1 |
Version CVS HEAD (?) |
2 |
|
3 |
* plugin changes: |
4 |
- LV2 Persist + Files support (patch by David Robillard) |
5 |
|
6 |
* packaging changes: |
7 |
- fixed building with newer MinGW-w64 |
8 |
- Mac OS X: support the new dir for Core Audio SDK |
9 |
- Mac OS X: fixed building outside source directory |
10 |
- made sure all source files for hostplugins are included when |
11 |
doing "make dist" |
12 |
- removed empty directories left from the cvs to svn migration |
13 |
- fixed compilation with gcc 4.6.1 |
14 |
- another "make dist" fix, for LV2 plugin |
15 |
- made --enable-pthread-testcancel default on Mac OS X |
16 |
|
17 |
* general changes: |
18 |
- Refactoring: moved the independent code from |
19 |
the Gigasampler format engine to base classes |
20 |
- command line option '--profile' is currently disabled, since the |
21 |
respective profiling code is currently broken |
22 |
- Introduced support for optional environment variable |
23 |
"LINUXSAMPLER_PLUGIN_DIR" which allows to override the directory |
24 |
where the sampler shall look for instrument editor plugins |
25 |
(patch by Luis Garrido, slightly modified). |
26 |
- implemented Roland GS NRPN 1ArrH which allows to set volume per note |
27 |
- implemented Roland GS NRPN 1CrrH which allows to set pan per note |
28 |
- implemented Roland GS NRPN 1DrrH which allows to set reverb send per |
29 |
note (in this implementation of the sampler its simply hard routed to |
30 |
the 1st effect send of the sampler channel, no matter what the actual |
31 |
effect type is) |
32 |
- implemented Roland GS NRPN 1ErrH which allows to set chorus send per |
33 |
note (in this implementation of the sampler its simply hard routed to |
34 |
the 2nd effect send of the sampler channel, no matter what the actual |
35 |
effect type is) |
36 |
- implemented support for internal LADSPA effects |
37 |
- raised limit of program change queue from 100 to 512 |
38 |
(as suggested by Alex Stone) |
39 |
|
40 |
* Gigasampler format engine: |
41 |
- implemented the "round robin keyboard" dimension |
42 |
- fixed round robin and random dimensions for cases when number of |
43 |
dimension zones is not a power of two |
44 |
- made round robin use a counter for each region instead of each |
45 |
key |
46 |
|
47 |
* SFZ format engine: |
48 |
- Initial implementation (not usable yet) |
49 |
- added support for v2 multiple stage envelope generators |
50 |
- added a fine-tuned v1 envelope generator instead of using the |
51 |
one from the gig engine |
52 |
- fixed memory leak and memory handling errors |
53 |
- added support for trigger=first, trigger=legato and sw_previous |
54 |
- allow non-numerical key values ("C#4" for example) |
55 |
- "key" opcode now sets pitch_keycenter too |
56 |
- fixed error when unloading instrument with same sample used by |
57 |
multiple regions |
58 |
- added some opcode aliases, like loopmode for loop_mode, to be |
59 |
more compatible |
60 |
- added support for trigger=release and rt_decay |
61 |
- added support for off_mode=normal |
62 |
- added support for random, seq_position, seq_length and volume |
63 |
- added v1 LFO opcodes to parser (no support in engine yet) |
64 |
- added support for amp_veltrack and amp_velcurve_N |
65 |
- fine-tuned the default velocity curve |
66 |
- added support for transpose |
67 |
- fixed crash when using sw_down/up |
68 |
- improved logic for sw_lokey/hikey/up/down/last |
69 |
- added more v1 aliases to parser |
70 |
- reduced memory usage for sfz data |
71 |
- RT-safeness: avoid malloc in audio thread |
72 |
- fixed a bug that could cause voice stealing to fail |
73 |
- optimized sample lookup |
74 |
- improved support for exclusive groups (group, off_by and |
75 |
off_mode) |
76 |
- added support for controller triggered regions (on_locc/on_hicc) |
77 |
- added support for loop_mode=one_shot |
78 |
- fixed playback of 16 bit wav files on big endian CPUs |
79 |
- added support for Ogg Vorbis sample files |
80 |
- use loop markers from sample file if loop_start and loop_end are |
81 |
not set in sfz file |
82 |
- implemeted filters. Filter types: lowpass, bandpass, bandreject |
83 |
and highpass. 1, 2, 4 and 6 pole filters. Opcodes: fil_type, |
84 |
cutoff, resonance, fil_veltrack, fil_keytrack, fil_keycenter, |
85 |
cutoff_cc, cutoff_chanaft. |
86 |
- bugfix: zero ampeg_sustain didn't work |
87 |
- added support for velocity effect on amplifier envelope time |
88 |
(ampeg_vel2attack, ampeg_vel2decay, ampeg_vel2sustain and |
89 |
ampeg_vel2release) |
90 |
|
91 |
* SoundFont format engine: |
92 |
- Initial implementation (not usable yet) |
93 |
- RT-safeness: avoid malloc in audio thread |
94 |
- fixed a bug that could cause voice stealing to fail |
95 |
|
96 |
* Gigasampler format engine: |
97 |
- bugfix: pitch LFO controller "internal+aftertouch" was broken |
98 |
- bugfix: filter keyboard tracking was broken |
99 |
- filter performance fix (an unnecessary copy was made of the |
100 |
filter parameters in each sub fragment) |
101 |
|
102 |
* Host plugins: |
103 |
- AU bugfix: failed to destroy its audio/MIDI devices |
104 |
- Listen to all interfaces on Mac OS X (INADDR_ANY) |
105 |
- VST bugfix: If the host called resume() before and after |
106 |
