum voices and disk streams can now be altered at runtime
203 - fixed CPU feature detection on x86_64 (maybe fixes #108)
204 - automatic stacktrace mechanism is now turned off by default and can
205 be switched on by command line option "--stacktrace" (the automatic
206 stacktrace mechanism seems to be broken on most systems at the moment)
207 - C++ API method InstrumentManager::LaunchInstrumentEditor() now returns
208 a pointer to the launched InstrumentEditor object
209 - added optional 3rd party user data parameter for following
210 liblinuxsampler C++ API methods: InstrumentEditor::Main(),
211 InstrumentEditor::Launch(),
212 InstrumentManager::LaunchInstrumentEditor()
213 - theoretical fix: made SynchronizedConfig follow C++0x memory
214 model more strictly
215 - fixes for using large audio device buffers
216 - Windows: add the installation directory to the DLL search path
217 when loading an editor plugin (solves problems with VST and
218 gigedit on systems with other GTK versions installed)
220 * audio driver:
221 - removed the nonsense audio channel constraint (which was hard coded to
222 max. 100 audio channels) for most audio drivers
223 - JACK audio driver did not offer a device parameter "SAMPLERATE" as
224 opposed to the LSCP specs
225 - bugfix: the SAMPLERATE parameter of some drivers (e.g. JACK)
226 reflected the wrong value
227 - fixed a memory management bug in ASIO driver
228 - Makefile fix: JACK_CFLAGS wasn't used
229 - JACK: use jack_client_open instead of the deprecated
230 jack_client_new
231 - added (experimental) CoreAudio driver
232 - applied old fixes to the ASIO driver that were included in the
233 previous binary release but accidentally never committed to CVS
234 (fixes #117)
235 - fixes for ASIO on mingw-w64 (iasio wrapper is not needed on
236 win64)
237 - VST: added support for sample rate and buffer size changes
238 - VST: close editor (Fantasia) when the VST is removed
239 - VST: avoid opening Fantasia more than once for each VST instance
240 - VST: export main function as "main" on Linux too (fix for
241 energyXT)
242 - VST: changed number of output channels from one stereo to 16
243 stereo
244 - added channel routing, fxsends and midi maps to the settings
245 stored in the plugin state
246 - performance optimization of AudioChannel::MixTo() and
247 AudioChannel::CopyTo() methods using GCC vector exensions
248 (if available)
249 - ASIO fixes: avoid initializing the device twice, avoid throwing
250 exception when getting parameters from a disconnected device
252 * MIDI driver:
253 - added JACK MIDI driver
254 - dispatch bank select as ordinary CC as well, the user might seriously
255 want to (mis)use it for some purpose ("fixed" in all current MIDI
256 input drivers: ALSA, CoreMIDI, JACK, MidiShare, MME)
257 - bugfix: pitch bend wasn't working with JackMidi, VST, LV2, MME,
258 CoreMidi or AU
259 - fixed mingw-w64 compilation error in MME driver
260 - made program change handling in MIDI thread real-time safe by
261 moving the logic to a non-RT thread
262 - fixed minor memory leak in ALSA driver
264 * instruments database:
265 - avoid time consuming samples scanning when adding instruments
266 to the instruments database
267 - added support for handling lost files in the instruments database
268 - Implemented option for adding instruments in separate directories
269 in the instruments database
270 (patch by Chris Cherrett & Andrew Williams, a bit adjusted)
271 - work-around for missing fnmatch function on Windows to make
272 instrument database compilable
273 - added instrument database support on Windows
274 works with both standalone mode and VST plugin.
