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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1681 - (show annotations) (download)
Wed Feb 13 15:10:13 2008 UTC (16 years, 1 month ago) by schoenebeck
File size: 21238 byte(s)
- fixed di-harmonic triangle LFO implementation (this LFO implementation
  is only used by very few systems, most pick the int math
  implementation, which is usually faster)

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 - fixed di-harmonic triangle LFO implementation (this LFO implementation
24 is only used by very few systems, most pick the int math
25 implementation, which is usually faster)
26
27 * audio driver:
28 - removed the nonsense audio channel constraint (which was hard coded to
29 max. 100 audio channels) for most audio drivers
30 - JACK audio driver did not offer a device parameter "SAMPLERATE" as
31 opposed to the LSCP specs
32 - bugfix: the SAMPLERATE parameter of some drivers (e.g. JACK)
33 reflected the wrong value
34 - fixed a memory management bug in ASIO driver
35
36 * MIDI driver:
37 - added JACK MIDI driver
38
39 * instruments database:
40 - avoid time consuming samples scanning when adding instruments
41 to the instruments database
42
43 * Gigasampler format engine:
44 - fixed a memory leak that could happen when a channel was deleted
45 while notes were playing
46 - made it possible to load an instrument even if the audio thread
47 isn't running
48
49 Version 0.5.1 (6 December 2007)
50
51 * packaging changes:
52 - added autoconf checks for pthread library
53 - added autoconf check for pthread bug found on certain NPTL-enabled
54 glibc versions (see Gentoo bug report #194076)
55 - added autoconf checks for MS Windows
56
57 * general changes:
58 - bugfix: the thread used by an editor plugin didn't die when the
59 editor closed
60 - bugfix: Ringbuffer.h: fill_write_space_with_null() did not zero
61 out all the space. operator--() did not apply size_mask after
62 decrementing the read_ptr. DEFAULT_WRAP_ELEMENTS set to 0 in
63 order to avoid problems with the _NonVolatileReader functions.
64 - bugfix: Stream.h: added a missing divide by BytesPerSample in
65 GetWriteSpace(). Since this function is currently only used in
66 the stream's qsort() compare function, it didn't trigger any
67 bugs.
68 - Resampler.h, Synthesizer.h: cubic interpolation now works in
69 24bit mode too. Faster method to read 24bit words on little
70 endian machines (x86): replaced 3 byte reads + shifts with a 1
71 unaligned 32bit read and shift
72 - experimental support for MS Windows (MIDI input via MME, AUDIO
73 output via ASIO)
74 - made handling of SIGINT signal (Ctrl-C) a bit more robust
75 - support for monitoring the total number of active disk streams
76 (new LSCP commands: GET TOTAL_STREAM_COUNT,
77 SUBSCRIBE TOTAL_STREAM_COUNT, UNSUBSCRIBE TOTAL_STREAM_COUNT)
78
79 * AUDIO driver:
80 - added Windows ASIO low latency audio driver
81
82 * MIDI driver:
83 - added MME Windows MIDI driver
84
85 * LSCP server:
86 - added support for Windows style path / filenames, however with
87 forward slash path separators instead of backslash
88 (i.e. "C:/foo/bar.gig")
89 - allow naughty liblscp to send non-string device parameters within
90 apostrophes as well
91 - added new LSCP commands: "GET FILE INSTRUMENTS <file>",
92 "LIST FILE INSTRUMENTS <file>" and
93 "GET FILE INSTRUMENT INFO <file> <index>" for retrieving informations
94 about an arbitrary instrument file on the system where the sampler is
95 running on
96
97 Version 0.5.0 (15 October 2007)
98
99 * packaging changes:
100 - config.h is not going to be installed along with liblinuxsampler's
101 API header files anymore
102 - only the API relevant header (and source) files will be exported to
103 the Doxygen API documentation (explicitly listed in Doxyfile.in)
104 - added completely new XCode project files for Mac OSX which is now
105 capable to execute our autoconf environment, thus no need anymore
