1 |
Version CVS HEAD (?) |
Version CVS HEAD (?) |
2 |
|
|
3 |
* packaging changes: |
* packaging changes: |
4 |
|
- fixed building with newer MinGW-w64 |
5 |
|
|
6 |
|
* general changes: |
7 |
|
- Refactoring: moved the independent code from |
8 |
|
the Gigasampler format engine to base classes |
9 |
|
- command line option '--profile' is currently disabled, since the |
10 |
|
respective profiling code is currently broken |
11 |
|
- Introduced support for optional environment variable |
12 |
|
"LINUXSAMPLER_PLUGIN_DIR" which allows to override the directory |
13 |
|
where the sampler shall look for instrument editor plugins |
14 |
|
(patch by Luis Garrido, slightly modified). |
15 |
|
|
16 |
|
* Gigasampler format engine: |
17 |
|
- implemented the "round robin keyboard" dimension |
18 |
|
- fixed round robin and random dimensions for cases when number of |
19 |
|
dimension zones is not a power of two |
20 |
|
- made round robin use a counter for each region instead of each |
21 |
|
key |
22 |
|
|
23 |
|
* SFZ format engine: |
24 |
|
- Initial implementation (not usable yet) |
25 |
|
- added support for v2 multiple stage envelope generators |
26 |
|
- added a fine-tuned v1 envelope generator instead of using the |
27 |
|
one from the gig engine |
28 |
|
- fixed memory leak and memory handling errors |
29 |
|
- added support for trigger=first, trigger=legato and sw_previous |
30 |
|
- allow non-numerical key values ("C#4" for example) |
31 |
|
- "key" opcode now sets pitch_keycenter too |
32 |
|
- fixed error when unloading instrument with same sample used by |
33 |
|
multiple regions |
34 |
|
- added some opcode aliases, like loopmode for loop_mode, to be |
35 |
|
more compatible |
36 |
|
- added support for trigger=release and rt_decay |
37 |
|
- added support for off_mode=normal |
38 |
|
- added support for random, seq_position, seq_length and volume |
39 |
|
- added v1 LFO opcodes to parser (no support in engine yet) |
40 |
|
- added support for amp_veltrack and amp_velcurve_N |
41 |
|
- fine-tuned the default velocity curve |
42 |
|
- added support for transpose |
43 |
|
- fixed crash when using sw_down/up |
44 |
|
- improved logic for sw_lokey/hikey/up/down/last |
45 |
|
- added more v1 aliases to parser |
46 |
|
- reduced memory usage for sfz data |
47 |
|
- RT-safeness: avoid malloc in audio thread |
48 |
|
- fixed a bug that could cause voice stealing to fail |
49 |
|
|
50 |
|
* SoundFont format engine: |
51 |
|
- Initial implementation (not usable yet) |
52 |
|
- RT-safeness: avoid malloc in audio thread |
53 |
|
- fixed a bug that could cause voice stealing to fail |
54 |
|
|
55 |
|
* Host plugins: |
56 |
|
- AU bugfix: failed to destroy its audio/MIDI devices |
57 |
|
- Listen to all interfaces on Mac OS X (INADDR_ANY) |
58 |
|
- VST bugfix: If the host called resume() before and after |
59 |
|
changing sample rate or block size, the number of channels was |
60 |
|
incorrectly set to two. This caused silence in Cubase 5. |
61 |
|
|
62 |
|
* MIDI driver: |
63 |
|
- ALSA MIDI driver supports now "NAME" device parameter, for overriding |
64 |
|
the ALSA sequencer client name |
65 |
|
|
66 |
|
* LSCP server: |
67 |
|
- added support for sending MIDI CC messages via LSCP command |
68 |
|
"SEND CHANNEL MIDI_DATA CC <sampler-chan> <ctrl> <val>" |
69 |
|
|
70 |
|
* Bug fixes: |
71 |
|
- Fixed crash which may occur when MIDI key + transpose is out of range |
72 |
|
|
73 |
|
Version 1.0.0 (31 July 2009) |
74 |
|
|
75 |
|
* packaging changes: |
76 |
|
- autoconf bugfix: the PKG_CONFIG variable wasn't initialized properly, |
77 |
|
causing e.g. the libgig test to fail when |
78 |
|
"./configure --disable-jack-driver" was used |
79 |
|
(patch by Alexis Ballier) |
80 |
|
- fixed compilation with gcc 4.3 |
81 |
|
- fixes for building on OS X (thanks to Ebrahim Mayat for testing) |
82 |
|
- fixed configure so it detects x86_64 (#107) |
83 |
|
- fixes for building with newer MinGW versions |
84 |
|
- fix for building with bison 2.4 (#111) |
85 |
|
- fixed building with libgig installed in a non-standard directory |
86 |
|
- minor fix in configure for mmsystem.h detection on MinGW |
87 |
|
- Windows: look for editor plugins and Fantasia using base |
88 |
|