275 the instruments DB file is located in
276 %USERPROFILE%\.linuxsampler\instruments.db which allows different
277 databases for each windows user
278 if no DB is present it automatically creates the .linuxsampler subdir
279 and then creates an empty DB
280 - fixed recursive import, which was broken on Windows
282 * Gigasampler format engine:
283 - fixed a memory leak that could happen when a channel was deleted
284 while notes were playing
285 - made it possible to load an instrument even if the audio thread
286 isn't running
287 - added partial support for the "Controller Triggered" MIDI rule,
288 enough for piano gigs with pedal noise samples
289 - minor fix: only mark FX sends as being modified if really the
290 respective FX send MIDI controller was used
291 - added support for GM global device volume SysEx message
292 (can be selected at compile time to either apply globally to the
293 whole sampler [default] or only to the sampler channels that are
294 connected to the respective MIDI input port the SysEx message arrived
295 on)
296 - bugfix: notes triggered at position 0 in the audio buffer were
297 sometimes wrongly killed in the same buffer, causing no sound to
298 be played
299 - added support for chromatic / drumkit mode Roland GS Sysex message
300 which is usally used to switch a part between chromatic sounds and
301 drumkit sounds (as opposed to their common default setting of MIDI
302 part 10 being a drumkit part and all other ones chromatic parts), in
303 LS however one can switch between the first 16 MIDI instrument maps
304 defined for the sampler
305 - made it possible to create multiple sample channels even if the
306 audio thread isn't running
307 - fixed crash when removing channel with active voices (#116)
308 - bugfix: on sample reference changes (instrument editor), only
309 un-cache the respective sample if it's really not used by any
310 sampler engine anymore
311 - re-cache samples in case they were changed by an instrument editor,
312 e.g. when a sample was added while playing (#82)
313 - fixed hanging notes which occured when note-off event had the exact
314 same time stamp as the voice's note-on event and occured both in the
315 same audio fragment cycle (fixes bug #112)
316 - added support for the "fine tune" and "pitch bend range"
317 instrument-level gig parameters
318 - fixed minor artifacts in pitch bend handling
319 - added support for GS Reset SysEx message
320 - allow gig files to use unlimited downward pitch shifting
321 - added a limit check for upward pitch shifting
322 - bugfix: sometimes, when playing a note twice fast, the second
323 note was silent
324 - fixed crash happening when a pitch bend event arrived at the
325 same time a new instrument was loading
327 * LSCP server:
328 - added new LSCP event "CHANNEL_MIDI" which can be used by frontends to
329 react on MIDI data arriving on certain sampler channels (so far only
330 Note-On and Note-Off events are sent via this LSCP event)
331 - added new LSCP event "DEVICE_MIDI" which can be used by frontends to
332 react on MIDI data arriving on certain MIDI input devices (so far only
333 Note-On and Note-Off events are sent via this LSCP event)
334 - added new LSCP commands: FIND LOST DB_INSTRUMENT_FILES and
336 - added new LSCP command: SEND CHANNEL MIDI_DATA which can be used by
337 frontends to send MIDI messages to specific sampler channel
338 - added two additional fields to GET FILE INSTRUMENT INFO command -
340 - bugfix: the bank number provided by MIDI_INSTRUMENT_INFO notifications
341 was incorrect
342 - Optimized the retrieval of the MIDI instrument mappings
343 - added new LSCP command "SET VOICES" to globally alter the maximum
344 amount of voices, added new LSCP event "GLOBAL_INFO:VOICES" which will
345 be triggered respectively
346 - added new LSCP command "SET STREAMS" to globally alter the maximum
347 amount of disk streams, added new LSCP event "GLOBAL_INFO:STREAMS"
348 which will be triggered respectively
349 - bugfix: retry if "select" returns EINTR (this fixes a crash when
350 a gigedit file dialog is opened)
351 - close all connections when LSCPServer is deleted
352 - hard close of all sockets on exit
353 - bugfix: SET CHANNEL MIDI_INPUT_TYPE didn't work with the MME
354 driver
356 * Bug fixes:
357 - fixed a crash which occurs when removing a sampler channel waiting
358 to start instrument loading after another channel
359 - fixed a crash which occurs when removing a sampler channel with
360 instrument loading in progress (bug #113)
361 - fixed termination caused by uncaught exception when adding MIDI
362 instrument with PERSISTENT load mode
363 - fixed possible iterator invalidations when resetting the sampler
364 - fixed memory leaks when issuing the following LSCP commands:
369 - fixed possible compilation error when sqlite is not present
370 - fixed orphaned pointers when setting maximum voices limit (bug #118)
371 - fixed crash when changing the audio output device of a sampler
372 channel with loaded instrument and start playing notes
373 - fixed endless loop in Engine::SuspendAll() (bug #120)
374 - fixed a low-level atomic load function that was broken on 64-bit
375 PowerPC, which probably could cause crashes on that platform
376 - fixed a memory management error which could cause a crash when a
377 plugin was unloaded
378 - bugfix: two private structs had the same name, which could cause
379 problems if the linker chose the wrong constructor
380 - fixed low-level ConditionServer usage bug that caused lockups on
381 Windows
384 Version 0.5.1 (6 December 2007)
386 * packaging changes:
387 - added autoconf checks for pthread library
388 - added autoconf check for pthread bug found on certain NPTL-enabled
389 glibc versions (see Gentoo bug report #194076)
390 - added autoconf checks for MS Windows
392 * general changes:
393 - bugfix: the thread used by an editor plugin didn't die when the
394 editor closed
395 - bugfix: Ringbuffer.h: fill_write_space_with_null() did not zero
396 out all the space. operator--() did not apply size_mask after
397 decrementing the read_ptr. DEFAULT_WRAP_ELEMENTS set to 0 in
398 order to avoid problems with the _NonVolatileReader functions.
