ViewVC logotype

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 1792 - (show annotations) (download)
Sun Nov 16 19:19:26 2008 UTC (12 years, 2 months ago) by persson
File size: 24578 byte(s)
* fixes for building on OS X

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

  ViewVC Help
Powered by ViewVC