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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 3581 - (show annotations) (download)
Fri Aug 30 11:40:25 2019 UTC (5 months, 2 weeks ago) by schoenebeck
File size: 87170 byte(s)
NKSP: Allow more wider support of standard measuring units & 'final'ness.

* Raised compiler requirement to be C++14 compliant (due to severe
  restrictions regarding C-style aggregate initializer lists in C++11
  which are now massively used throughout the code base).

* NKSP VM API: Allow units and 'final'ness to be returned as result from
  built-in functions (added methods VMFunction::returnUnitType() and
  VMFunction::returnsFinal() for that purpose which must be implemented by
  built-in function implementations).

* NKSP language: Allow metric unit prefixes of numeric scalar and array
  variables to be changed freely at runtime (unlike unit types like Hz etc.
  which are still sticky parse-time features of variables which cannot be
  changed at runtime for the intentional sake of determinism).

* NKSP language: 'final' values are prohibited for array variables for now
  (attempt causes a parsers error).

* NKSP language: expressions with unit types (e.g. Hz) are prohibited for
  conditions of runtime control structures like if(), while(), select()
  (attempt causes a parser error).

* NKSP VM API: Allow built-in functions to perform their own, individual
  parse time checks of arguments going to be passed to the function at
  runtime (added method VMFunction::checkArgs() for that purpose).

* NKSP language: raise parser warning if only one operand of binary
  operators (like logical 'or' comparison) contain a 'final' value (because
  it would always yield in a 'final' result in such cases).

* NKSP language: Allow comparison (=, #, <, >, <=, >=) of values with
  different metric unit prefixes, which will behave as expected (e.g.
  result of expression '1000us < 2ms' is true).

* NKSP language: Allow adding values with different metric unit prefixes
  (e.g. result of expression '100Hz + 5kHz' is '5100Hz').

* NKSP language: Allow subtracting values with different metric unit
  prefixes (e.g. result of expression '1ms - 20us' is '980us').

* NKSP language: Allow multiplying with any metric unit prefixes
  (e.g. result of expression '2k * 3ms' is '6s'), however multiplications
  with unit types on both sides (e.g. '2s * 2s') is still prohibited since
  we don't have any considerable practical use for a term like '4s^2'
  (hence any attempt multiplying two unit types still causes parser error).

* NKSP language: Allow dividing by any metric unit prefixes and allow
  division of same unit type on both sides (e.g. expression '8kHz / 1000Hz'
  yields in unit free result '8'). So this is now a way to cast units away
  e.g. for passing the result to other expressions, certain function calls
  or variables which are not accepting any units (or that specific unit).

* NKSP language: integer arrays and real number arrays can now be converted
  to strings (e.g. for dumping their content with message() calls for
  script debugging purposes).

* NKSP language: expressions and variables with units are now correctly
  casted to strings (e.g. with message() calls).

* NKSP language: comparing real numbers for equalness (e.g. '~foo = 3.1') or
  unequalness (e.g. '~foo # 3.1') is now less strict and takes the expected
  floating point tolerances into account.

* NKSP VM API: Added methods VMScalarNumberExpr::evalCastInt() and
  VMScalarNumberExpr::evalCastReal().

* NKSP VM API: Added base class 'VMNumberArrayExpr' for classes
  'VMIntArrayExpr' and 'VMRealArrayExpr'.

* NKSP VM API: replaced all unitPrefix() (parse time) methods by
  unitFactor() (runtime) methods.

* Built-in function "exit()" supports now returning units and 'final'ness
  exclusively for test cases.

* The following built-in functions support now units as well: "abs()",
  "random()", "inc()", "dec()", "in_range()", "min()", "max()",
  "real_to_int()", "int()", "int_to_real()" and "real()".

* Built-in functions "array_equal()", "search()" and "sort()" support now
  real number arrays (correctly) as well.

* Added individual parse time checks of arguments to be passed to built-in
  functions "random()", "inc()", "dec()", "in_range()", "min()", "max()",
  "array_equal()" and "search()" specific for their individual purposes.

* Test cases: Added massive amount of NKSP test cases for standard
  measuring units and 'final' operator usage cases.

* Test cases: Added NKSP test cases for (floating point tolerance aware)
  real number equalness / unequalness comparison.

* Bumped version (2.1.1.svn8).