399 - bugfix: Stream.h: added a missing divide by BytesPerSample in
400 GetWriteSpace(). Since this function is currently only used in
401 the stream's qsort() compare function, it didn't trigger any
402 bugs.
403 - Resampler.h, Synthesizer.h: cubic interpolation now works in
404 24bit mode too. Faster method to read 24bit words on little
405 endian machines (x86): replaced 3 byte reads + shifts with a 1
406 unaligned 32bit read and shift
407 - experimental support for MS Windows (MIDI input via MME, AUDIO
408 output via ASIO)
409 - made handling of SIGINT signal (Ctrl-C) a bit more robust
410 - support for monitoring the total number of active disk streams
411 (new LSCP commands: GET TOTAL_STREAM_COUNT,
414 * AUDIO driver:
415 - added Windows ASIO low latency audio driver
417 * MIDI driver:
418 - added MME Windows MIDI driver
420 * LSCP server:
421 - added support for Windows style path / filenames, however with
422 forward slash path separators instead of backslash
423 (i.e. "C:/foo/bar.gig")
424 - allow naughty liblscp to send non-string device parameters within
425 apostrophes as well
426 - added new LSCP commands: "GET FILE INSTRUMENTS <file>",
427 "LIST FILE INSTRUMENTS <file>" and
428 "GET FILE INSTRUMENT INFO <file> <index>" for retrieving informations
429 about an arbitrary instrument file on the system where the sampler is
430 running on
432 Version 0.5.0 (15 October 2007)
434 * packaging changes:
435 - config.h is not going to be installed along with liblinuxsampler's
436 API header files anymore
437 - only the API relevant header (and source) files will be exported to
438 the Doxygen API documentation (explicitly listed in Doxyfile.in)
439 - added completely new XCode project files for Mac OSX which is now
440 capable to execute our autoconf environment, thus no need anymore
441 to maintain the compile time configuration file (osx/version.h) for
442 OSX manually (patch by Toshi Nagata)
443 - fixed buggy boolean --enable-foo / --disable-foo configure script
444 parameters
445 - global.h now only covers global definitions that are needed for the
446 C++ API header files, all implementation internal global definitions
447 are now in global_private.h
448 - atomic.h is not exposed to the C++ API anymore
449 - no need to include config.h anymore for using LS's API header files
450 - fixed warnings in API doc generation
452 * general changes:
453 - replaced the old, confusing MIDI program change mechanism by a
454 flexible MIDI instrument mapper which allows to map arbitrary
455 (MIDI bank, MIDI program) pairs with arbitrary
456 (engine type, instrument file, file index) triplets which will be
457 loaded on the respective channel when such MIDI program change
458 messages arrive, beside that, each entry allows to define a life-time
459 strategy for the instrument, so the user can define whether the
460 instrument should i.e. be permanently loaded (to be able to switch
461 quickly among sounds) or i.e. loaded just on demand when the
462 respective program change arrives (to safe RAM space), as well as a
463 global volume factor for each entry, so the user can adjust the volume
464 dynamic of his mapped instrument collection without having to modify
465 the instrument files, also one can manage arbitrary amount of such
466 MIDI instrument maps and assign each sampler channel individually
467 a certain map, so that i.e. separation of normal instruments and
468 drumkits is possible
469 - new notification events for tracking changes to audio/MIDI devices,
470 MIDI instrument maps, MIDI instruments, FX sends, global volume.
471 - sampler was limited to load max. 200 instruments in the background
472 due to a constant size RingBuffer FIFO which is now replaced by a
473 dynamic (unlimited) size std::list FIFO
474 - added FX sends, these allow to route audio signals to arbitrary audio
475 output channels for being processed by external effect processors
476 (i.e. jack-rack), the send levels are controllable via arbitrary MIDI
477 controllers
478 - global (sampler wide) volume can now be controlled at runtime
479 - Implemented new, improved notification system
480 - fixed compilation errors regarding OSX
481 (patch by Toshi Nagata)
482 - implemented instruments database
483 - added support for escape sequences to the instruments database
484 - added highly experimental support for on-the-fly instrument editing
485 within the sampler's process (by using instrument editor plugins),
486 you'll notice the new "Registered instrument editors:" message on
487 startup, added a new LSCP command:
488 "EDIT CHANNEL INSTRUMENT <sampler-channel>"
489 to spawn a matching instrument editor for the instrument on the
490 given sampler channel, the plugin path can be overridden at compile
491 time with ./configure --enable-plugin-dir=/some/dir
492 - added experimental code for synchronizing instrument editors hosted
493 in the sampler's process to safely edit instruments while playing
494 without a crash (hopefully) by either suspending single regions
495 wherever possible or if unavoidable whole engine(s)
496 - fixed several issues in fundamental "Thread" class: set scheduling
497 policy and priority on thread level, set a minimum stack size for
498 thread (TODO: a reasonable value yet to be tested), bugfix: non-RT
499 threads simply inherited properties of starting thread instead of
500 setting their own policy and priority
501 - minor fix in our automatic stack trace mechanism on crashes, the
502 main process did not wait for the stack trace process to finish
503 its output
504 - fixed some minor memory leaks
505 - reenabled assembly features support, at the moment only for
506 enabling a fast denormal FPU mode (x86 platforms supporting SSE2)
507 - minor assembly fix in x86 features detection (don't use the PIC
508 register, to avoid relocations in the text segment at runtime)
509 - POSIX callback functions of Thread.h are hidden
511 * MIDI driver:
512 - dispatch bank select (MSB and LSB) messages
514 * audio driver:
515 - the ALSA audio output driver parameters now reflect the correct
516 parameter value ranges for the respective selected sound card
517 (patch by Till Wimmer, a bit fixed and extended)
519 * Gigasampler format engine:
520 - if a filter is used and EG2 finishes before EG1, let the voice
521 die when EG2 has finished (this fixes a problem with clicks and
522 voice starvation for some gigs)
523 - playback is no longer disabled during instrument loading
524 - all notes playing on a channel that changes its instrument keep
525 playing with the old instrument until they get a note off
526 command
527 - EG fix: a release value of zero could cause noises or crash
528 - handle MIDI coarse tuning messages (MIDI RPN #0 MSB #2 LSB)
529 - EG fine tuning: when attack is zero the EG starts at a level
530 above max sustain level, which means that there is a decay phase
531 even if sustain is 100%
532 - more EG fixes: the level could sometimes go below zero and cause
533 noises or crashes
534 - minor fix of EGDecay (patch by Toshi Nagata)
535 - fixed compiler error when --enable-override-filter-type was
536 supplied to the configure script (fixes #46)
537 - disk thread: queue sizes are now proportional to CONFIG_MAX_STREAMS
538 instead of a fix value
539 - behavior fix: on MIDI CC# 65 (portamento on / off), 126 (mono mode),
540 127 (solo mode) only kill voices if the respective mode really
541 changed
543 * LSCP server:
544 - fixed compile time error for old Bison versions
545 (i.e. v1.28 found on MacOS 10.4, patch by Toshi Nagata)
546 - parser now supports extended ASCII character set
547 (up to ASCII code 255, i.e. includes now umlauts and accents)
548 - filename arguments in LSCP commands now allow to use escape
549 sequences, that is directly literal as one of: \', \", \\, \n, \r,
550 \f, \t, \v, or as octal ASCII code value like \132, or as hex ASCII
551 code value like \xf2) (fixes bug #24)
552 - the following LSCP commands now also support escape sequences for at
553 least one of their text-based parameters (i.e. name, description):
560 - returns verbose syntax errors (line and column where syntax error
561 occured, the unexpected character and the actually expected, possible
562 character(s), the latter only if less than 5 possibilities)
563 - made sure that LSCP syntax is not affected by gigedit locale
564 settings
565 - bugfix regarding strings parameter lists: all comma separated lists
566 of strings were treated as being one string containing commas
567 (fixes #57)
569 * Bug fixes:
570 - fixed crash occurring on certain LSCP scripts (Bug 39)
571 - another thread safety fix for lscp "load engine" and "set
572 channel audio output device"
573 - fixed a crash which occurs when reassigning the same engine
574 on a sampler channel with connected MIDI device
575 - fixed a crash which occurs when changing the number of ports of a MIDI
576 device connected to a sampler channel to number less then or equal
577 to the index of the port to which the sampler channel is connected.