directory of liblinuxsampler dll (look in the same directory and one |
89 |
|
directory above) |
90 |
|
- configure script fix: removed unconditional use of SSE |
91 |
|
- fixed building with sqlite installed in a non-standard directory |
92 |
|
- when cross-compiling, don't try to create instruments.db |
93 |
|
- fix for new mingw-w64 version, which has usleep |
94 |
|
|
95 |
|
* general changes: |
96 |
|
- bugfix: on some POSIX systems instrument editor plugins refused to |
97 |
|
load as we used a non-portable Linux specific struct field |
98 |
|
(fixes bug #70, patch by Ronald Baljeu) |
99 |
|
- fixed endless loop which occured when loading an instrument editor |
100 |
|
plugin DLL on Windows systems failed |
101 |
|
- fixed memory leaks that occurred when liblinuxsampler was unloaded |
102 |
|
- optimized the SynchronizedConfig class so it doesn't wait |
103 |
|
unnecessarily long after an update |
104 |
|
- added support for notifying instrument editors on note-on / note-off |
105 |
|
events (e.g. to highlight the pressed keys on the virtual keyboard |
106 |
|
of gigedit) |
107 |
|
- added support for triggering notes by instrument editors (see above) |
108 |
|
- be verbose on DLL load errors (on Linux) |
109 |
|
- fixed di-harmonic triangle LFO implementation (this LFO implementation |
110 |
|
is only used by very few systems, most pick the int math |
111 |
|
implementation, which is usually faster) |
112 |
|
- fixes for audio drivers with varying buffer sizes |
113 |
|
- experimental support for running LinuxSampler as a DSSI, LV2 and |
114 |
|
VST plugin |
115 |
|
- notification events for stream/voice count statistics are now sent |
116 |
|
only when there are actual changes |
117 |
|
- added memory ordering constraints to improve stability on |
118 |
|
multi-core and multi-cpu systems |
119 |
|
- maximum voices and disk streams can now be altered at runtime |
120 |
|
- fixed CPU feature detection on x86_64 (maybe fixes #108) |
121 |
|
- automatic stacktrace mechanism is now turned off by default and can |
122 |
|
be switched on by command line option "--stacktrace" (the automatic |
123 |
|
stacktrace mechanism seems to be broken on most systems at the moment) |
124 |
|
- C++ API method InstrumentManager::LaunchInstrumentEditor() now returns |
125 |
|
a pointer to the launched InstrumentEditor object |
126 |
|
- added optional 3rd party user data parameter for following |
127 |
|
liblinuxsampler C++ API methods: InstrumentEditor::Main(), |
128 |
|
InstrumentEditor::Launch(), |
129 |
|
InstrumentManager::LaunchInstrumentEditor() |
130 |
|
- theoretical fix: made SynchronizedConfig follow C++0x memory |
131 |
|
model more strictly |
132 |
|
- fixes for using large audio device buffers |
133 |
|
- Windows: add the installation directory to the DLL search path |
134 |
|
when loading an editor plugin (solves problems with VST and |
135 |
|
gigedit on systems with other GTK versions installed) |
136 |
|
|
137 |
|
* audio driver: |
138 |
|
- removed the nonsense audio channel constraint (which was hard coded to |
139 |
|
max. 100 audio channels) for most audio drivers |
140 |
|
- JACK audio driver did not offer a device parameter "SAMPLERATE" as |
141 |
|
opposed to the LSCP specs |
142 |
|
- bugfix: the SAMPLERATE parameter of some drivers (e.g. JACK) |
143 |
|
reflected the wrong value |
144 |
|
- fixed a memory management bug in ASIO driver |
145 |
|
- Makefile fix: JACK_CFLAGS wasn't used |
146 |
|
- JACK: use jack_client_open instead of the deprecated |
147 |
|
jack_client_new |
148 |
|
- added (experimental) CoreAudio driver |
149 |
|
- applied old fixes to the ASIO driver that were included in the |
150 |
|
previous binary release but accidentally never committed to CVS |
151 |
|
(fixes #117) |
152 |
|
- fixes for ASIO on mingw-w64 (iasio wrapper is not needed on |
153 |
|
win64) |
154 |
|
- VST: added support for sample rate and buffer size changes |
155 |
|
- VST: close editor (Fantasia) when the VST is removed |
156 |
|
- VST: avoid opening Fantasia more than once for each VST instance |
157 |
|
- VST: export main function as "main" on Linux too (fix for |
158 |
|
energyXT) |
159 |
|
- VST: changed number of output channels from one stereo to 16 |
160 |
|
stereo |
161 |
|
