ViewVC logotype

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 1604 - (show annotations) (download)
Mon Dec 31 15:33:54 2007 UTC (12 years, 9 months ago) by schoenebeck
File size: 19933 byte(s)
* bugfix: JACK audio driver did not offer a device parameter
  "SAMPLERATE" as opposed to the LSCP specs which requires
  all audio drivers to provide it

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

  ViewVC Help
Powered by ViewVC