1 Version SVN trunk (?)
2
3 * general changes:
4 - Fixed compiler error in Pool.h.
5 - Require C++14 compiler support.
6 - Autoconf: Added m4/ax_cxx_compile_stdcxx.m4 macro which is used
7 for checking in configure for C++14 support (as mandatory
8 requirement) and automatically adds compiler argument if required
9 (e.g. -std=C++14).
10 - RTMath: Implemented floating point comparison methods
11 fEqual32(float,float) and fEqual64(double,double) which take the
12 expected floating point tolerances into account.
13
14 * Real-time instrument scripts:
15 - Added method ScriptVM::setExitResultEnabled() which allows to
16 explicitly enable the built-in exit() function to optionally accept
17 one function argument; the value of the passed exit() function
18 argument will then become available by calling
19 VMExecContext::exitResult() after script execution.
20 - 64 bit support for NKSP integer scripts variables (declare $foo).
21 - Variable names, function names and preprocessor condition names must start
22 with a regular character (a-z or A-Z); starting them with a digit or
23 underscore is not allowed.
24 - NKSP parser fix: equal comparison operator "=" and not equal comparison
25 operator "#" must only accept integer operands.
26 - NKSP language: Implemented support for standard units like Hertz, seconds,
27 Bel including support for metric unit prefixes; so one can now e.g.
28 conveniently use numbers in scripts like "5us" meaning "5 microseconds",
29 or e.g. "12kHz" meaning "12 kilo Hertz", or e.g. "-14mdB" meaning
30 "minus 14 Millidecibel", or e.g. "28c" meaning "28 cents" (for tuning).
31 - NKSP language: Introduced "final" operator "!" which is specifically
32 intended for synthesis parameter values to denote that the synthesis
33 parameter value is intended to be the "final" value for that synthesis
34 parameter that should explicitly be used by the engine and thus causing
35 the sampler engine to ignore all other modulation sources for the same
36 synthesis parameter (like e.g. LFO, EG); by simply prefixing a value,
37 variable or formula with this new "!" operator the expression is marked as
38 being "final".
39 - NKSP script editor API: Added support for detecting standard unit tokens
40 and their potential metric prefix token.
41 - NKSP language: Added support for NKSP real number literals and
42 arithmetic operations on them (e.g. "(3.9 + 2.9) / 12.3 - 42.0").
43 - NKSP language: Added support for NKSP real number (floating point) script
44 variables (declare ~foo).
45 - NKSP language: Added support for NKSP real number (floating point) array
46 script variables (declare ?foo[]).
47 - Built-in script function "message()" accepts now real number argument as
48 well.
49 - Added built-in script function "real_to_int()" and its short hand form
50 "int()" for casting from real number to integer in NKSP scripts.
51 - Added built-in script function "int_to_real()" and its short hand form
52 "real()" for casting from integer to real number in NKSP scripts.
53 - Allow built-in exit() function to potentially accept real number type
54 argument as well.
55 - Built-in script functions may have a different return type depending on
56 the arguments passed to the function.
57 - Built-in script function "abs()" optionally accepts and returns real
58 number.
59 - Built-in script functions "min()" and "max()" optionally accept real
60 number arguments and return real number as result in that case.
61 - NKSP VM API: Allow units and 'final'ness to be returned as result from
62 built-in functions (added methods VMFunction::returnUnitType() and
63 VMFunction::returnsFinal() for that purpose which must be implemented by
64 built-in function implementations).
65 - NKSP language: Allow metric unit prefixes of numeric scalar and array
66 variables to be changed freely at runtime (unlike unit types like Hz etc.
67 which are still sticky, parse-time features of variables which cannot be
68 changed at runtime for the intentional sake of determinism).
69 - NKSP language: 'final' values are prohibited for array variables for now
70 (attempt causes a parsers error).
71 - NKSP language: expressions with unit types (e.g. Hz) are prohibited for
72 conditions of runtime control structures like if(), while(), select()
73 (attempt causes a parser error).
74 - NKSP VM API: Allow built-in functions to perform their own, individual
75 parse time checks of arguments going to be passed to the function at
76 runtime (added method VMFunction::checkArgs() for that purpose).
77 - NKSP language: raise parser warning if only one operand of binary
78 operators (like logical 'or' comparison) contain a 'final' value (because
79 it would always yield in a 'final' result in such cases).
80 - NKSP language: Allow comparison (=, #, <, >, <=, >=) of values with
81 different metric unit prefixes, which will behave as expected (e.g.
82 result of expression '1000us < 2ms' is true).
83 - NKSP language: Allow adding values with different metric unit prefixes
84 (e.g. result of expression '100Hz + 5kHz' is '5100Hz').
85 - NKSP language: Allow subtracting values with different metric unit
86 prefixes (e.g. result of expression '1ms - 20us' is '980us').
87 - NKSP language: Allow multiplying with any metric unit prefixes
88 (e.g. result of expression '2k * 3ms' is '6s'), however multiplications
89 with unit types on both sides (e.g. '2s * 2s') is still prohibited since
90 we don't have any considerable practical use for a term like '4s^2'