changing sample rate or block size, the number of channels was |
107 |
incorrectly set to two. This caused silence in Cubase 5. |
108 |
- save engine type (gig, sfz or sf2) in plugin state |
109 |
- VST: when opening Fantasia, look for both 32 and 64 bit Java on |
110 |
64 bit Windows |
111 |
|
112 |
* MIDI driver: |
113 |
- ALSA MIDI driver supports now "NAME" device parameter, for overriding |
114 |
the ALSA sequencer client name |
115 |
- removed limit of maximum amount of MIDI ports per MIDI device, since |
116 |
there is no reason for this limit |
117 |
|
118 |
* audio driver: |
119 |
- ASIO driver fixes for newer gcc versions (fix from PortAudio) |
120 |
|
121 |
* LSCP server: |
122 |
- added support for sending MIDI CC messages via LSCP command |
123 |
"SEND CHANNEL MIDI_DATA CC <sampler-chan> <ctrl> <val>" |
124 |
- added LSCP command "GET AVAILABLE_EFFECTS" |
125 |
- added LSCP command "LIST AVAILABLE_EFFECTS" |
126 |
- added LSCP command "GET EFFECT INFO <effect-index>" |
127 |
- added LSCP command "CREATE EFFECT_INSTANCE <effect-index>" |
128 |
- added LSCP command |
129 |
"CREATE EFFECT_INSTANCE <effect-system> <module> <effect-name>" |
130 |
- added LSCP command "DESTROY EFFECT_INSTANCE <effect-instance>" |
131 |
- added LSCP command "GET EFFECT_INSTANCES" |
132 |
- added LSCP command "LIST EFFECT_INSTANCES" |
133 |
- added LSCP command "GET EFFECT_INSTANCE INFO <effect-instance>" |
134 |
- added LSCP command |
135 |
"GET EFFECT_INSTANCE_INPUT_CONTROL INFO <effect-instance> <input-control>" |
136 |
- added LSCP command "SET EFFECT_INSTANCE_INPUT_CONTROL VALUE |
137 |
<effect-instance> <input-control> <value>" |
138 |
- added LSCP command "GET SEND_EFFECT_CHAINS <audio-device>" |
139 |
- added LSCP command "LIST SEND_EFFECT_CHAINS <audio-device>" |
140 |
- added LSCP command "ADD SEND_EFFECT_CHAIN <audio-device>" |
141 |
- added LSCP command |
142 |
"REMOVE SEND_EFFECT_CHAIN <audio-device> <effect-chain>" |
143 |
- added LSCP command |
144 |
"GET SEND_EFFECT_CHAIN INFO <audio-device> <effect-chain>" |
145 |
- added LSCP command "APPEND SEND_EFFECT_CHAIN EFFECT <audio-device> |
146 |
<effect-chain> <effect-instance>" |
147 |
- added LSCP command "INSERT SEND_EFFECT_CHAIN EFFECT <audio-device> |
148 |
<effect-chain> <effect-chain-pos> <effect-instance>" |
149 |
- added LSCP command "REMOVE SEND_EFFECT_CHAIN EFFECT <audio-device> |
150 |
<effect-chain> <chain-pos>" |
151 |
- added LSCP command "SET FX_SEND EFFECT <sampler_channel> |
152 |
<fx_send_id> <effect_chain> <chain_pos>" |
153 |
- added LSCP command "REMOVE FX_SEND EFFECT <sampler_channel> <fx_send_id>" |
154 |
|
155 |
* Bug fixes: |
156 |
- Fixed crash which may occur when MIDI key + transpose is out of range |
157 |
- minor valgrind fixes |
158 |
- fixed crash which occured when changing an already deployed sampler |
159 |
channel to a different engine type |
160 |
- fixed crash when deleting a sampler channel or changing engine |
161 |
type while an instrument load was in progress |
162 |
- bugfix: playing a note while changing the instrument could cause |
163 |
a crash, or give "Handing back unknown region" error messages |
164 |
- bugfix: calling SET VOICES (which Fantasia does on start and |
165 |
refresh) could cause instruments to be unloaded |
166 |
- Mac OS X: fixed name collision of enum in EffectControl and |
167 |
wrong name of destructor in AudioOutputDeviceCoreAudio.cpp |
168 |
- Mac OS X: fixed hanging threads |
169 |
- Windows: disabled the previous thread fix on non-Mac systems, as |
170 |
it caused hanging threads on Windows |
171 |
|
172 |
Version 1.0.0 (31 July 2009) |
173 |
|
174 |
* packaging changes: |
175 |
- autoconf bugfix: the PKG_CONFIG variable wasn't initialized properly, |
176 |
causing e.g. the libgig test to fail when |
177 |
"./configure --disable-jack-driver" was used |
178 |
(patch by Alexis Ballier) |
179 |
- fixed compilation with gcc 4.3 |
180 |
- fixes for building on OS X (thanks to Ebrahim Mayat for testing) |
181 |
- fixed configure so it detects x86_64 (#107) |
182 |
- fixes for building with newer MinGW versions |
183 |
- fix for building with bison 2.4 (#111) |
184 |
- fixed building with libgig installed in a non-standard directory |
185 |
- minor fix in configure for mmsystem.h detection on MinGW |
186 |
- Windows: look for editor plugins and Fantasia using base |
187 |
directory of liblinuxsampler dll (look in the same directory and one |
188 |
directory above) |
189 |
- configure script fix: removed unconditional use of SSE |
190 |
- fixed building with sqlite installed in a non-standard directory |
191 |
- when cross-compiling, don't try to create instruments.db |
192 |
- fix for new mingw-w64 version, which has usleep |
193 |
|
194 |
* general changes: |
195 |
- bugfix: on some POSIX systems instrument editor plugins refused to |
196 |
load as we used a non-portable Linux specific struct field |
197 |
(fixes bug #70, patch by Ronald Baljeu) |
198 |
- fixed endless loop which occured when loading an instrument editor |
199 |
plugin DLL on Windows systems failed |
200 |
- fixed memory leaks that occurred when liblinuxsampler was unloaded |
201 |
- optimized the SynchronizedConfig class so it doesn't wait |
202 |
unnecessarily long after an update |
203 |
