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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1502 - (show annotations) (download)
Wed Nov 21 07:29:52 2007 UTC (16 years, 4 months ago) by senoner
File size: 18667 byte(s)
* win32 port, work in progress: added autoconf checks for Windows
 - win32 additions in linuxsampler.cpp

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

  ViewVC Help
Powered by ViewVC