ViewVC logotype

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 1835 - (show annotations) (download)
Mon Feb 16 17:56:50 2009 UTC (11 years, 7 months ago) by iliev
File size: 26195 byte(s)
* Adapted drivers/Plugin class to be used as a base class for the
  AudioUnit plugin and fixed orphaned pointers when deleting PluginGlobal
* Fixed possible iterator invalidations when resetting the sampler
* Fixed memory leaks when issuing the following LSCP commands:
* Fixed possible compilation error when sqlite is not present

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

  ViewVC Help
Powered by ViewVC