578 - The previous bindings were not been disconnected when altering
579 the ALSA_SEQ_BINDINGS parameter. Introduced a NONE keyword for
580 unsubscribing from all bindings (e.g. ALSA_SEQ_BINDINGS=NONE).
581 - The active stream/voice count statistic was incorrect.
582 - notification events were not been sent for some sampler
583 channel changes
584 - added default min and max values to restrict the number of allowed
585 audio output channels and MIDI input ports
586 - the connection to the PCM interface is now closed when destroying
587 an audio output device
588 - files with slash in their path or filename could not be loaded
590 * test cases:
591 - updated and fixed (haven't been touched in a while)
593 Version 0.4.0 (24 November 2006)
595 * packaging changes:
596 - changed deprecated copyright attribute to license; added ldconfig
597 to post-(un)install steps; added devel package for liblinuxsampler;
598 to linuxsampler.spec (RPM)
599 - install necessary development header files for allowing 3rd party
600 applications to link against liblinuxsampler
601 - liblinuxsampler's API documentation can be generated with 'make docs'
602 (Doxygen required)
603 - added benchmark to automatically detect the best triangle LFO
604 implementation (currently either an integer math solution or a
605 di-harmonic approximation), automatic detection can be overriden
606 with --enable-signed-triang-algo=x and --enable-unsigned-triang-algo=x
607 configure script argument though (mandatory for cross-compilation)
608 - do not automatically pick optimized gcc flags if the user already
609 provided some on his own (as CXXFLAGS)
610 - added compile time option to disable processing of All-Notes-Off MIDI
611 messages
612 - added compile time options to allow disabling the various audio and
613 MIDI drivers
614 - fixed automatic GCC CPU switch detection on PPC
615 (patch by Ebrahim Mayat)
617 * Gigasampler format engine:
618 - extensive synthesis optimization
619 (reimplementation of EGs and LFO(s), removed synthesis parameter
620 prerendering and the synthesis parameter matrix in general, splitting
621 each audio fragment into subfragments now where each subfragment uses
622 constant pitch and filter coefficients. The volume coefficient is
623 linearly interpolated inside a subfragment, unless
624 --disable-interpolate-volume is set.)
625 - fine tuning of the EG modulation parameters
626 - improved filter cutoff calculation by adding support for the
627 following gig parameters: Cutoff freq (used when no cutoff
628 controller is defined), Control invert, Minimum cutoff, Velocity
629 curve and Velocity range. The keyboard tracking now scales
630 cutoff frequency instead of resonance.
631 - added support for gig parameter Resonance.
632 - fixed bug in sysex handling (patch by Juan Linietsky)
633 - added global volume attenuation of -9 dB (0.35f) to prevent clipping
634 which can be overridden with --enable-global-attenuation
635 - EG fixes: made the length of "attack hold" stage more
636 accurate. Release stage can now start before attack stage
637 ends. Cancel release didn't work when sustain was zero. Attack
638 time now has a minimal value to prevent clicks.
639 - fixed pitch changes larger than one octave
640 - fixed EG3 (pitch envelope) synthesis which was neutral all the time
641 - implemented portamento mode and solo mode (a.k.a 'mono mode'):
642 all modes can be altered via standard GM messages, that is CC5 for
643 altering portamento time, CC65 for enabling / disabling portamento
644 mode, CC126 for enabling solo mode and CC127 for disabling solo mode
645 - fine tuning of the curves for volume (CC7), pan (CC10 and gig
646 parameter) and crossfade
647 - added support for the "attenuation controller threshold" gig
648 parameter
649 - added smoothing of volume changes caused by control change
650 messages
651 - sample loop parameters are now taken from the DimensionRegion
652 instead of the wave chunk
653 - fixed keyswitching for v3 gigs with a number of keyswitch splits
654 not equal to a power of two
655 - reimplementation of the filter algorithm. The new filter is more
656 accurate and supports all gig filter types, including bandreject
657 and lowpass turbo.