- added support for notifying instrument editors on note-on / note-off |
204 |
events (e.g. to highlight the pressed keys on the virtual keyboard |
205 |
of gigedit) |
206 |
- added support for triggering notes by instrument editors (see above) |
207 |
- be verbose on DLL load errors (on Linux) |
208 |
- fixed di-harmonic triangle LFO implementation (this LFO implementation |
209 |
is only used by very few systems, most pick the int math |
210 |
implementation, which is usually faster) |
211 |
- fixes for audio drivers with varying buffer sizes |
212 |
- experimental support for running LinuxSampler as a DSSI, LV2 and |
213 |
VST plugin |
214 |
- notification events for stream/voice count statistics are now sent |
215 |
only when there are actual changes |
216 |
- added memory ordering constraints to improve stability on |
217 |
multi-core and multi-cpu systems |
218 |
- maximum voices and disk streams can now be altered at runtime |
219 |
- fixed CPU feature detection on x86_64 (maybe fixes #108) |
220 |
- automatic stacktrace mechanism is now turned off by default and can |
221 |
be switched on by command line option "--stacktrace" (the automatic |
222 |
stacktrace mechanism seems to be broken on most systems at the moment) |
223 |
- C++ API method InstrumentManager::LaunchInstrumentEditor() now returns |
224 |
a pointer to the launched InstrumentEditor object |
225 |
- added optional 3rd party user data parameter for following |
226 |
liblinuxsampler C++ API methods: InstrumentEditor::Main(), |
227 |
InstrumentEditor::Launch(), |
228 |
InstrumentManager::LaunchInstrumentEditor() |
229 |
- theoretical fix: made SynchronizedConfig follow C++0x memory |
230 |
model more strictly |
231 |
- fixes for using large audio device buffers |
232 |
- Windows: add the installation directory to the DLL search path |
233 |
when loading an editor plugin (solves problems with VST and |
234 |
gigedit on systems with other GTK versions installed) |
235 |
|
236 |
* audio driver: |
237 |
- removed the nonsense audio channel constraint (which was hard coded to |
238 |
max. 100 audio channels) for most audio drivers |
239 |
- JACK audio driver did not offer a device parameter "SAMPLERATE" as |
240 |
opposed to the LSCP specs |
241 |
- bugfix: the SAMPLERATE parameter of some drivers (e.g. JACK) |
242 |
reflected the wrong value |
243 |
- fixed a memory management bug in ASIO driver |
244 |
- Makefile fix: JACK_CFLAGS wasn't used |
245 |
- JACK: use jack_client_open instead of the deprecated |
246 |
jack_client_new |
247 |
- added (experimental) CoreAudio driver |
248 |
- applied old fixes to the ASIO driver that were included in the |
249 |
previous binary release but accidentally never committed to CVS |
250 |
(fixes #117) |
251 |
- fixes for ASIO on mingw-w64 (iasio wrapper is not needed on |
252 |
win64) |
253 |
- VST: added support for sample rate and buffer size changes |
254 |
- VST: close editor (Fantasia) when the VST is removed |
255 |
- VST: avoid opening Fantasia more than once for each VST instance |
256 |
- VST: export main function as "main" on Linux too (fix for |
257 |
energyXT) |
258 |
- VST: changed number of output channels from one stereo to 16 |
259 |
stereo |
260 |
- added channel routing, fxsends and midi maps to the settings |
261 |
stored in the plugin state |
262 |
- performance optimization of AudioChannel::MixTo() and |
263 |
AudioChannel::CopyTo() methods using GCC vector exensions |
264 |
(if available) |
265 |
- ASIO fixes: avoid initializing the device twice, avoid throwing |
266 |
exception when getting parameters from a disconnected device |
267 |
|
268 |
* MIDI driver: |
269 |
- added JACK MIDI driver |
270 |
- dispatch bank select as ordinary CC as well, the user might seriously |
271 |
want to (mis)use it for some purpose ("fixed" in all current MIDI |
272 |
input drivers: ALSA, CoreMIDI, JACK, MidiShare, MME) |
273 |
- bugfix: pitch bend wasn't working with JackMidi, VST, LV2, MME, |
274 |
CoreMidi or AU |
275 |
- fixed mingw-w64 compilation error in MME driver |
276 |
- made program change handling in MIDI thread real-time safe by |
277 |
moving the logic to a non-RT thread |
278 |
- fixed minor memory leak in ALSA driver |
279 |
|
280 |
* instruments database: |
281 |
- avoid time consuming samples scanning when adding instruments |
282 |
to the instruments database |
283 |
- added support for handling lost files in the instruments database |
284 |
- Implemented option for adding instruments in separate directories |
285 |
in the instruments database |
286 |
(patch by Chris Cherrett & Andrew Williams, a bit adjusted) |
287 |
- work-around for missing fnmatch function on Windows to make |
288 |
instrument database compilable |
289 |
- added instrument database support on Windows |
290 |
works with both standalone mode and VST plugin. |
291 |
the instruments DB file is located in |
292 |
%USERPROFILE%\.linuxsampler\instruments.db which allows different |
293 |
databases for each windows user |
294 |
if no DB is present it automatically creates the .linuxsampler subdir |
295 |
and then creates an empty DB |
296 |
- fixed recursive import, which was broken on Windows |
297 |
|
298 |
* Gigasampler format engine: |
299 |
- fixed a memory leak that could happen when a channel was deleted |
