/[svn]/linuxsampler/trunk/ChangeLog
ViewVC logotype

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1659 - (show annotations) (download)
Sun Feb 3 00:13:27 2008 UTC (16 years, 1 month ago) by schoenebeck
File size: 21051 byte(s)
* added support for triggering notes by instrument editors
  (still some cleanup / refactoring ahead, but it should work)
* bumped version to 0.5.1.1cvs

1 Version CVS HEAD (?)
2
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
9 * general changes:
10 - bugfix: on some POSIX systems instrument editor plugins refused to
11 load as we used a non-portable Linux specific struct field
12 (fixes bug #70, patch by Ronald Baljeu)
13 - fixed endless loop which occured when loading an instrument editor
14 plugin DLL on Windows systems failed
15 - fixed memory leaks that occurred when liblinuxsampler was unloaded
16 - optimized the SynchronizedConfig class so it doesn't wait
17 unnecessarily long after an update
18 - added support for notifying instrument editors on note-on / note-off
19 events (e.g. to highlight the pressed keys on the virtual keyboard
20 of gigedit)
21 - added support for triggering notes by instrument editors (see above)
22 - be verbose on DLL load errors (on Linux)
23
24 * audio driver:
25 - removed the nonsense audio channel constraint (which was hard coded to
26 max. 100 audio channels) for most audio drivers
27 - JACK audio driver did not offer a device parameter "SAMPLERATE" as
28 opposed to the LSCP specs
29 - bugfix: the SAMPLERATE parameter of some drivers (e.g. JACK)
30 reflected the wrong value
31 - fixed a memory management bug in ASIO driver
32
33 * MIDI driver:
34 - added JACK MIDI driver
35
36 * instruments database:
37 - avoid time consuming samples scanning when adding instruments
38 to the instruments database
39
40 * Gigasampler format engine:
41 - fixed a memory leak that could happen when a channel was deleted
42 while notes were playing
43 - made it possible to load an instrument even if the audio thread
44 isn't running
45
46 Version 0.5.1 (6 December 2007)
47
48 * packaging changes:
49 - added autoconf checks for pthread library
50 - added autoconf check for pthread bug found on certain NPTL-enabled
51 glibc versions (see Gentoo bug report #194076)
52 - added autoconf checks for MS Windows
53
54 * general changes:
55 - bugfix: the thread used by an editor plugin didn't die when the
56 editor closed
57 - bugfix: Ringbuffer.h: fill_write_space_with_null() did not zero
58 out all the space. operator--() did not apply size_mask after
59 decrementing the read_ptr. DEFAULT_WRAP_ELEMENTS set to 0 in
60 order to avoid problems with the _NonVolatileReader functions.
61 - bugfix: Stream.h: added a missing divide by BytesPerSample in
62 GetWriteSpace(). Since this function is currently only used in
63 the stream's qsort() compare function, it didn't trigger any
64 bugs.