91 (hence any attempt multiplying two unit types still causes parser error).
92 - NKSP language: Allow dividing by any metric unit prefixes and allow
93 division of same unit type on both sides (e.g. expression '8kHz / 1000Hz'
94 yields in unit free result '8'). So this is now a way to cast units away
95 e.g. for passing the result to other expressions, certain function calls
96 or variables which are not accepting any units (or that specific unit).
97 - NKSP language: integer arrays and real number arrays can now be converted
98 to strings (e.g. for dumping their content with message() calls for
99 script debugging purposes).
100 - NKSP language: expressions and variables with units are now correctly
101 casted to strings (e.g. with message() calls).
102 - NKSP language: comparing real numbers for equalness (e.g. '~foo = 3.1') or
103 unequalness (e.g. '~foo # 3.1') is now less strict and takes the expected
104 floating point tolerances into account.
105 - NKSP VM API: Added methods VMScalarNumberExpr::evalCastInt() and
106 VMScalarNumberExpr::evalCastReal().
107 - NKSP VM API: Added base class 'VMNumberArrayExpr' for classes
108 'VMIntArrayExpr' and 'VMRealArrayExpr'.
109 - NKSP VM API: replaced all unitPrefix() (parse time) methods by
110 unitFactor() (runtime) methods.
111 - Built-in function "exit()" supports now returning units and 'final'ness
112 for test cases.
113 - The following built-in functions support now units as well: "abs()",
114 "random()", "inc()", "dec()", "in_range()", "min()", "max()",
115 "real_to_int()", "int()", "int_to_real()" and "real()".
116 - Built-in functions "array_equal()", "search()" and "sort()" support now
117 real number arrays (correctly) as well.
118 - Added individual parse time checks of arguments to be passed to built-in
119 functions "random()", "inc()", "dec()", "in_range()", "min()", "max()",
120 "array_equal()" and "search()" specific for their individual purposes.
121
122 * test cases:
123 - Fixed compiler errors in test cases.
124 - Updated README for how to compile & run test cases.
125 - Updated test case
126 MutexTest::testDoubleLockStillBlocksConcurrentThread() to latest
127 expected behaviour of the Mutex class implementation (recursive
128 mutex type).
129 - Added test cases for NKSP core language aspects and core built-in
130 functions.
131 - Fixed thread tests segfaulting on Linux.
132 - NKSP: Added real number test cases for built-in functions exit(),
133 int_to_real(), real(), real_to_int() and int(), as well as for the
134 plus, minus and negate language operators.
135 - Added massive amount of NKSP test cases for standard measuring units and
136 'final' operator usage cases.
137 - Added NKSP test cases for (floating point tolerance aware) real number
138 equalness / unequalness comparison.
139
140 Version 2.1.1 (27 Jul 2019)
141
142 * Real-time instrument scripts:
143 - Fixed behavior of built-in NKSP functions change_sustain(),
144 change_cutoff_attack(), change_cutoff_decay(), change_cutoff_sustain()
145 and change_cutoff_release().
146
147 * general changes:
148 - Only play release trigger samples on sustain pedal up if this behaviour
149 was explicitly requested by the instrument (otherwise only on note-off).
150 - Fixed compiler warnings.
151 - Fixed compilation error when cross-compiling to Mac.
152 - FX Sends: Provide more useful error messages on routing problems
153 (see bug #169).
154 - LSCP doc: Be more clear describing the two distinct approaches
155 of using external vs. internal effects (see bug #169).
156 - "optional" class: Fixed comparison operators.
157
158 * Gigasampler/GigaStudio format engine:
159 - Format extension: If requested by instrument then don't play release
160 trigger sample on note-off events.
161
162 * SFZ format engine:
163 - Fixed memory leak when releasing samples
164 (fixes bug #307, patch by Jacek Roszkowski)
165 - Fixed potential crash when a sample is shared by more than one region
166 (fixes bug #308, patch by Jacek Roszkowski).
167 - Opcode 'sample': Added support for built-in sample '*silence'
168 (fixes bug #310, patch by Jacek Roszkowski).
169
170 Version 2.1.0 (25 Nov 2017)
171
172 * SFZ format engine:
173 - added support for <global>, <master> and #define (patch by Alby M)
174 - Removed code duplication in SFZ file loading code.
175 - Added support for sfz extension opcode 'script' which may be used to
176 load real-time instrument script file (NKSP script language).
177 - Implemented opcode set_ccN (initial patch by Giovanni Senatore).
178 - Fixed unintended volume fade-in of voices under certain conditions.
179 - sfz parser: allow missing space between header and opcode
180
181 * Gigasampler/GigaStudio format engine:
182 - Fixed clicks and pumping noise with Lowpass Turbo filter on very low
183 cutoff settings.
184 - Got rid of resembling an ancient GSt misbehavior which did not pitch at
185 all if an up-pitch of more than 40 semi tones was requested (I don't
186 think there is any stock gig sound that requires this behavior to
187 resemble its original sound).
188 - Added support for controlling whether the individual EGADSR stages may
189 be aborted (as LinuxSampler extension to the original GigaStudio 4
190 format).
191
192 * general changes:
193 - fixed printf type errors (mostly in debug messages)
194 - use unique_ptr instead of auto_ptr when building with C++11
195 - Added RTAVLTree class which is a real-time safe ordered multi-map, thus
196 allowing to sort data efficiently in real-time safe manner.
197 - RTList class: added methods for moving/inserting elements to arbitrary