300 |
while notes were playing |
301 |
- made it possible to load an instrument even if the audio thread |
302 |
isn't running |
303 |
- added partial support for the "Controller Triggered" MIDI rule, |
304 |
enough for piano gigs with pedal noise samples |
305 |
- minor fix: only mark FX sends as being modified if really the |
306 |
respective FX send MIDI controller was used |
307 |
- added support for GM global device volume SysEx message |
308 |
(can be selected at compile time to either apply globally to the |
309 |
whole sampler [default] or only to the sampler channels that are |
310 |
connected to the respective MIDI input port the SysEx message arrived |
311 |
on) |
312 |
- bugfix: notes triggered at position 0 in the audio buffer were |
313 |
sometimes wrongly killed in the same buffer, causing no sound to |
314 |
be played |
315 |
- added support for chromatic / drumkit mode Roland GS Sysex message |
316 |
which is usally used to switch a part between chromatic sounds and |
317 |
drumkit sounds (as opposed to their common default setting of MIDI |
318 |
part 10 being a drumkit part and all other ones chromatic parts), in |
319 |
LS however one can switch between the first 16 MIDI instrument maps |
320 |
defined for the sampler |
321 |
- made it possible to create multiple sample channels even if the |
322 |
audio thread isn't running |
323 |
- fixed crash when removing channel with active voices (#116) |
324 |
- bugfix: on sample reference changes (instrument editor), only |
325 |
un-cache the respective sample if it's really not used by any |
326 |
sampler engine anymore |
327 |
- re-cache samples in case they were changed by an instrument editor, |
328 |
e.g. when a sample was added while playing (#82) |
329 |
- fixed hanging notes which occured when note-off event had the exact |
330 |
same time stamp as the voice's note-on event and occured both in the |
331 |
same audio fragment cycle (fixes bug #112) |
332 |
- added support for the "fine tune" and "pitch bend range" |
333 |
instrument-level gig parameters |
334 |
- fixed minor artifacts in pitch bend handling |
335 |
- added support for GS Reset SysEx message |
336 |
- allow gig files to use unlimited downward pitch shifting |
337 |
- added a limit check for upward pitch shifting |
338 |
- bugfix: sometimes, when playing a note twice fast, the second |
339 |
note was silent |
340 |
- fixed crash happening when a pitch bend event arrived at the |
341 |
same time a new instrument was loading |
342 |
|
343 |
* LSCP server: |
344 |
- added new LSCP event "CHANNEL_MIDI" which can be used by frontends to |
345 |
react on MIDI data arriving on certain sampler channels (so far only |
346 |
Note-On and Note-Off events are sent via this LSCP event) |
347 |
- added new LSCP event "DEVICE_MIDI" which can be used by frontends to |
348 |
react on MIDI data arriving on certain MIDI input devices (so far only |
349 |
Note-On and Note-Off events are sent via this LSCP event) |
350 |
- added new LSCP commands: FIND LOST DB_INSTRUMENT_FILES and |
351 |
SET DB_INSTRUMENT FILE_PATH |
352 |
- added new LSCP command: SEND CHANNEL MIDI_DATA which can be used by |
353 |
frontends to send MIDI messages to specific sampler channel |
354 |
- added two additional fields to GET FILE INSTRUMENT INFO command - |
355 |
KEY_BINDINGS and KEYSWITCH_BINDINGS |
356 |
- bugfix: the bank number provided by MIDI_INSTRUMENT_INFO notifications |
357 |
was incorrect |
358 |
- Optimized the retrieval of the MIDI instrument mappings |
359 |
- added new LSCP command "SET VOICES" to globally alter the maximum |
360 |
amount of voices, added new LSCP event "GLOBAL_INFO:VOICES" which will |
361 |
be triggered respectively |
362 |
- added new LSCP command "SET STREAMS" to globally alter the maximum |
363 |
amount of disk streams, added new LSCP event "GLOBAL_INFO:STREAMS" |
364 |
which will be triggered respectively |
365 |
- bugfix: retry if "select" returns EINTR (this fixes a crash when |
366 |
a gigedit file dialog is opened) |
367 |
- close all connections when LSCPServer is deleted |
368 |
- hard close of all sockets on exit |
369 |
- bugfix: SET CHANNEL MIDI_INPUT_TYPE didn't work with the MME |
370 |
driver |
371 |
|
372 |
* Bug fixes: |
373 |
- fixed a crash which occurs when removing a sampler channel waiting |
374 |
to start instrument loading after another channel |
375 |
- fixed a crash which occurs when removing a sampler channel with |
376 |
instrument loading in progress (bug #113) |
377 |
- fixed termination caused by uncaught exception when adding MIDI |
378 |
instrument with PERSISTENT load mode |
379 |
- fixed possible iterator invalidations when resetting the sampler |
380 |
- fixed memory leaks when issuing the following LSCP commands: |
381 |
GET AUDIO_OUTPUT_DRIVER INFO |
382 |
GET AUDIO_OUTPUT_DRIVER_PARAMETER INFO |
383 |
GET MIDI_INPUT_DRIVER INFO |
384 |
GET MIDI_INPUT_DRIVER_PARAMETER INFO |
385 |
- fixed possible compilation error when sqlite is not present |
386 |
- fixed orphaned pointers when setting maximum voices limit (bug #118) |
387 |
- fixed crash when changing the audio output device of a sampler |
388 |
channel with loaded instrument and start playing notes |
389 |