106 to maintain the compile time configuration file (osx/version.h) for
107 OSX manually (patch by Toshi Nagata)
108 - fixed buggy boolean --enable-foo / --disable-foo configure script
109 parameters
110 - global.h now only covers global definitions that are needed for the
111 C++ API header files, all implementation internal global definitions
112 are now in global_private.h
113 - atomic.h is not exposed to the C++ API anymore
114 - no need to include config.h anymore for using LS's API header files
115 - fixed warnings in API doc generation
116
117 * general changes:
118 - replaced the old, confusing MIDI program change mechanism by a
119 flexible MIDI instrument mapper which allows to map arbitrary
120 (MIDI bank, MIDI program) pairs with arbitrary
121 (engine type, instrument file, file index) triplets which will be
122 loaded on the respective channel when such MIDI program change
123 messages arrive, beside that, each entry allows to define a life-time
124 strategy for the instrument, so the user can define whether the
125 instrument should i.e. be permanently loaded (to be able to switch
126 quickly among sounds) or i.e. loaded just on demand when the
127 respective program change arrives (to safe RAM space), as well as a
128 global volume factor for each entry, so the user can adjust the volume
129 dynamic of his mapped instrument collection without having to modify
130 the instrument files, also one can manage arbitrary amount of such
131 MIDI instrument maps and assign each sampler channel individually
132 a certain map, so that i.e. separation of normal instruments and
133 drumkits is possible
134 - new notification events for tracking changes to audio/MIDI devices,
135 MIDI instrument maps, MIDI instruments, FX sends, global volume.
136 - sampler was limited to load max. 200 instruments in the background
137 due to a constant size RingBuffer FIFO which is now replaced by a
138 dynamic (unlimited) size std::list FIFO
139 - added FX sends, these allow to route audio signals to arbitrary audio
140 output channels for being processed by external effect processors
141 (i.e. jack-rack), the send levels are controllable via arbitrary MIDI
142 controllers
143 - global (sampler wide) volume can now be controlled at runtime
144 - Implemented new, improved notification system
145 - fixed compilation errors regarding OSX
146 (patch by Toshi Nagata)
147 - implemented instruments database
148 - added support for escape sequences to the instruments database
149 - added highly experimental support for on-the-fly instrument editing
150 within the sampler's process (by using instrument editor plugins),
151 you'll notice the new "Registered instrument editors:" message on
152 startup, added a new LSCP command:
153 "EDIT CHANNEL INSTRUMENT <sampler-channel>"
154 to spawn a matching instrument editor for the instrument on the
155 given sampler channel, the plugin path can be overridden at compile
156 time with ./configure --enable-plugin-dir=/some/dir
157 - added experimental code for synchronizing instrument editors hosted
158 in the sampler's process to safely edit instruments while playing
159 without a crash (hopefully) by either suspending single regions
160 wherever possible or if unavoidable whole engine(s)
161 - fixed several issues in fundamental "Thread" class: set scheduling
162 policy and priority on thread level, set a minimum stack size for
163 thread (TODO: a reasonable value yet to be tested), bugfix: non-RT
164 threads simply inherited properties of starting thread instead of
165 setting their own policy and priority
166 - minor fix in our automatic stack trace mechanism on crashes, the
167 main process did not wait for the stack trace process to finish
168 its output
169 - fixed some minor memory leaks
170 - reenabled assembly features support, at the moment only for
171 enabling a fast denormal FPU mode (x86 platforms supporting SSE2)
