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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1541 - (show annotations) (download)
Tue Dec 4 18:09:26 2007 UTC (12 years ago) by iliev
File size: 19194 byte(s)
- Added support for monitoring the total number of active disk streams
  (new LSCP commands: GET TOTAL_STREAM_COUNT,
  SUBSCRIBE TOTAL_STREAM_COUNT, UNSUBSCRIBE TOTAL_STREAM_COUNT)

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

  ViewVC Help
Powered by ViewVC