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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1321 - (show annotations) (download)
Tue Sep 4 01:12:49 2007 UTC (16 years, 6 months ago) by schoenebeck
File size: 15522 byte(s)
* added highly experimental code for synchronizing instrument editors
  hosted in the sampler's process to safely edit instruments while playing
  without a crash (hopefully) by either suspending single regions wherever
  possible or - if unavoidable - whole engine(s)
* disk thread: queue sizes are now proportional to CONFIG_MAX_STREAMS
  instead of fix values
* removed legacy Makefiles in meanwhile deleted src/lib directory and its
  subdirectories
* bumped version to 0.4.0.7cvs

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

  ViewVC Help
Powered by ViewVC