- fixed endless loop in Engine::SuspendAll() (bug #120) |
390 |
- fixed a low-level atomic load function that was broken on 64-bit |
391 |
PowerPC, which probably could cause crashes on that platform |
392 |
- fixed a memory management error which could cause a crash when a |
393 |
plugin was unloaded |
394 |
- bugfix: two private structs had the same name, which could cause |
395 |
problems if the linker chose the wrong constructor |
396 |
- fixed low-level ConditionServer usage bug that caused lockups on |
397 |
Windows |
398 |
|
399 |
|
400 |
Version 0.5.1 (6 December 2007) |
401 |
|
402 |
* packaging changes: |
403 |
- added autoconf checks for pthread library |
404 |
- added autoconf check for pthread bug found on certain NPTL-enabled |
405 |
glibc versions (see Gentoo bug report #194076) |
406 |
- added autoconf checks for MS Windows |
407 |
|
408 |
* general changes: |
409 |
- bugfix: the thread used by an editor plugin didn't die when the |
410 |
editor closed |
411 |
- bugfix: Ringbuffer.h: fill_write_space_with_null() did not zero |
412 |
out all the space. operator--() did not apply size_mask after |
413 |
decrementing the read_ptr. DEFAULT_WRAP_ELEMENTS set to 0 in |
414 |
order to avoid problems with the _NonVolatileReader functions. |
415 |
- bugfix: Stream.h: added a missing divide by BytesPerSample in |
416 |
GetWriteSpace(). Since this function is currently only used in |
417 |
the stream's qsort() compare function, it didn't trigger any |
418 |
bugs. |
419 |
- Resampler.h, Synthesizer.h: cubic interpolation now works in |
420 |
24bit mode too. Faster method to read 24bit words on little |
421 |
endian machines (x86): replaced 3 byte reads + shifts with a 1 |
422 |
unaligned 32bit read and shift |
423 |
- experimental support for MS Windows (MIDI input via MME, AUDIO |
424 |
output via ASIO) |
425 |
- made handling of SIGINT signal (Ctrl-C) a bit more robust |
426 |
- support for monitoring the total number of active disk streams |
427 |
(new LSCP commands: GET TOTAL_STREAM_COUNT, |
428 |
SUBSCRIBE TOTAL_STREAM_COUNT, UNSUBSCRIBE TOTAL_STREAM_COUNT) |
429 |
|
430 |
* AUDIO driver: |
431 |
- added Windows ASIO low latency audio driver |
432 |
|
433 |
* MIDI driver: |
434 |
- added MME Windows MIDI driver |
435 |
|
436 |
* LSCP server: |
437 |
- added support for Windows style path / filenames, however with |
438 |
forward slash path separators instead of backslash |
439 |
(i.e. "C:/foo/bar.gig") |
440 |
- allow naughty liblscp to send non-string device parameters within |
441 |
apostrophes as well |
442 |
- added new LSCP commands: "GET FILE INSTRUMENTS <file>", |
443 |
"LIST FILE INSTRUMENTS <file>" and |
444 |
"GET FILE INSTRUMENT INFO <file> <index>" for retrieving informations |
445 |
about an arbitrary instrument file on the system where the sampler is |
446 |
running on |
447 |
|
448 |
Version 0.5.0 (15 October 2007) |
449 |
|
450 |
* packaging changes: |
451 |
- config.h is not going to be installed along with liblinuxsampler's |
452 |
API header files anymore |
453 |
- only the API relevant header (and source) files will be exported to |
454 |
the Doxygen API documentation (explicitly listed in Doxyfile.in) |
455 |
- added completely new XCode project files for Mac OSX which is now |
456 |
capable to execute our autoconf environment, thus no need anymore |
457 |
to maintain the compile time configuration file (osx/version.h) for |
458 |
OSX manually (patch by Toshi Nagata) |
459 |
- fixed buggy boolean --enable-foo / --disable-foo configure script |
460 |
parameters |
461 |
- global.h now only covers global definitions that are needed for the |
462 |
C++ API header files, all implementation internal global definitions |
463 |
are now in global_private.h |
464 |
- atomic.h is not exposed to the C++ API anymore |
465 |
- no need to include config.h anymore for using LS's API header files |
466 |
- fixed warnings in API doc generation |
467 |
|
468 |
* general changes: |
469 |
- replaced the old, confusing MIDI program change mechanism by a |
470 |
flexible MIDI instrument mapper which allows to map arbitrary |
471 |
(MIDI bank, MIDI program) pairs with arbitrary |
472 |
(engine type, instrument file, file index) triplets which will be |
473 |
loaded on the respective channel when such MIDI program change |
474 |
messages arrive, beside that, each entry allows to define a life-time |
475 |
strategy for the instrument, so the user can define whether the |
476 |
instrument should i.e. be permanently loaded (to be able to switch |
477 |
quickly among sounds) or i.e. loaded just on demand when the |
478 |
respective program change arrives (to safe RAM space), as well as a |
479 |
global volume factor for each entry, so the user can adjust the volume |
480 |
dynamic of his mapped instrument collection without having to modify |
481 |
the instrument files, also one can manage arbitrary amount of such |
482 |
MIDI instrument maps and assign each sampler channel individually |
483 |
a certain map, so that i.e. separation of normal instruments and |
484 |
drumkits is possible |
485 |
- new notification events for tracking changes to audio/MIDI devices, |
486 |
MIDI instrument maps, MIDI instruments, FX sends, global volume. |
487 |
- sampler was limited to load max. 200 instruments in the background |