198 position within a list.
199 - RTList class: added method fromPtr() for creating an Iterator object from
200 a raw element pointer.
201 - Attempt to partly fix resetting engine channels vs. resetting engine, an
202 overall cleanup of the Reset*(), ConnectAudioDevice(),
203 DisconnectAudioDevice() API methods would still be desirable though,
204 because the current situation is still inconsistent and error prone.
205 - Active voices are now internally grouped to "Note" objects, instead of
206 being directly assigned to a keyboard key. This allows more fine graded
207 processing of voices, which is i.e. required for certain instrument
208 script features.
209 - Fix: Release trigger voices were not spawned on sustain pedal up
210 (CC #64) events.
211 - Fix: Release trigger voices can now also distinguish correctly between
212 note-off and sustain pedal up events.
213 - All engines: Increased ramp speed of volume smoother and pan smoother
214 (while slow rate of crossfade smoother is preserved) to allow quick
215 volume and pan changes by instrument scripts for instance.
216 - gig/sf2/sfz: Fixed aftertouch channel pressure events being processed
217 properly and thus were ignored for certain uses.
218 - Added new C++ API method
219 VirtualMidiDevice::SendChannelPressureToSampler().
220 - windows, 32-bit: fixed potential crashes by making sure the stack in
221 sub threads is 16-byte aligned
222 - fixed numerous compiler warnings
223 - Fixed invalid (note-on) event ID being assigned to new Note objects.
224 - Revised fundamental C++ classes "Thread", "Mutex" and "Condition" which
225 fixes potential undefined behavior.
226 - Fixed Note object leak when triggering notes on keys which did not
227 have a valid sample mapped (fixes bug #252).
228 - Fixed compilation errors when compiling with CONFIG_DEVMODE enabled.
229 - linuxsampler binary fix: option --create-instruments-db ignored
230 subsequent optional argument due to glibc's implementation oddity
231 which expects a "=" sign, but no space between them.
232
233 * packaging changes:
234 - removed unnecessary dependency to libuuid
235 (originated by libgig's usage of it)
236 - Automake: set environment variable GCC_COLORS=auto to allow GCC to
237 auto detect whether it (sh/c)ould output its messages in color.
238 - Debian: Fixed packaging error about invalid "Source-Version"
239 substitution variable.
240 - Debian: Raised Debian compatibility level to Debian 9 "Stretch".
241 - Debian: Added build dependency to libsqlite3-dev for building
242 linuxsampler with instruments DB support.
243
244 * Real-time instrument scripts:
245 - Implemented scheduler for delayed MIDI events and for suspended scripts.
246 - Built-in script function "wait()": implemented support for function's
247 "duration-us" argument, thus scripts using this function are now
248 correctly resumed after the requested amount of microseconds.
249 - Built-in script function "play_note()": implemented support for
250 function's "duration-us" argument, thus notes triggered with this
251 argument are now correctly released after the requested amount of
252 microseconds.
253 - Fix: script events were not cleared when engine channel was reset,
254 potentially causing undefined behavior.
255 - Fixed crash which happened when trying to reference an undeclared
256 variable.
257 - Built-in script function "play_note()": Added support for passing
258 special value -1 for "duration-us" argument, which will cause the
259 triggered note to be released once the original note was released.
260 - Instrument script classes now exported with the liblinuxsampler C++ API.
261 - Added new API method ScriptVM::syntaxHighlighting() which provides
262 a convenient syntax highlighting backend for external instrument
263 script editor applications.
264 - Added new C++ API class "ScriptVMFactory".
265 - Extended parser issues to provide not only first line and first
266 column, but also last line and last column of issue (thus marking
267 the precise span of the issue within the source code).
268 - Reload script automatically after being modified by an instrument
269 editor.
270 - NKSP language grammar correction: allow empty event handler bodies
271 like "on note end on".
272 - Implemented built-in script function "change_vol()".
273 - Implemented built-in script function "change_tune()".
274 - Implemented built-in script function "change_pan()".
275 - Implemented built-in script function "change_cutoff()".
276 - Implemented built-in script function "change_reso()".
277 - Implemented built-in script function "event_status()".
278 - Added built-in script constants "$EVENT_STATUS_INACTIVE" and
279 "$EVENT_STATUS_NOTE_QUEUE" both for being used as flags for
280 "event_status()" function.
281 - NKSP language: Added support for bitwise operators ".or.", ".and."
282 and ".not.".
283 - NKSP language scanner: Fixed IDs matching to require at least one
284 character (i.e. when matching function names or variable names).
285 - NKSP language scanner: disabled unusued rules.
286 - Fixed behavior of play_note() and note_off() functions which must
287 be distinguished engine internally from "real" MIDI note on/off
288 events in order to avoid misbehaviors like hanging notes.
289 - Implemented built-in script variable "$KSP_TIMER".
290 - Implemented built-in script variable "$NKSP_REAL_TIMER".
291 - Implemented built-in script variable "$NKSP_PERF_TIMER".
292 - Implemented built-in script variable "$ENGINE_UPTIME".
