ViewVC logotype

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 1602 - (show annotations) (download)
Sat Dec 29 22:02:29 2007 UTC (16 years, 2 months ago) by schoenebeck
File size: 19699 byte(s)
* fixed endless loop which occured when loading an instrument editor
  plugin DLL on Windows systems failed (note: completely untested yet :) )

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

  ViewVC Help
Powered by ViewVC