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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1603 - (show annotations) (download)
Mon Dec 31 10:17:51 2007 UTC (16 years, 3 months ago) by iliev
File size: 19827 byte(s)
* avoid time consuming samples scanning when
  adding instruments to the instruments database

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

  ViewVC Help
Powered by ViewVC