658 - real support for 24 bit samples - samples are not truncated to
659 16 bits anymore
660 - support for aftertouch (channel pressure, not polyphonic
661 aftertouch)
662 - LFO1 behaviour fixed (dampening from max volume instead of
663 amplifying from 0)
665 * LSCP server:
666 - fixed application exit on broken pipe error (fixes bug #20)
667 - fixed the notification messages delay due to lack of
668 network activity (fixes bug #26)
669 - fixed parser bug which occured on space(s) within device parameters
672 * audio driver:
673 - added aRts audio output driver (by no means RT safe)
675 * MIDI driver:
676 - fixed legacy sysex code which caused dispatching of MIDI SysEx
677 messages several times instead of once
678 - API extension for MIDI drivers which already supply exact time stamps
679 for events (i.e. for offline rendering based MIDI drivers)
680 - added checks for bad MIDI data
682 * linuxsampler application:
683 - show available sampler engine types on startup
685 * general changes:
686 - support for muting sampler channels and solo mode of the same, two new
687 LSCP commands ("SET CHANNEL MUTE" and "SET CHANNEL SOLO") and two new
688 fields ("MUTE" and "SOLO") for command "GET CHANNEL INFO" were
689 introduced for this, the behavior is the same like on a mixer console
690 (patch by Grigor Iliev, a bit adjusted). Also added configure option
691 --enable-process-muted-channels which can be used to enable the
692 processing of muted channels.
693 - support for sostenuto pedal
694 - support for monitoring the total number of active voices
697 - fixed some memory management errors
698 - fixed some concurrency problems that could lead to crashes when
699 LSCP commands were executed
700 - fixed crash when instrument loading failed previously on the same
701 sampler channel (fixes bug #36)
703 Version 0.3.3 (15 July 2005)
705 * packaging changes:
706 - fixed compilation with gcc 4.0
708 * Gigasampler format engine:
709 - LFO preprocessing was performed even though the respective LFO was
710 disabled by the instrument patch (minor efficiency fix)
711 - if period time of chosen audio device is too small (< MIN_RELEASE_TIME)
712 for volume ramp downs in the same fragment (needed for current voice
713 stealing implementation) then simply inform the user with a warning
714 about possible click sounds and reduce the volume ramp down
715 appropriately instead of cancelling the audio device connection
717 Version 0.3.2 (24 June 2005)
719 * packaging changes:
720 - updated autotools build files to compile on OS X
721 (conditional compilation of CoreMIDI and MidiShare drivers)
722 - hand-crafted assembly optimization code can be disabled with
723 './configure --disable-asm' (definitely not recommended)
724 - fixed 'make dist' rule to include all necessary files
725 - require automake (>= 1.5) for 'make -f Makefile.cvs'
726 (due to 'dist-bzip2' automake option)
728 * Gigasampler format engine:
729 - support for the gig parameters for "release velocity response" curves
730 - fine tuning of the EGADSR envelope
731 - volume of release triggered samples now depends on note-on velocity,
732 note length and gig parameter "release trigger decay" instead of
733 note-off velocity.
734 - revised voice stealing
735 (fixes crash and endless loop caused by voice stealing)
736 - don't reset scale tuning on instrument or audio output device change
737 - handle key group conflicts right at the beginning of each fragment
738 instead of when the respective voice is actually launched
739 (fixes undefined behavior if stolen voices belonged to a key group -
740 this case was followed by a "killed voice survived" error message)
741 - fixed minor issue with null/silence samples
742 (those stole voices even though they don't need a voice at all which
743 resulted in "voice stealing didn't work out" messages)
744 - don't reset volume, pan, pitch and MIDI controller values on
745 instrument or audio output device change
747 * LSCP server:
748 - fixed some crashes (patch by Grigor Iliev, fixes #19)
749 - fixed LSCP event "CHANNEL_INFO" notification
750 (e.g. did not notify on volume changes or MIDI program change events)
752 * linuxsampler application:
753 - added command line parameters --lscp-addr and --lscp-port to override
754 default IP address and port of LSCP server
756 Version 0.3.1 (24 May 2005)
758 * initial release

  ViewVC Help
Powered by ViewVC