488 |
due to a constant size RingBuffer FIFO which is now replaced by a |
489 |
dynamic (unlimited) size std::list FIFO |
490 |
- added FX sends, these allow to route audio signals to arbitrary audio |
491 |
output channels for being processed by external effect processors |
492 |
(i.e. jack-rack), the send levels are controllable via arbitrary MIDI |
493 |
controllers |
494 |
- global (sampler wide) volume can now be controlled at runtime |
495 |
- Implemented new, improved notification system |
496 |
- fixed compilation errors regarding OSX |
497 |
(patch by Toshi Nagata) |
498 |
- implemented instruments database |
499 |
- added support for escape sequences to the instruments database |
500 |
- added highly experimental support for on-the-fly instrument editing |
501 |
within the sampler's process (by using instrument editor plugins), |
502 |
you'll notice the new "Registered instrument editors:" message on |
503 |
startup, added a new LSCP command: |
504 |
"EDIT CHANNEL INSTRUMENT <sampler-channel>" |
505 |
to spawn a matching instrument editor for the instrument on the |
506 |
given sampler channel, the plugin path can be overridden at compile |
507 |
time with ./configure --enable-plugin-dir=/some/dir |
508 |
- added experimental code for synchronizing instrument editors hosted |
509 |
in the sampler's process to safely edit instruments while playing |
510 |
without a crash (hopefully) by either suspending single regions |
511 |
wherever possible or if unavoidable whole engine(s) |
512 |
- fixed several issues in fundamental "Thread" class: set scheduling |
513 |
policy and priority on thread level, set a minimum stack size for |
514 |
thread (TODO: a reasonable value yet to be tested), bugfix: non-RT |
515 |
threads simply inherited properties of starting thread instead of |
516 |
setting their own policy and priority |
517 |
- minor fix in our automatic stack trace mechanism on crashes, the |
518 |
main process did not wait for the stack trace process to finish |
519 |
its output |
520 |
- fixed some minor memory leaks |
521 |
- reenabled assembly features support, at the moment only for |
522 |
enabling a fast denormal FPU mode (x86 platforms supporting SSE2) |
523 |
- minor assembly fix in x86 features detection (don't use the PIC |
524 |
register, to avoid relocations in the text segment at runtime) |
525 |
- POSIX callback functions of Thread.h are hidden |
526 |
|
527 |
* MIDI driver: |
528 |
- dispatch bank select (MSB and LSB) messages |
529 |
|
530 |
* audio driver: |
531 |
- the ALSA audio output driver parameters now reflect the correct |
532 |
parameter value ranges for the respective selected sound card |
533 |
(patch by Till Wimmer, a bit fixed and extended) |
534 |
|
535 |
* Gigasampler format engine: |
536 |
- if a filter is used and EG2 finishes before EG1, let the voice |
537 |
die when EG2 has finished (this fixes a problem with clicks and |
538 |
voice starvation for some gigs) |
539 |
- playback is no longer disabled during instrument loading |
540 |
- all notes playing on a channel that changes its instrument keep |
541 |
playing with the old instrument until they get a note off |
542 |
command |
543 |
- EG fix: a release value of zero could cause noises or crash |
544 |
- handle MIDI coarse tuning messages (MIDI RPN #0 MSB #2 LSB) |
545 |
- EG fine tuning: when attack is zero the EG starts at a level |
546 |
above max sustain level, which means that there is a decay phase |
547 |
even if sustain is 100% |
548 |
- more EG fixes: the level could sometimes go below zero and cause |
549 |
noises or crashes |
550 |
- minor fix of EGDecay (patch by Toshi Nagata) |
551 |
- fixed compiler error when --enable-override-filter-type was |
552 |
supplied to the configure script (fixes #46) |
553 |
- disk thread: queue sizes are now proportional to CONFIG_MAX_STREAMS |
554 |
instead of a fix value |
555 |
- behavior fix: on MIDI CC# 65 (portamento on / off), 126 (mono mode), |
556 |
127 (solo mode) only kill voices if the respective mode really |
557 |
changed |
558 |
|
559 |
* LSCP server: |
560 |
- fixed compile time error for old Bison versions |
561 |
(i.e. v1.28 found on MacOS 10.4, patch by Toshi Nagata) |
562 |
- parser now supports extended ASCII character set |
563 |
(up to ASCII code 255, i.e. includes now umlauts and accents) |
564 |
- filename arguments in LSCP commands now allow to use escape |
565 |
sequences, that is directly literal as one of: \', \", \\, \n, \r, |
566 |
\f, \t, \v, or as octal ASCII code value like \132, or as hex ASCII |
567 |
code value like \xf2) (fixes bug #24) |
568 |
- the following LSCP commands now also support escape sequences for at |
569 |
least one of their text-based parameters (i.e. name, description): |
570 |
"ADD MIDI_INSTRUMENT_MAP", "MAP MIDI_INSTRUMENT", |
571 |
"SET MIDI_INSTRUMENT_MAP NAME", "SET FX_SEND NAME", "CREATE FX_SEND", |
572 |
"SET DB_INSTRUMENT_DIRECTORY NAME", |
573 |
"SET DB_INSTRUMENT_DIRECTORY DESCRIPTION", "SET DB_INSTRUMENT NAME", |
574 |
"SET DB_INSTRUMENT DESCRIPTION", "FIND DB_INSTRUMENTS", |
575 |
"FIND DB_INSTRUMENT_DIRECTORIES" |
576 |
- returns verbose syntax errors (line and column where syntax error |
577 |