- added channel routing, fxsends and midi maps to the settings |
162 |
|
stored in the plugin state |
163 |
|
- performance optimization of AudioChannel::MixTo() and |
164 |
|
AudioChannel::CopyTo() methods using GCC vector exensions |
165 |
|
(if available) |
166 |
|
- ASIO fixes: avoid initializing the device twice, avoid throwing |
167 |
|
exception when getting parameters from a disconnected device |
168 |
|
|
169 |
|
* MIDI driver: |
170 |
|
- added JACK MIDI driver |
171 |
|
- dispatch bank select as ordinary CC as well, the user might seriously |
172 |
|
want to (mis)use it for some purpose ("fixed" in all current MIDI |
173 |
|
input drivers: ALSA, CoreMIDI, JACK, MidiShare, MME) |
174 |
|
- bugfix: pitch bend wasn't working with JackMidi, VST, LV2, MME, |
175 |
|
CoreMidi or AU |
176 |
|
- fixed mingw-w64 compilation error in MME driver |
177 |
|
- made program change handling in MIDI thread real-time safe by |
178 |
|
moving the logic to a non-RT thread |
179 |
|
- fixed minor memory leak in ALSA driver |
180 |
|
|
181 |
|
* instruments database: |
182 |
|
- avoid time consuming samples scanning when adding instruments |
183 |
|
to the instruments database |
184 |
|
- added support for handling lost files in the instruments database |
185 |
|
- Implemented option for adding instruments in separate directories |
186 |
|
in the instruments database |
187 |
|
(patch by Chris Cherrett & Andrew Williams, a bit adjusted) |
188 |
|
- work-around for missing fnmatch function on Windows to make |
189 |
|
instrument database compilable |
190 |
|
- added instrument database support on Windows |
191 |
|
works with both standalone mode and VST plugin. |
192 |
|
the instruments DB file is located in |
193 |
|
%USERPROFILE%\.linuxsampler\instruments.db which allows different |
194 |
|
databases for each windows user |
195 |
|
if no DB is present it automatically creates the .linuxsampler subdir |
196 |
|
and then creates an empty DB |
197 |
|
- fixed recursive import, which was broken on Windows |
198 |
|
|
199 |
|
* Gigasampler format engine: |
200 |
|
- fixed a memory leak that could happen when a channel was deleted |
201 |
|
while notes were playing |
202 |
|
- made it possible to load an instrument even if the audio thread |
203 |
|
isn't running |
204 |
|
- added partial support for the "Controller Triggered" MIDI rule, |
205 |
|
enough for piano gigs with pedal noise samples |
206 |
|
- minor fix: only mark FX sends as being modified if really the |
207 |
|
respective FX send MIDI controller was used |
208 |
|
- added support for GM global device volume SysEx message |
209 |
|
(can be selected at compile time to either apply globally to the |
210 |
|
whole sampler [default] or only to the sampler channels that are |
211 |
|
connected to the respective MIDI input port the SysEx message arrived |
212 |
|
on) |
213 |
|
- bugfix: notes triggered at position 0 in the audio buffer were |
214 |
|
sometimes wrongly killed in the same buffer, causing no sound to |
215 |
|
be played |
216 |
|
- added support for chromatic / drumkit mode Roland GS Sysex message |
217 |
|
which is usally used to switch a part between chromatic sounds and |
218 |
|
drumkit sounds (as opposed to their common default setting of MIDI |
219 |
|
part 10 being a drumkit part and all other ones chromatic parts), in |
220 |
|
LS however one can switch between the first 16 MIDI instrument maps |
221 |
|
defined for the sampler |
222 |
|
- made it possible to create multiple sample channels even if the |
223 |
|
audio thread isn't running |
224 |
|
- fixed crash when removing channel with active voices (#116) |
225 |
|
- bugfix: on sample reference changes (instrument editor), only |
226 |
|
un-cache the respective sample if it's really not used by any |
227 |
|
sampler engine anymore |
228 |
|
- re-cache samples in case they were changed by an instrument editor, |
229 |
|
e.g. when a sample was added while playing (#82) |
230 |
|
- fixed hanging notes which occured when note-off event had the exact |
231 |
|
same time stamp as the voice's note-on event and occured both in the |
232 |
|
same audio fragment cycle (fixes bug #112) |
233 |
|
- added support for the "fine tune" and "pitch bend range" |
234 |
|
instrument-level gig parameters |
235 |
|