65 - Resampler.h, Synthesizer.h: cubic interpolation now works in
66 24bit mode too. Faster method to read 24bit words on little
67 endian machines (x86): replaced 3 byte reads + shifts with a 1
68 unaligned 32bit read and shift
69 - experimental support for MS Windows (MIDI input via MME, AUDIO
70 output via ASIO)
71 - made handling of SIGINT signal (Ctrl-C) a bit more robust
72 - support for monitoring the total number of active disk streams
73 (new LSCP commands: GET TOTAL_STREAM_COUNT,
74 SUBSCRIBE TOTAL_STREAM_COUNT, UNSUBSCRIBE TOTAL_STREAM_COUNT)
75
76 * AUDIO driver:
77 - added Windows ASIO low latency audio driver
78
79 * MIDI driver:
80 - added MME Windows MIDI driver
81
82 * LSCP server:
83 - added support for Windows style path / filenames, however with
84 forward slash path separators instead of backslash
85 (i.e. "C:/foo/bar.gig")
86 - allow naughty liblscp to send non-string device parameters within
87 apostrophes as well
88 - added new LSCP commands: "GET FILE INSTRUMENTS <file>",
89 "LIST FILE INSTRUMENTS <file>" and
90 "GET FILE INSTRUMENT INFO <file> <index>" for retrieving informations
91 about an arbitrary instrument file on the system where the sampler is
92 running on
93
94 Version 0.5.0 (15 October 2007)
95
96 * packaging changes:
97 - config.h is not going to be installed along with liblinuxsampler's
98 API header files anymore
99 - only the API relevant header (and source) files will be exported to
100 the Doxygen API documentation (explicitly listed in Doxyfile.in)
101 - added completely new XCode project files for Mac OSX which is now
102 capable to execute our autoconf environment, thus no need anymore
103 to maintain the compile time configuration file (osx/version.h) for
104 OSX manually (patch by Toshi Nagata)
105 - fixed buggy boolean --enable-foo / --disable-foo configure script
106 parameters
107 - global.h now only covers global definitions that are needed for the
108 C++ API header files, all implementation internal global definitions
109 are now in global_private.h
110 - atomic.h is not exposed to the C++ API anymore
111 - no need to include config.h anymore for using LS's API header files
112 - fixed warnings in API doc generation
113
114 * general changes:
115 - replaced the old, confusing MIDI program change mechanism by a
116 flexible MIDI instrument mapper which allows to map arbitrary
117 (MIDI bank, MIDI program) pairs with arbitrary
118 (engine type, instrument file, file index) triplets which will be
119 loaded on the respective channel when such MIDI program change
120 messages arrive, beside that, each entry allows to define a life-time
121 strategy for the instrument, so the user can define whether the
122 instrument should i.e. be permanently loaded (to be able to switch
123 quickly among sounds) or i.e. loaded just on demand when the
124 respective program change arrives (to safe RAM space), as well as a
125 global volume factor for each entry, so the user can adjust the volume
126 dynamic of his mapped instrument collection without having to modify
127 the instrument files, also one can manage arbitrary amount of such
128 MIDI instrument maps and assign each sampler channel individually
129 a certain map, so that i.e. separation of normal instruments and
130 drumkits is possible
131 - new notification events for tracking changes to audio/MIDI devices,
132 MIDI instrument maps, MIDI instruments, FX sends, global volume.
133 - sampler was limited to load max. 200 instruments in the background
134 due to a constant size RingBuffer FIFO which is now replaced by a
135 dynamic (unlimited) size std::list FIFO
136 - added FX sends, these allow to route audio signals to arbitrary audio
137 output channels for being processed by external effect processors
138 (i.e. jack-rack), the send levels are controllable via arbitrary MIDI
139 controllers
140 - global (sampler wide) volume can now be controlled at runtime
141 - Implemented new, improved notification system
142 - fixed compilation errors regarding OSX
143 (patch by Toshi Nagata)
144 - implemented instruments database
145 - added support for escape sequences to the instruments database
146 - added highly experimental support for on-the-fly instrument editing
147 within the sampler's process (by using instrument editor plugins),
148 you'll notice the new "Registered instrument editors:" message on
149 startup, added a new LSCP command:
150 "EDIT CHANNEL INSTRUMENT <sampler-channel>"
151 to spawn a matching instrument editor for the instrument on the
152 given sampler channel, the plugin path can be overridden at compile
153 time with ./configure --enable-plugin-dir=/some/dir
154 - added experimental code for synchronizing instrument editors hosted