occured, the unexpected character and the actually expected, possible |
578 |
character(s), the latter only if less than 5 possibilities) |
579 |
- made sure that LSCP syntax is not affected by gigedit locale |
580 |
settings |
581 |
- bugfix regarding strings parameter lists: all comma separated lists |
582 |
of strings were treated as being one string containing commas |
583 |
(fixes #57) |
584 |
|
585 |
* Bug fixes: |
586 |
- fixed crash occurring on certain LSCP scripts (Bug 39) |
587 |
- another thread safety fix for lscp "load engine" and "set |
588 |
channel audio output device" |
589 |
- fixed a crash which occurs when reassigning the same engine |
590 |
on a sampler channel with connected MIDI device |
591 |
- fixed a crash which occurs when changing the number of ports of a MIDI |
592 |
device connected to a sampler channel to number less then or equal |
593 |
to the index of the port to which the sampler channel is connected. |
594 |
- The previous bindings were not been disconnected when altering |
595 |
the ALSA_SEQ_BINDINGS parameter. Introduced a NONE keyword for |
596 |
unsubscribing from all bindings (e.g. ALSA_SEQ_BINDINGS=NONE). |
597 |
- The active stream/voice count statistic was incorrect. |
598 |
- notification events were not been sent for some sampler |
599 |
channel changes |
600 |
- added default min and max values to restrict the number of allowed |
601 |
audio output channels and MIDI input ports |
602 |
- the connection to the PCM interface is now closed when destroying |
603 |
an audio output device |
604 |
- files with slash in their path or filename could not be loaded |
605 |
|
606 |
* test cases: |
607 |
- updated and fixed (haven't been touched in a while) |
608 |
|
609 |
Version 0.4.0 (24 November 2006) |
610 |
|
611 |
* packaging changes: |
612 |
- changed deprecated copyright attribute to license; added ldconfig |
613 |
to post-(un)install steps; added devel package for liblinuxsampler; |
614 |
to linuxsampler.spec (RPM) |
615 |
- install necessary development header files for allowing 3rd party |
616 |
applications to link against liblinuxsampler |
617 |
- liblinuxsampler's API documentation can be generated with 'make docs' |
618 |
(Doxygen required) |
619 |
- added benchmark to automatically detect the best triangle LFO |
620 |
implementation (currently either an integer math solution or a |
621 |
di-harmonic approximation), automatic detection can be overriden |
622 |
with --enable-signed-triang-algo=x and --enable-unsigned-triang-algo=x |
623 |
configure script argument though (mandatory for cross-compilation) |
624 |
- do not automatically pick optimized gcc flags if the user already |
625 |
provided some on his own (as CXXFLAGS) |
626 |
- added compile time option to disable processing of All-Notes-Off MIDI |
627 |
messages |
628 |
- added compile time options to allow disabling the various audio and |
629 |
MIDI drivers |
630 |
- fixed automatic GCC CPU switch detection on PPC |
631 |
(patch by Ebrahim Mayat) |
632 |
|
633 |
* Gigasampler format engine: |
634 |
- extensive synthesis optimization |
635 |
(reimplementation of EGs and LFO(s), removed synthesis parameter |
636 |
prerendering and the synthesis parameter matrix in general, splitting |
637 |
each audio fragment into subfragments now where each subfragment uses |
638 |
constant pitch and filter coefficients. The volume coefficient is |
639 |
linearly interpolated inside a subfragment, unless |
640 |
--disable-interpolate-volume is set.) |
641 |
- fine tuning of the EG modulation parameters |
642 |
- improved filter cutoff calculation by adding support for the |
643 |
following gig parameters: Cutoff freq (used when no cutoff |
644 |
controller is defined), Control invert, Minimum cutoff, Velocity |
645 |
curve and Velocity range. The keyboard tracking now scales |
646 |
cutoff frequency instead of resonance. |
647 |
- added support for gig parameter Resonance. |
648 |
- fixed bug in sysex handling (patch by Juan Linietsky) |
649 |
- added global volume attenuation of -9 dB (0.35f) to prevent clipping |
650 |
which can be overridden with --enable-global-attenuation |
651 |
- EG fixes: made the length of "attack hold" stage more |
652 |
accurate. Release stage can now start before attack stage |
653 |
ends. Cancel release didn't work when sustain was zero. Attack |
654 |
time now has a minimal value to prevent clicks. |
655 |
- fixed pitch changes larger than one octave |
656 |
- fixed EG3 (pitch envelope) synthesis which was neutral all the time |
657 |
- implemented portamento mode and solo mode (a.k.a 'mono mode'): |
658 |
all modes can be altered via standard GM messages, that is CC5 for |
659 |
altering portamento time, CC65 for enabling / disabling portamento |
660 |
mode, CC126 for enabling solo mode and CC127 for disabling solo mode |
661 |
- fine tuning of the curves for volume (CC7), pan (CC10 and gig |
662 |
parameter) and crossfade |
663 |
- added support for the "attenuation controller threshold" gig |
664 |
parameter |
665 |
- added smoothing of volume changes caused by control change |
666 |
messages |
667 |
- sample loop parameters are now taken from the DimensionRegion |
668 |
instead of the wave chunk |
669 |
- fixed keyswitching for v3 gigs with a number of keyswitch splits |