- fixed minor artifacts in pitch bend handling |
236 |
|
- added support for GS Reset SysEx message |
237 |
|
- allow gig files to use unlimited downward pitch shifting |
238 |
|
- added a limit check for upward pitch shifting |
239 |
|
- bugfix: sometimes, when playing a note twice fast, the second |
240 |
|
note was silent |
241 |
|
- fixed crash happening when a pitch bend event arrived at the |
242 |
|
same time a new instrument was loading |
243 |
|
|
244 |
|
* LSCP server: |
245 |
|
- added new LSCP event "CHANNEL_MIDI" which can be used by frontends to |
246 |
|
react on MIDI data arriving on certain sampler channels (so far only |
247 |
|
Note-On and Note-Off events are sent via this LSCP event) |
248 |
|
- added new LSCP event "DEVICE_MIDI" which can be used by frontends to |
249 |
|
react on MIDI data arriving on certain MIDI input devices (so far only |
250 |
|
Note-On and Note-Off events are sent via this LSCP event) |
251 |
|
- added new LSCP commands: FIND LOST DB_INSTRUMENT_FILES and |
252 |
|
SET DB_INSTRUMENT FILE_PATH |
253 |
|
- added new LSCP command: SEND CHANNEL MIDI_DATA which can be used by |
254 |
|
frontends to send MIDI messages to specific sampler channel |
255 |
|
- added two additional fields to GET FILE INSTRUMENT INFO command - |
256 |
|
KEY_BINDINGS and KEYSWITCH_BINDINGS |
257 |
|
- bugfix: the bank number provided by MIDI_INSTRUMENT_INFO notifications |
258 |
|
was incorrect |
259 |
|
- Optimized the retrieval of the MIDI instrument mappings |
260 |
|
- added new LSCP command "SET VOICES" to globally alter the maximum |
261 |
|
amount of voices, added new LSCP event "GLOBAL_INFO:VOICES" which will |
262 |
|
be triggered respectively |
263 |
|
- added new LSCP command "SET STREAMS" to globally alter the maximum |
264 |
|
amount of disk streams, added new LSCP event "GLOBAL_INFO:STREAMS" |
265 |
|
which will be triggered respectively |
266 |
|
- bugfix: retry if "select" returns EINTR (this fixes a crash when |
267 |
|
a gigedit file dialog is opened) |
268 |
|
- close all connections when LSCPServer is deleted |
269 |
|
- hard close of all sockets on exit |
270 |
|
- bugfix: SET CHANNEL MIDI_INPUT_TYPE didn't work with the MME |
271 |
|
driver |
272 |
|
|
273 |
|
* Bug fixes: |
274 |
|
- fixed a crash which occurs when removing a sampler channel waiting |
275 |
|
to start instrument loading after another channel |
276 |
|
- fixed a crash which occurs when removing a sampler channel with |
277 |
|
instrument loading in progress (bug #113) |
278 |
|
- fixed termination caused by uncaught exception when adding MIDI |
279 |
|
instrument with PERSISTENT load mode |
280 |
|
- fixed possible iterator invalidations when resetting the sampler |
281 |
|
- fixed memory leaks when issuing the following LSCP commands: |
282 |
|
GET AUDIO_OUTPUT_DRIVER INFO |
283 |
|
GET AUDIO_OUTPUT_DRIVER_PARAMETER INFO |
284 |
|
GET MIDI_INPUT_DRIVER INFO |
285 |
|
GET MIDI_INPUT_DRIVER_PARAMETER INFO |
286 |
|
- fixed possible compilation error when sqlite is not present |
287 |
|
- fixed orphaned pointers when setting maximum voices limit (bug #118) |
288 |
|
- fixed crash when changing the audio output device of a sampler |
289 |
|
channel with loaded instrument and start playing notes |
290 |
|
- fixed endless loop in Engine::SuspendAll() (bug #120) |
291 |
|
- fixed a low-level atomic load function that was broken on 64-bit |
292 |
|
PowerPC, which probably could cause crashes on that platform |
293 |
|
- fixed a memory management error which could cause a crash when a |
294 |
|
plugin was unloaded |
295 |
|
- bugfix: two private structs had the same name, which could cause |
296 |
|
problems if the linker chose the wrong constructor |
297 |
|
- fixed low-level ConditionServer usage bug that caused lockups on |
298 |
|
Windows |
299 |
|
|
300 |
|
|
301 |
|
Version 0.5.1 (6 December 2007) |
302 |
|
|
303 |
|
* packaging changes: |
304 |
- added autoconf checks for pthread library |
- added autoconf checks for pthread library |
305 |
- added autoconf check for pthread bug found on certain NPTL-enabled |
- added autoconf check for pthread bug found on certain NPTL-enabled |
306 |
glibc versions (see Gentoo bug report #194076) |
glibc versions (see Gentoo bug report #194076) |