293 - Implemented built-in script function "inc()".
294 - Implemented built-in script function "dec()".
295 - NKSP language fix: division expressions were evaluated too often.
296 - NKSP language fix: string concatenation operator was right
297 associative instead of left (to right).
298 - NKSP language correction: allow empty statements in entire language
299 (i.e. "if end if").
300 - Implemented built-in script function "stop_wait()".
301 - Implemented built-in script variable "$NI_CALLBACK_ID".
302 - Implemented built-in script variable "$NI_CALLBACK_TYPE".
303 - Implemented built-in script variable "$NKSP_IGNORE_WAIT".
304 - Added support for read-only built-in variables (respectively
305 handled by the script parser).
306 - Added built-in script constant "$NI_CB_TYPE_INIT".
307 - Added built-in script constant "$NI_CB_TYPE_NOTE".
308 - Added built-in script constant "$NI_CB_TYPE_RELEASE".
309 - Added built-in script constant "$NI_CB_TYPE_CONTROLLER".
310 - NKSP Language: Added support for user defined script functions.
311 - Implemented built-in script function "change_attack()".
312 - Implemented built-in script function "change_decay()".
313 - Implemented built-in script function "change_release()".
314 - Fixed all change_*() built-in script functions to apply their
315 synthesis parameter changes immediately in case the respective note
316 was triggered at the same time, instead of scheduling the parameter
317 change, especially because it would cause some parameter types's
318 changes either to be ramped (i.e. change_vol()) or other types even
319 to have not effect at all (i.e. change_attack()).
320 - Implemented built-in script function "sh_left()".
321 - Implemented built-in script function "sh_right()".
322 - Implemented built-in script function "min()".
323 - Implemented built-in script function "max()".
324 - NKSP Fix: "init" event handler was not always executed when a script
325 was loaded.
326 - NKSP built-in wait() script function: abort script execution if
327 a negative or zero wait time was passed as argument (since this is
328 a common indication of a bug either of the script or even of the
329 engine, which could lead to RT instability or even worse).
330 - ScriptVM: Implemented automatic suspension of RT safety
331 threatening scripts.
332 - Provide more user friendly error messages on syntax errors.
333 - Fixed NKSP parser warning "Not a statement" when assigning an
334 initializer list to an array variable.
335 - Implemented built-in script array variable "%ALL_EVENTS".
336 - Implemented built-in script function "in_range()".
337 - Implemented built-in script function "change_amp_lfo_depth()".
338 - Implemented built-in script function "change_amp_lfo_freq()".
339 - Implemented built-in script function "change_pitch_lfo_depth()".
340 - Implemented built-in script function "change_pitch_lfo_freq()".
341 - Implemented built-in script function "change_vol_time()".
342 - Implemented built-in script function "change_tune_time()".
343 - Implemented built-in script function "fade_in()".
344 - Implemented built-in script function "fade_out()".
345 - Fixed acceptance of wrong data type of parameters passed to built-in
346 script functions "change_vol()", "change_tune()", "change_pan()",
347 "change_cutoff()", "change_reso()", "change_attack()", "change_decay()",
348 "change_release()", "change_amp_lfo_depth()", "change_amp_lfo_freq()",
349 "change_pitch_lfo_depth()" and "change_pitch_lfo_freq()".
350 - Added built-in script function "get_event_par()" and implemented some
351 of its possible parameter selections.
352 - Added built-in script function "set_event_par()" and implemented some
353 of its possible parameter selections.
354 - Fixed a bunch of scheduler time related bugs.
355 - Fixed polyphonic variables not being reset to zero after usage.
356 - Built-in "ignore_event()" function: argument is now optional, like with
357 built-in function "ignore_controller()".
358 - Implemented built-in script function "change_velo()".
359 - Implemented built-in script function "change_note()".
360 - Adjusted behavior of "change_vol()" and "change_tune()" to a more
361 intuitive behavior if used in combination with "change_vol_time()" or
362 "change_tune_time()" respectively: now tuning/volume changes are only
363 assigned (without delay) immediately to a new note if the respective
364 timing function has not been called before, otherwise the volume/tuning
365 changes are automatically faded (before, only the event's time stamp was
366 relevant).
367 - Implemented built-in script function "array_equal()".
368 - Implemented built-in script function "search()".
369 - Implemented built-in script function "sort()".
370 - NKSP Fix: Never suspend "init" event handlers.
371 - Implemented built-in script function "same_region()" (currently only
372 available for gig format engine).
373 - Added built-in script constant "$NKSP_LINEAR".
374 - Added built-in script constant "$NKSP_EASE_IN_EASE_OUT".
375 - Implemented built-in script function "change_vol_curve()".
376 - Implemented built-in script function "change_tune_curve()".
377 - built-in "play_note()" function now supports a sample playback start
378 offset with argument 3, where special value -1 means to use the regular
379 sample offset as defined by the instrument file.
380 - Built-in array variable %KEY_DOWN[] is now a read-only variable.
381 - Built-in variable $EVENT_NOTE is now a read-only variable.
382 - Built-in variable $EVENT_VELOCITY is now a read-only variable.