670 |
not equal to a power of two |
671 |
- reimplementation of the filter algorithm. The new filter is more |
672 |
accurate and supports all gig filter types, including bandreject |
673 |
and lowpass turbo. |
674 |
- real support for 24 bit samples - samples are not truncated to |
675 |
16 bits anymore |
676 |
- support for aftertouch (channel pressure, not polyphonic |
677 |
aftertouch) |
678 |
- LFO1 behaviour fixed (dampening from max volume instead of |
679 |
amplifying from 0) |
680 |
|
681 |
* LSCP server: |
682 |
- fixed application exit on broken pipe error (fixes bug #20) |
683 |
- fixed the notification messages delay due to lack of |
684 |
network activity (fixes bug #26) |
685 |
- fixed parser bug which occured on space(s) within device parameters |
686 |
(e.g. "SET AUDIO_OUTPUT_CHANNEL_PARAMETER 0 0 SOMEPARAM='foo bar'") |
687 |
|
688 |
* audio driver: |
689 |
- added aRts audio output driver (by no means RT safe) |
690 |
|
691 |
* MIDI driver: |
692 |
- fixed legacy sysex code which caused dispatching of MIDI SysEx |
693 |
messages several times instead of once |
694 |
- API extension for MIDI drivers which already supply exact time stamps |
695 |
for events (i.e. for offline rendering based MIDI drivers) |
696 |
- added checks for bad MIDI data |
697 |
|
698 |
* linuxsampler application: |
699 |
- show available sampler engine types on startup |
700 |
|
701 |
* general changes: |
702 |
- support for muting sampler channels and solo mode of the same, two new |
703 |
LSCP commands ("SET CHANNEL MUTE" and "SET CHANNEL SOLO") and two new |
704 |
fields ("MUTE" and "SOLO") for command "GET CHANNEL INFO" were |
705 |
introduced for this, the behavior is the same like on a mixer console |
706 |
(patch by Grigor Iliev, a bit adjusted). Also added configure option |
707 |
--enable-process-muted-channels which can be used to enable the |
708 |
processing of muted channels. |
709 |
- support for sostenuto pedal |
710 |
- support for monitoring the total number of active voices |
711 |
(new LSCP commands: GET TOTAL_VOICE_COUNT, GET TOTAL_VOICE_COUNT_MAX, |
712 |
SUBSCRIBE TOTAL_VOICE_COUNT, UNSUBSCRIBE TOTAL_VOICE_COUNT) |
713 |
- fixed some memory management errors |
714 |
- fixed some concurrency problems that could lead to crashes when |
715 |
LSCP commands were executed |
716 |
- fixed crash when instrument loading failed previously on the same |
717 |
sampler channel (fixes bug #36) |
718 |
|
719 |
Version 0.3.3 (15 July 2005) |
720 |
|
721 |
* packaging changes: |
722 |
- fixed compilation with gcc 4.0 |
723 |
|
724 |
* Gigasampler format engine: |
725 |
- LFO preprocessing was performed even though the respective LFO was |
726 |
disabled by the instrument patch (minor efficiency fix) |
727 |
- if period time of chosen audio device is too small (< MIN_RELEASE_TIME) |
728 |
for volume ramp downs in the same fragment (needed for current voice |
729 |
stealing implementation) then simply inform the user with a warning |
730 |
about possible click sounds and reduce the volume ramp down |
731 |
appropriately instead of cancelling the audio device connection |
732 |
|
733 |
Version 0.3.2 (24 June 2005) |
734 |
|
735 |
* packaging changes: |
736 |
- updated autotools build files to compile on OS X |
737 |
(conditional compilation of CoreMIDI and MidiShare drivers) |
738 |
- hand-crafted assembly optimization code can be disabled with |
739 |
'./configure --disable-asm' (definitely not recommended) |
740 |
- fixed 'make dist' rule to include all necessary files |
741 |
- require automake (>= 1.5) for 'make -f Makefile.cvs' |
742 |
(due to 'dist-bzip2' automake option) |
743 |
|
744 |
* Gigasampler format engine: |
745 |
- support for the gig parameters for "release velocity response" curves |
746 |
- fine tuning of the EGADSR envelope |
747 |
- volume of release triggered samples now depends on note-on velocity, |
748 |
note length and gig parameter "release trigger decay" instead of |
749 |
note-off velocity. |
750 |
- revised voice stealing |
751 |
(fixes crash and endless loop caused by voice stealing) |
752 |
- don't reset scale tuning on instrument or audio output device change |
753 |
- handle key group conflicts right at the beginning of each fragment |
754 |
instead of when the respective voice is actually launched |
755 |
(fixes undefined behavior if stolen voices belonged to a key group - |
756 |
this case was followed by a "killed voice survived" error message) |
757 |
- fixed minor issue with null/silence samples |
758 |
(those stole voices even though they don't need a voice at all which |
759 |
resulted in "voice stealing didn't work out" messages) |
760 |
- don't reset volume, pan, pitch and MIDI controller values on |
761 |
instrument or audio output device change |
762 |
|
763 |
* LSCP server: |
764 |
- fixed some crashes (patch by Grigor Iliev, fixes #19) |
765 |
- fixed LSCP event "CHANNEL_INFO" notification |
766 |
(e.g. did not notify on volume changes or MIDI program change events) |
767 |
|
768 |
* linuxsampler application: |
769 |
- added command line parameters --lscp-addr and --lscp-port to override |
770 |
default IP address and port of LSCP server |
771 |
|
772 |
Version 0.3.1 (24 May 2005) |
773 |
|
774 |
* initial release |