ViewVC logotype

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 1789 - (show annotations) (download)
Sat Nov 1 19:01:27 2008 UTC (11 years, 6 months ago) by iliev
File size: 24404 byte(s)
* notification events for stream/voice count statistics are now sent
  only when there are actual changes
* use atomic_t for stream/voice count statistics

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

  ViewVC Help
Powered by ViewVC