172 - minor assembly fix in x86 features detection (don't use the PIC
173 register, to avoid relocations in the text segment at runtime)
174 - POSIX callback functions of Thread.h are hidden
175
176 * MIDI driver:
177 - dispatch bank select (MSB and LSB) messages
178
179 * audio driver:
180 - the ALSA audio output driver parameters now reflect the correct
181 parameter value ranges for the respective selected sound card
182 (patch by Till Wimmer, a bit fixed and extended)
183
184 * Gigasampler format engine:
185 - if a filter is used and EG2 finishes before EG1, let the voice
186 die when EG2 has finished (this fixes a problem with clicks and
187 voice starvation for some gigs)
188 - playback is no longer disabled during instrument loading
189 - all notes playing on a channel that changes its instrument keep
190 playing with the old instrument until they get a note off
191 command
192 - EG fix: a release value of zero could cause noises or crash
193 - handle MIDI coarse tuning messages (MIDI RPN #0 MSB #2 LSB)
194 - EG fine tuning: when attack is zero the EG starts at a level
195 above max sustain level, which means that there is a decay phase
196 even if sustain is 100%
197 - more EG fixes: the level could sometimes go below zero and cause
198 noises or crashes
199 - minor fix of EGDecay (patch by Toshi Nagata)
200 - fixed compiler error when --enable-override-filter-type was
201 supplied to the configure script (fixes #46)
202 - disk thread: queue sizes are now proportional to CONFIG_MAX_STREAMS
203 instead of a fix value
204 - behavior fix: on MIDI CC# 65 (portamento on / off), 126 (mono mode),
205 127 (solo mode) only kill voices if the respective mode really
206 changed
207
208 * LSCP server:
209 - fixed compile time error for old Bison versions
210 (i.e. v1.28 found on MacOS 10.4, patch by Toshi Nagata)
211 - parser now supports extended ASCII character set
212 (up to ASCII code 255, i.e. includes now umlauts and accents)
213 - filename arguments in LSCP commands now allow to use escape
214 sequences, that is directly literal as one of: \', \", \\, \n, \r,
215 \f, \t, \v, or as octal ASCII code value like \132, or as hex ASCII
216 code value like \xf2) (fixes bug #24)
217 - the following LSCP commands now also support escape sequences for at
218 least one of their text-based parameters (i.e. name, description):
219 "ADD MIDI_INSTRUMENT_MAP", "MAP MIDI_INSTRUMENT",
220 "SET MIDI_INSTRUMENT_MAP NAME", "SET FX_SEND NAME", "CREATE FX_SEND",
221 "SET DB_INSTRUMENT_DIRECTORY NAME",
222 "SET DB_INSTRUMENT_DIRECTORY DESCRIPTION", "SET DB_INSTRUMENT NAME",
223 "SET DB_INSTRUMENT DESCRIPTION", "FIND DB_INSTRUMENTS",
224 "FIND DB_INSTRUMENT_DIRECTORIES"
225 - returns verbose syntax errors (line and column where syntax error
226 occured, the unexpected character and the actually expected, possible
227 character(s), the latter only if less than 5 possibilities)
228 - made sure that LSCP syntax is not affected by gigedit locale
229 settings
230 - bugfix regarding strings parameter lists: all comma separated lists
231 of strings were treated as being one string containing commas
232 (fixes #57)
233
234 * Bug fixes:
235 - fixed crash occurring on certain LSCP scripts (Bug 39)
236 - another thread safety fix for lscp "load engine" and "set
237 channel audio output device"
238 - fixed a crash which occurs when reassigning the same engine
239 on a sampler channel with connected MIDI device
240 - fixed a crash which occurs when changing the number of ports of a MIDI
241 device connected to a sampler channel to number less then or equal
242 to the index of the port to which the sampler channel is connected.
243 - The previous bindings were not been disconnected when altering
244 the ALSA_SEQ_BINDINGS parameter. Introduced a NONE keyword for
245 unsubscribing from all bindings (e.g. ALSA_SEQ_BINDINGS=NONE).
246 - The active stream/voice count statistic was incorrect.