155 in the sampler's process to safely edit instruments while playing
156 without a crash (hopefully) by either suspending single regions
157 wherever possible or if unavoidable whole engine(s)
158 - fixed several issues in fundamental "Thread" class: set scheduling
159 policy and priority on thread level, set a minimum stack size for
160 thread (TODO: a reasonable value yet to be tested), bugfix: non-RT
161 threads simply inherited properties of starting thread instead of
162 setting their own policy and priority
163 - minor fix in our automatic stack trace mechanism on crashes, the
164 main process did not wait for the stack trace process to finish
165 its output
166 - fixed some minor memory leaks
167 - reenabled assembly features support, at the moment only for
168 enabling a fast denormal FPU mode (x86 platforms supporting SSE2)
169 - minor assembly fix in x86 features detection (don't use the PIC
170 register, to avoid relocations in the text segment at runtime)
171 - POSIX callback functions of Thread.h are hidden
172
173 * MIDI driver:
174 - dispatch bank select (MSB and LSB) messages
175
176 * audio driver:
177 - the ALSA audio output driver parameters now reflect the correct
178 parameter value ranges for the respective selected sound card
179 (patch by Till Wimmer, a bit fixed and extended)
180
181 * Gigasampler format engine:
182 - if a filter is used and EG2 finishes before EG1, let the voice
183 die when EG2 has finished (this fixes a problem with clicks and
184 voice starvation for some gigs)
185 - playback is no longer disabled during instrument loading
186 - all notes playing on a channel that changes its instrument keep
187 playing with the old instrument until they get a note off
188 command
189 - EG fix: a release value of zero could cause noises or crash
190 - handle MIDI coarse tuning messages (MIDI RPN #0 MSB #2 LSB)
191 - EG fine tuning: when attack is zero the EG starts at a level
192 above max sustain level, which means that there is a decay phase
193 even if sustain is 100%
194 - more EG fixes: the level could sometimes go below zero and cause
195 noises or crashes
196 - minor fix of EGDecay (patch by Toshi Nagata)
197 - fixed compiler error when --enable-override-filter-type was
198 supplied to the configure script (fixes #46)
199 - disk thread: queue sizes are now proportional to CONFIG_MAX_STREAMS
200 instead of a fix value
201 - behavior fix: on MIDI CC# 65 (portamento on / off), 126 (mono mode),
202 127 (solo mode) only kill voices if the respective mode really
203 changed
204
205 * LSCP server:
206 - fixed compile time error for old Bison versions
207 (i.e. v1.28 found on MacOS 10.4, patch by Toshi Nagata)
208 - parser now supports extended ASCII character set
209 (up to ASCII code 255, i.e. includes now umlauts and accents)
210 - filename arguments in LSCP commands now allow to use escape
211 sequences, that is directly literal as one of: \', \", \\, \n, \r,
212 \f, \t, \v, or as octal ASCII code value like \132, or as hex ASCII
213 code value like \xf2) (fixes bug #24)
214 - the following LSCP commands now also support escape sequences for at
215 least one of their text-based parameters (i.e. name, description):
216 "ADD MIDI_INSTRUMENT_MAP", "MAP MIDI_INSTRUMENT",
217 "SET MIDI_INSTRUMENT_MAP NAME", "SET FX_SEND NAME", "CREATE FX_SEND",
218 "SET DB_INSTRUMENT_DIRECTORY NAME",
219 "SET DB_INSTRUMENT_DIRECTORY DESCRIPTION", "SET DB_INSTRUMENT NAME",
220 "SET DB_INSTRUMENT DESCRIPTION", "FIND DB_INSTRUMENTS",
221 "FIND DB_INSTRUMENT_DIRECTORIES"
222 - returns verbose syntax errors (line and column where syntax error
223 occured, the unexpected character and the actually expected, possible
224 character(s), the latter only if less than 5 possibilities)
225 - made sure that LSCP syntax is not affected by gigedit locale
226 settings
227 - bugfix regarding strings parameter lists: all comma separated lists
228 of strings were treated as being one string containing commas
229 (fixes #57)
230
231 * Bug fixes:
232 - fixed crash occurring on certain LSCP scripts (Bug 39)
233 - another thread safety fix for lscp "load engine" and "set
234 channel audio output device"
235 - fixed a crash which occurs when reassigning the same engine
236 on a sampler channel with connected MIDI device
237 - fixed a crash which occurs when changing the number of ports of a MIDI
238 device connected to a sampler channel to number less then or equal
239 to the index of the port to which the sampler channel is connected.
240 - The previous bindings were not been disconnected when altering
241 the ALSA_SEQ_BINDINGS parameter. Introduced a NONE keyword for
242 unsubscribing from all bindings (e.g. ALSA_SEQ_BINDINGS=NONE).
243 - The active stream/voice count statistic was incorrect.
