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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1653 - (show annotations) (download)
Wed Jan 30 01:51:46 2008 UTC (16 years, 2 months ago) by schoenebeck
File size: 20976 byte(s)
* added support for notifying instrument editors on note-on / note-off
  events (e.g. to highlight the pressed keys on the virtual keyboard
  of gigedit)
* fixed return value of recently added Thread::TestCancel() method
* be verbose on DLL load errors (on Linux)

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

  ViewVC Help
Powered by ViewVC