247 - notification events were not been sent for some sampler
248 channel changes
249 - added default min and max values to restrict the number of allowed
250 audio output channels and MIDI input ports
251 - the connection to the PCM interface is now closed when destroying
252 an audio output device
253 - files with slash in their path or filename could not be loaded
254
255 * test cases:
256 - updated and fixed (haven't been touched in a while)
257
258 Version 0.4.0 (24 November 2006)
259
260 * packaging changes:
261 - changed deprecated copyright attribute to license; added ldconfig
262 to post-(un)install steps; added devel package for liblinuxsampler;
263 to linuxsampler.spec (RPM)
264 - install necessary development header files for allowing 3rd party
265 applications to link against liblinuxsampler
266 - liblinuxsampler's API documentation can be generated with 'make docs'
267 (Doxygen required)
268 - added benchmark to automatically detect the best triangle LFO
269 implementation (currently either an integer math solution or a
270 di-harmonic approximation), automatic detection can be overriden
271 with --enable-signed-triang-algo=x and --enable-unsigned-triang-algo=x
272 configure script argument though (mandatory for cross-compilation)
273 - do not automatically pick optimized gcc flags if the user already
274 provided some on his own (as CXXFLAGS)
275 - added compile time option to disable processing of All-Notes-Off MIDI
276 messages
277 - added compile time options to allow disabling the various audio and
278 MIDI drivers
279 - fixed automatic GCC CPU switch detection on PPC
280 (patch by Ebrahim Mayat)
281
282 * Gigasampler format engine:
283 - extensive synthesis optimization
284 (reimplementation of EGs and LFO(s), removed synthesis parameter
285 prerendering and the synthesis parameter matrix in general, splitting
286 each audio fragment into subfragments now where each subfragment uses
287 constant pitch and filter coefficients. The volume coefficient is
288 linearly interpolated inside a subfragment, unless
289 --disable-interpolate-volume is set.)
290 - fine tuning of the EG modulation parameters
291 - improved filter cutoff calculation by adding support for the
292 following gig parameters: Cutoff freq (used when no cutoff
293 controller is defined), Control invert, Minimum cutoff, Velocity
294 curve and Velocity range. The keyboard tracking now scales
295 cutoff frequency instead of resonance.
296 - added support for gig parameter Resonance.
297 - fixed bug in sysex handling (patch by Juan Linietsky)
298 - added global volume attenuation of -9 dB (0.35f) to prevent clipping
299 which can be overridden with --enable-global-attenuation
300 - EG fixes: made the length of "attack hold" stage more
301 accurate. Release stage can now start before attack stage
302 ends. Cancel release didn't work when sustain was zero. Attack
303 time now has a minimal value to prevent clicks.
304 - fixed pitch changes larger than one octave
305 - fixed EG3 (pitch envelope) synthesis which was neutral all the time
306 - implemented portamento mode and solo mode (a.k.a 'mono mode'):
307 all modes can be altered via standard GM messages, that is CC5 for
308 altering portamento time, CC65 for enabling / disabling portamento
309 mode, CC126 for enabling solo mode and CC127 for disabling solo mode
310 - fine tuning of the curves for volume (CC7), pan (CC10 and gig
311 parameter) and crossfade
312 - added support for the "attenuation controller threshold" gig
313 parameter
314 - added smoothing of volume changes caused by control change
315 messages
316 - sample loop parameters are now taken from the DimensionRegion
317 instead of the wave chunk
318 - fixed keyswitching for v3 gigs with a number of keyswitch splits
319 not equal to a power of two
320 - reimplementation of the filter algorithm. The new filter is more
321 accurate and supports all gig filter types, including bandreject
322 and lowpass turbo.