383 - built-in "play_note()" function now accepts -2 for its fourth argument
384 (note duration) which means the life time of the note shall be sticked
385 to the requested note number of argument 1.
386 - Fix: built-in "play_note()" function now returns 0 as result value if
387 -1 was passed for its fourth argument (note duration) and the respective
388 parent note is already gone.
389 - Implemented built-in script function "change_play_pos()".
390 - NKSP language: Added support for user declared const array variables.
391 - NKSP language: Raise parser warning if array variable is accessed with
392 an index that exceeds the array's size.
393 - NKSP language fix: Unknown characters were not handled correctly.
394 - NKSP language: Added support for "synchronized .. end synchronized"
395 code blocks.
396 - Implemented built-in script function "abort()" which allows to abort
397 another script handler by passing its callback ID.
398 - Fixed potential memory access bug and potential undefined behavior of
399 "init" event handlers.
400 - Print a time stamp along to each call of built-in function "message()".
401 - ScriptVM API: Added VMParserContext::preprocessorComments() which allows
402 to retrieve all code blocks filtered out by the preprocessor.
403 - Added built-in script function "fork()".
404 - Added built-in array variable %NKSP_CALLBACK_CHILD_ID[].
405 - Added built-in variable $NKSP_CALLBACK_PARENT_ID.
406 - Fixed potential crash when accessing dynamic built-in array variables.
407 - Added built-in script function "callback_status()".
408 - Added built-in constant $CALLBACK_STATUS_TERMINATED.
409 - Added built-in constant $CALLBACK_STATUS_QUEUE.
410 - Added built-in constant $CALLBACK_STATUS_RUNNING.
411 - Removed max. value limitation of built-in functions "change_attack()",
412 "change_decay()" and "change_release()" to i.e. allow passing 2000000
413 for doubling the respective time.
414 - NKSP script editor syntax highlighting API: Fixed app termination due
415 to a lexer start condition stack underrun.
416 - NKSP preprocessor: Fixed wrong behavior on nested USE_CODE_IF() and
417 USE_CODE_IF_NOT() preprocessor statements.
418 - NKSP: Added built-in preprocessor condition NKSP_NO_MESSAGE, which
419 can be set to disable all subsequent built-in "message()" function calls
420 on preprocessor level.
421 - Implemented built-in script function "change_sustain()".
422 - NKSP script editor syntax highlighting API: catch all fatal lexer errors,
423 to avoid the editor app to crash on ill-formed text input.
424 - Added built-in script function "change_pan_time()".
425 - Added built-in script function "change_pan_curve()".
426 - Added built-in script function "change_cutoff_attack()".
427 - Added built-in script function "change_cutoff_decay()".
428 - Added built-in script function "change_cutoff_sustain()".
429 - Added built-in script function "change_cutoff_release()".
430 - Added built-in script function "change_cutoff_lfo_depth()".
431 - Added built-in script function "change_cutoff_lfo_freq()".
432
433 * Instruments DB:
434 - Fixed memory access bug of general DB access code which lead to
435 undefined behavior.
436 - Cleanup of instruments DB file creation and opening code.
437 - The instrument DB path of linuxsampler's --create-instruments-db argument
438 is now optional, if it is missing, then a default location is used.
439 - Added support for scanning SFZ (.sfz) files.
440 - Added support for scanning Sound Font (.sf2) files.
441 - Fixed undefined DB transaction behavior.
442
443 Version 2.0.0 (15 July 2015)
444
445 * packaging changes:
446 - fixed building with newer MinGW-w64
447 - Mac OS X: support the new dir for Core Audio SDK
448 - Mac OS X: fixed building outside source directory
449 - made sure all source files for hostplugins are included when
450 doing "make dist"
451 - removed empty directories left from the cvs to svn migration
452 - fixed compilation with gcc 4.6.1
453 - another "make dist" fix, for the LV2 plugin
454 - made --enable-pthread-testcancel default on Mac OS X
455 - minor Makefile fix for building lscpparser when build and source
456 directories are separate
457 - Mac OS X: made it possible to specify plugin installation dir to
458 configure
459 - Mac OS X: Makefile fix for the install-strip target
460 - fixed compilation with gcc 4.7
461 - fixed configure script error with old autoconf versions
462 - lsatomic.h: use gcc provided atomic functions if building with
463 gcc 4.7 and C++11
464 - modernized configure script
465 - fixed linkage error when building with
466 LDFLAGS="-Wl,--no-undefined" (#190)
467 - fixed compilation with Clang 3.2
468 - removed usage of deprecated Automake variable INCLUDES
469 - fixed building with C++11
470 - build fix: ChangeFlagRelaxed.h was missing in makefile
471 - build fix: libsndfile compiler flags were missing in some
472 makefiles
473 - fix for building with bison 3.0 (#202)
474 - Mac OS X: added temporary hack allowing to spawn gigedit as callback
475 on the process's main thread
476 - fixed build error on newer MinGW
477 - support building with older jack versions
478 - support building with spaces in vst sdk path
479 - enabled automake 'subdir-objects' option and moved external
480 source references (vst, au, asio) from makefiles to cpp files,
481 in order to get rid of warnings from automake 1.14
482
483 * general changes:
484 - Refactoring: moved the independent code from
485 the Gigasampler format engine to base classes