244 - notification events were not been sent for some sampler
245 channel changes
246 - added default min and max values to restrict the number of allowed
247 audio output channels and MIDI input ports
248 - the connection to the PCM interface is now closed when destroying
249 an audio output device
250 - files with slash in their path or filename could not be loaded
251
252 * test cases:
253 - updated and fixed (haven't been touched in a while)
254
255 Version 0.4.0 (24 November 2006)
256
257 * packaging changes:
258 - changed deprecated copyright attribute to license; added ldconfig
259 to post-(un)install steps; added devel package for liblinuxsampler;
260 to linuxsampler.spec (RPM)
261 - install necessary development header files for allowing 3rd party
262 applications to link against liblinuxsampler
263 - liblinuxsampler's API documentation can be generated with 'make docs'
264 (Doxygen required)
265 - added benchmark to automatically detect the best triangle LFO
266 implementation (currently either an integer math solution or a
267 di-harmonic approximation), automatic detection can be overriden
268 with --enable-signed-triang-algo=x and --enable-unsigned-triang-algo=x
269 configure script argument though (mandatory for cross-compilation)
270 - do not automatically pick optimized gcc flags if the user already
271 provided some on his own (as CXXFLAGS)
272 - added compile time option to disable processing of All-Notes-Off MIDI
273 messages
274 - added compile time options to allow disabling the various audio and
275 MIDI drivers
276 - fixed automatic GCC CPU switch detection on PPC
277 (patch by Ebrahim Mayat)
278
279 * Gigasampler format engine:
280 - extensive synthesis optimization
281 (reimplementation of EGs and LFO(s), removed synthesis parameter
282 prerendering and the synthesis parameter matrix in general, splitting
283 each audio fragment into subfragments now where each subfragment uses
284 constant pitch and filter coefficients. The volume coefficient is
285 linearly interpolated inside a subfragment, unless
286 --disable-interpolate-volume is set.)
287 - fine tuning of the EG modulation parameters
288 - improved filter cutoff calculation by adding support for the
289 following gig parameters: Cutoff freq (used when no cutoff
290 controller is defined), Control invert, Minimum cutoff, Velocity
291 curve and Velocity range. The keyboard tracking now scales
292 cutoff frequency instead of resonance.
293 - added support for gig parameter Resonance.
294 - fixed bug in sysex handling (patch by Juan Linietsky)
295 - added global volume attenuation of -9 dB (0.35f) to prevent clipping
296 which can be overridden with --enable-global-attenuation
297 - EG fixes: made the length of "attack hold" stage more
298 accurate. Release stage can now start before attack stage
299 ends. Cancel release didn't work when sustain was zero. Attack
300 time now has a minimal value to prevent clicks.
301 - fixed pitch changes larger than one octave
302 - fixed EG3 (pitch envelope) synthesis which was neutral all the time
303 - implemented portamento mode and solo mode (a.k.a 'mono mode'):
304 all modes can be altered via standard GM messages, that is CC5 for
305 altering portamento time, CC65 for enabling / disabling portamento
306 mode, CC126 for enabling solo mode and CC127 for disabling solo mode
307 - fine tuning of the curves for volume (CC7), pan (CC10 and gig
308 parameter) and crossfade
309 - added support for the "attenuation controller threshold" gig
310 parameter
311 - added smoothing of volume changes caused by control change
312 messages
313 - sample loop parameters are now taken from the DimensionRegion
314 instead of the wave chunk
315 - fixed keyswitching for v3 gigs with a number of keyswitch splits
316 not equal to a power of two
317 - reimplementation of the filter algorithm. The new filter is more
318 accurate and supports all gig filter types, including bandreject
319 and lowpass turbo.