323 - real support for 24 bit samples - samples are not truncated to
324 16 bits anymore
325 - support for aftertouch (channel pressure, not polyphonic
326 aftertouch)
327 - LFO1 behaviour fixed (dampening from max volume instead of
328 amplifying from 0)
329
330 * LSCP server:
331 - fixed application exit on broken pipe error (fixes bug #20)
332 - fixed the notification messages delay due to lack of
333 network activity (fixes bug #26)
334 - fixed parser bug which occured on space(s) within device parameters
335 (e.g. "SET AUDIO_OUTPUT_CHANNEL_PARAMETER 0 0 SOMEPARAM='foo bar'")
336
337 * audio driver:
338 - added aRts audio output driver (by no means RT safe)
339
340 * MIDI driver:
341 - fixed legacy sysex code which caused dispatching of MIDI SysEx
342 messages several times instead of once
343 - API extension for MIDI drivers which already supply exact time stamps
344 for events (i.e. for offline rendering based MIDI drivers)
345 - added checks for bad MIDI data
346
347 * linuxsampler application:
348 - show available sampler engine types on startup
349
350 * general changes:
351 - support for muting sampler channels and solo mode of the same, two new
352 LSCP commands ("SET CHANNEL MUTE" and "SET CHANNEL SOLO") and two new
353 fields ("MUTE" and "SOLO") for command "GET CHANNEL INFO" were
354 introduced for this, the behavior is the same like on a mixer console
355 (patch by Grigor Iliev, a bit adjusted). Also added configure option
356 --enable-process-muted-channels which can be used to enable the
357 processing of muted channels.
358 - support for sostenuto pedal
359 - support for monitoring the total number of active voices
360 (new LSCP commands: GET TOTAL_VOICE_COUNT, GET TOTAL_VOICE_COUNT_MAX,
361 SUBSCRIBE TOTAL_VOICE_COUNT, UNSUBSCRIBE TOTAL_VOICE_COUNT)
362 - fixed some memory management errors
363 - fixed some concurrency problems that could lead to crashes when
364 LSCP commands were executed
365 - fixed crash when instrument loading failed previously on the same
366 sampler channel (fixes bug #36)
367
368 Version 0.3.3 (15 July 2005)
369
370 * packaging changes:
371 - fixed compilation with gcc 4.0
372
373 * Gigasampler format engine:
374 - LFO preprocessing was performed even though the respective LFO was
375 disabled by the instrument patch (minor efficiency fix)
376 - if period time of chosen audio device is too small (< MIN_RELEASE_TIME)
377 for volume ramp downs in the same fragment (needed for current voice
378 stealing implementation) then simply inform the user with a warning
379 about possible click sounds and reduce the volume ramp down
380 appropriately instead of cancelling the audio device connection
381
382 Version 0.3.2 (24 June 2005)
383
384 * packaging changes:
385 - updated autotools build files to compile on OS X
386 (conditional compilation of CoreMIDI and MidiShare drivers)
387 - hand-crafted assembly optimization code can be disabled with
388 './configure --disable-asm' (definitely not recommended)
389 - fixed 'make dist' rule to include all necessary files
390 - require automake (>= 1.5) for 'make -f Makefile.cvs'
391 (due to 'dist-bzip2' automake option)
392
393 * Gigasampler format engine:
394 - support for the gig parameters for "release velocity response" curves
395 - fine tuning of the EGADSR envelope
396 - volume of release triggered samples now depends on note-on velocity,
397 note length and gig parameter "release trigger decay" instead of
398 note-off velocity.
399 - revised voice stealing
400 (fixes crash and endless loop caused by voice stealing)
401 - don't reset scale tuning on instrument or audio output device change
402 - handle key group conflicts right at the beginning of each fragment
403 instead of when the respective voice is actually launched
404 (fixes undefined behavior if stolen voices belonged to a key group -
405 this case was followed by a "killed voice survived" error message)
406 - fixed minor issue with null/silence samples
407 (those stole voices even though they don't need a voice at all which
408 resulted in "voice stealing didn't work out" messages)
409 - don't reset volume, pan, pitch and MIDI controller values on
410 instrument or audio output device change
411
412 * LSCP server:
413 - fixed some crashes (patch by Grigor Iliev, fixes #19)
414 - fixed LSCP event "CHANNEL_INFO" notification
415 (e.g. did not notify on volume changes or MIDI program change events)
416
417 * linuxsampler application:
418 - added command line parameters --lscp-addr and --lscp-port to override
419 default IP address and port of LSCP server
420
421 Version 0.3.1 (24 May 2005)
422
423 * initial release

  ViewVC Help
Powered by ViewVC