486 - command line option '--profile' is currently disabled, since the
487 respective profiling code is currently broken
488 - Introduced support for optional environment variable
489 "LINUXSAMPLER_PLUGIN_DIR" which allows to override the directory
490 where the sampler shall look for instrument editor plugins
491 (patch by Luis Garrido, slightly modified).
492 - implemented Roland GS NRPN 1ArrH which allows to set volume per note
493 - implemented Roland GS NRPN 1CrrH which allows to set pan per note
494 - implemented Roland GS NRPN 1DrrH which allows to set reverb send per
495 note (in this implementation of the sampler its simply hard routed to
496 the 1st effect send of the sampler channel, no matter what the actual
497 effect type is)
498 - implemented Roland GS NRPN 1ErrH which allows to set chorus send per
499 note (in this implementation of the sampler its simply hard routed to
500 the 2nd effect send of the sampler channel, no matter what the actual
501 effect type is)
502 - implemented support for internal LADSPA effects
503 - raised limit of program change queue from 100 to 512
504 (as suggested by Alex Stone)
505 - implemented sine LFO, pulse LFO and saw LFO
506 - experimental support for per voice equalization
507 - added command line option --exec-after-init
508 - Introduced new C++ API method:
509 EngineChannel::InstrumentFileName(int index)
510 allowing to retrieve the whole list of files used for the loaded
511 instrument on an engine channel (a.k.a. part). Some GigaStudio
512 instruments for example are splitted over several files like
513 "Foo.gig", "Foo.gx01", "Foo.gx02", ...
514 - Added new C++ API method Sampler::GetGlobalMaxVoices().
515 - Added new C++ API method Sampler::GetGlobalMaxStreams().
516 - Added new C++ API method Sampler::SetGlobalMaxVoices().
517 - Added new C++ API method Sampler::SetGlobalMaxStreams().
518 - Various "const" and "restrict" optimizations.
519 - all engines: add pan CC value to instrument pan parameter before
520 applying panning, instead of using two separate pan functions in
521 series (#182)
522 - added a lock guard class for exception safe mutex handling and
523 used it everywhere appropriate
524 - Immediately apply scale tuning changes to active voices.
525 - Exposed scale tuning to C++ API (along to the already existing standard
526 SysEx way).
527 - lsatomic.h fixes: seq_cst load and store were suboptimal for x86
528 and broken for ppc64. (Seq_cst loads and stores are actually not
529 used in LS, so the bug wasn't noticable.)
530 - lsatomic.h: added ARMv7 support
531 - Added support for multiple MIDI input ports per sampler channel (and
532 added various new C++ methods for this new feature / design change, old
533 C++ API methods for managing SamplerChannel's MIDI inputs are now marked
534 as deprecated but are still there and should provide full behavior
535 backward compatibility).
536 - AbstractEngine::GSChecksum(): don't allocate memory on the stack (was
537 unsafe and caused compilation error with clang 2.x).
538 - Bugfix: only process the latest MIDI program change event.
539 - Introducing the LSCP shell, which provides convenient control of the
540 sampler from the command line by providing LSCP aware features.
541 - VirtualMidiDevice: Added support for program change.
542 - VirtualMidiDevice: Added support for bank select (MSB & LSB).
543 - VirtualMidiDevice: Added support for pitch bend.
544 - Aftertouch: extended API to explicitly handle channel pressure and
545 polyphonic key pressure events (so far polyphonic pressure was not
546 supported at all, and channel pressure was rerouted as CC128 but not
547 used so far).
548 - Added initial support for real-time instrument scripts. The script VM
549 code is shared by all sampler engine implemementations, however only the
550 gig file format currently provides support for storing instrument scripts
551 (as LinuxSampler extension to the original GigaStudio 4 format).
552
553 * Gigasampler format engine:
554 - implemented the "round robin keyboard" dimension
555 - fixed round robin and random dimensions for cases when number of
556 dimension zones is not a power of two
557 - made round robin use a counter for each region instead of each
558 key
559 - bugfix: pitch LFO controller "internal+aftertouch" was broken
560 - bugfix: filter keyboard tracking was broken
561 - filter performance fix (an unnecessary copy was made of the
562 filter parameters in each sub fragment)
563 - handle special case when pan parameter in gig file has max or
564 min value
565 - Exclusive Groups: don't ever stop voices of the same note,
566 doesn't sound naturally with a drumkit
567 - fixed EG1 modulation when attack or release is zero
568 - Fixed support for 'aftertouch' attenuation controller.
569 - Fixed crash that happened with velocity split sounds under certain
570 conditions (see also previous commit on libgig).
571 - fixed behaviour of filter LFO
572
573 * SFZ format engine:
574 - Initial implementation (not usable yet)
575 - added support for v2 multiple stage envelope generators
576 - added a fine-tuned v1 envelope generator instead of using the
577 one from the gig engine
578 - fixed memory leak and memory handling errors
579 - added support for trigger=first, trigger=legato and sw_previous
580 - allow non-numerical key values ("C#4" for example)
581 - "key" opcode now sets pitch_keycenter too
582 - fixed error when unloading instrument with same sample used by
583 multiple regions
584 - added some opcode aliases, like loopmode for loop_mode, to be