320 - real support for 24 bit samples - samples are not truncated to
321 16 bits anymore
322 - support for aftertouch (channel pressure, not polyphonic
323 aftertouch)
324 - LFO1 behaviour fixed (dampening from max volume instead of
325 amplifying from 0)
326
327 * LSCP server:
328 - fixed application exit on broken pipe error (fixes bug #20)
329 - fixed the notification messages delay due to lack of
330 network activity (fixes bug #26)
331 - fixed parser bug which occured on space(s) within device parameters
332 (e.g. "SET AUDIO_OUTPUT_CHANNEL_PARAMETER 0 0 SOMEPARAM='foo bar'")
333
334 * audio driver:
335 - added aRts audio output driver (by no means RT safe)
336
337 * MIDI driver:
338 - fixed legacy sysex code which caused dispatching of MIDI SysEx
339 messages several times instead of once
340 - API extension for MIDI drivers which already supply exact time stamps
341 for events (i.e. for offline rendering based MIDI drivers)
342 - added checks for bad MIDI data
343
344 * linuxsampler application:
345 - show available sampler engine types on startup
346
347 * general changes:
348 - support for muting sampler channels and solo mode of the same, two new
349 LSCP commands ("SET CHANNEL MUTE" and "SET CHANNEL SOLO") and two new
350 fields ("MUTE" and "SOLO") for command "GET CHANNEL INFO" were
351 introduced for this, the behavior is the same like on a mixer console
352 (patch by Grigor Iliev, a bit adjusted). Also added configure option
353 --enable-process-muted-channels which can be used to enable the
354 processing of muted channels.
355 - support for sostenuto pedal
356 - support for monitoring the total number of active voices
357 (new LSCP commands: GET TOTAL_VOICE_COUNT, GET TOTAL_VOICE_COUNT_MAX,
358 SUBSCRIBE TOTAL_VOICE_COUNT, UNSUBSCRIBE TOTAL_VOICE_COUNT)
359 - fixed some memory management errors
360 - fixed some concurrency problems that could lead to crashes when
361 LSCP commands were executed
362 - fixed crash when instrument loading failed previously on the same
363 sampler channel (fixes bug #36)
364
365 Version 0.3.3 (15 July 2005)
366
367 * packaging changes:
368 - fixed compilation with gcc 4.0
369
370 * Gigasampler format engine:
371 - LFO preprocessing was performed even though the respective LFO was
372 disabled by the instrument patch (minor efficiency fix)
373 - if period time of chosen audio device is too small (< MIN_RELEASE_TIME)
374 for volume ramp downs in the same fragment (needed for current voice
375 stealing implementation) then simply inform the user with a warning
376 about possible click sounds and reduce the volume ramp down
377 appropriately instead of cancelling the audio device connection
378
379 Version 0.3.2 (24 June 2005)
380
381 * packaging changes:
382 - updated autotools build files to compile on OS X
383 (conditional compilation of CoreMIDI and MidiShare drivers)
384 - hand-crafted assembly optimization code can be disabled with
385 './configure --disable-asm' (definitely not recommended)
386 - fixed 'make dist' rule to include all necessary files
387 - require automake (>= 1.5) for 'make -f Makefile.cvs'
388 (due to 'dist-bzip2' automake option)
389
390 * Gigasampler format engine:
391 - support for the gig parameters for "release velocity response" curves
392 - fine tuning of the EGADSR envelope
393 - volume of release triggered samples now depends on note-on velocity,
394 note length and gig parameter "release trigger decay" instead of
395 note-off velocity.
396 - revised voice stealing
397 (fixes crash and endless loop caused by voice stealing)
398 - don't reset scale tuning on instrument or audio output device change
399 - handle key group conflicts right at the beginning of each fragment
400 instead of when the respective voice is actually launched
401 (fixes undefined behavior if stolen voices belonged to a key group -
402 this case was followed by a "killed voice survived" error message)
403 - fixed minor issue with null/silence samples
404 (those stole voices even though they don't need a voice at all which
405 resulted in "voice stealing didn't work out" messages)
406 - don't reset volume, pan, pitch and MIDI controller values on
407 instrument or audio output device change
408
409 * LSCP server:
410 - fixed some crashes (patch by Grigor Iliev, fixes #19)
411 - fixed LSCP event "CHANNEL_INFO" notification
412 (e.g. did not notify on volume changes or MIDI program change events)
413
414 * linuxsampler application:
415 - added command line parameters --lscp-addr and --lscp-port to override
416 default IP address and port of LSCP server
417
418 Version 0.3.1 (24 May 2005)
419
420 * initial release

  ViewVC Help
Powered by ViewVC