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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1454 - (show annotations) (download)
Fri Oct 19 17:52:15 2007 UTC (16 years, 5 months ago) by schoenebeck
File size: 17300 byte(s)
* added autoconf checks for pthread library
* added autoconf check for pthread bug found on certain NPTL-enabled
  glibc versions (see Gentoo bug report #194076)

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

  ViewVC Help
Powered by ViewVC