585 more compatible
586 - added support for trigger=release and rt_decay
587 - added support for off_mode=normal
588 - added support for random, seq_position, seq_length and volume
589 - added v1 LFO opcodes to parser (no support in engine yet)
590 - added support for amp_veltrack and amp_velcurve_N
591 - fine-tuned the default velocity curve
592 - added support for transpose
593 - fixed crash when using sw_down/up
594 - improved logic for sw_lokey/hikey/up/down/last
595 - added more v1 aliases to parser
596 - reduced memory usage for sfz data
597 - RT-safeness: avoid malloc in audio thread
598 - fixed a bug that could cause voice stealing to fail
599 - optimized sample lookup
600 - improved support for exclusive groups (group, off_by and
601 off_mode)
602 - added support for controller triggered regions (on_locc/on_hicc)
603 - added support for loop_mode=one_shot
604 - fixed playback of 16 bit wav files on big endian CPUs
605 - added support for Ogg Vorbis sample files
606 - use loop markers from sample file if loop_start and loop_end are
607 not set in sfz file
608 - implemeted filters. Filter types: lowpass, bandpass, bandreject
609 and highpass. 1, 2, 4 and 6 pole filters. Opcodes: fil_type,
610 cutoff, resonance, fil_veltrack, fil_keytrack, fil_keycenter,
611 cutoff_cc, cutoff_chanaft.
612 - bugfix: zero ampeg_sustain didn't work
613 - added support for velocity effect on amplifier envelope time
614 (ampeg_vel2attack, ampeg_vel2decay, ampeg_vel2sustain and
615 ampeg_vel2release)
616 - added support for EG hold (ampeg_hold)
617 - added support for sample offset (offset)
618 - use the newly introduced signal units model
619 - implemented opcodes lfoN_delay,
620 lfoN_freq, lfoN_pan, lfoN_cutoff, lfoN_resonance
621 - implemented opcodes ampeg_delay, ampeg_vel2delay
622 pitcheg_delay, pitcheg_start, pitcheg_attack, pitcheg_hold,
623 pitcheg_decay, pitcheg_sustain, pitcheg_release, pitcheg_vel2delay,
624 pitcheg_vel2attack, pitcheg_vel2hold, pitcheg_vel2decay,
625 pitcheg_vel2sustain, pitcheg_vel2release, pitcheg_depth
626 - implemented opcodes pitchlfo_delay, pitchlfo_freq,
627 pitchlfo_depth, fillfo_delay, fillfo_freq, fillfo_depth,
628 amplfo_delay, amplfo_freq, amplfo_depth
629 - implemented opcodes fileg_delay, fileg_start, fileg_attack,
630 fileg_hold, fileg_decay, fileg_sustain, fileg_release, fileg_vel2delay,
631 fileg_vel2attack, fileg_vel2hold, fileg_vel2decay, fileg_vel2sustain,
632 fileg_vel2release
633 - implemented opcode lfoN_wave
634 - implemented opcode pitchlfo_depthccN
635 - implemented opcodes lfoN_phase, lfoN_phase_onccX,
636 lfoN_pitch, lfoN_pitch_onccX
637 - implemented opcodes pitchlfo_fade,
638 fillfo_fade, amplfo_fade, lfoN_fade, lfoN_fade_onccX
639 - implemented opcodes fillfo_freqccN,
640 pitchlfo_freqccN, amplfo_freqccN, lfoN_freq_onccX
641 - implemented opcodes ampeg_delayccN, ampeg_startccN,
642 ampeg_attackccN, ampeg_holdccN, ampeg_decayccN, ampeg_sustainccN,
643 ampeg_releaseccN, egN_timeX_onccY, egN_levelX_onccY
644 - lfoN_* and egN_* opcodes defined in group sections
645 are now taken into account
646 - implemented curves
647 - implemented opcodes volume_onccN, volume_curveccN
648 - implemented opcode volume_smoothccN
649 - implemented opcodes fillfo_depthccN, amplfo_depthccN,
650 lfoN_volume, lfoN_volume_onccX, lfoN_volume_smoothccX,
651 lfoN_freq_smoothccX, lfoN_pitch_smoothccX, lfoN_pan_onccX,
652 lfoN_pan_smoothccX, lfoN_cutoff_onccX, lfoN_cutoff_smoothccX,
653 lfoN_resonance_onccX, lfoN_resonance_smoothccX, lfoN_delay_onccX
654 - implemented opcode end
655 - implemented opcodes egN_amplitude_onccX,
656 egN_volume, egN_volume_onccX, egN_cutoff, egN_cutoff_onccX,
657 egN_pitch, egN_pitch_onccX, egN_resonance, egN_resonance_onccX
658 - implemented opcodes xfin_lokey, xfin_hikey,
659 xfout_lokey, xfout_hikey, xf_keycurve, xfin_lovel, xfin_hivel,
660 xfout_lovel, xfout_hivel, xf_velcurve, xfin_loccN, xfin_hiccN,
661 xfout_loccN, xfout_hiccN, xf_cccurve
662 - implemented opcodes pan_onccN, pan_smoothccN,
663 pan_curveccN, egN_pan, egN_pan_curve, egN_pan_onccX,
664 egN_pan_curveccX, pitch_veltrack
665 - when failed to parse a sfz file
666 print the line number on which the error occurs
667 - use common pool of CC objects to minimize RAM usage
668 - implemented opcodes amplfo_delay_onccN,
669 amplfo_fade_onccN, fillfo_delay_onccN, fillfo_fade_onccN,
670 pitchlfo_delay_onccN, pitchlfo_fade_onccN
671 - implemented opcodes fileg_delay_onccN,
672 fileg_start_onccN, fileg_attack_onccN, fileg_hold_onccN,
673 fileg_decay_onccN, fileg_sustain_onccN, fileg_release_onccN,
674 fileg_depth_onccN, pitcheg_delay_onccN, pitcheg_start_onccN,
675 pitcheg_attack_onccN, pitcheg_hold_onccN, pitcheg_decay_onccN,
676 pitcheg_sustain_onccN, pitcheg_release_onccN, pitcheg_depth_onccN
677 - implemented automatic aliasing *ccN <-> *_onccN
678 - *lfo_freqccN wasn't working when the respective *lfo_freq
679 was not set or was set to zero
680 - lfoN_freq_onccX wasn't working when lfoN_freq
681 was not set or was set to zero
682 - implemented opcodes resonance_onccN, resonance_smoothccN,
683 resonance_curveccN, cutoff_smoothccN, cutoff_curveccN
684 - implemented opcodes fillfo_depthchanaft,
685 fillfo_freqchanaft, amplfo_depthchanaft, amplfo_freqchanaft,
686 pitchlfo_depthchanaft, pitchlfo_freqchanaft
687 - implemented opcodes pitch_onccN,
688 pitch_curveccN, pitch_smoothccN, pitch_stepccN
689 - implemented opcodes volume_stepccN, pan_stepccN,
690 cutoff_stepccN, resonance_stepccN, lfoN_freq_stepccX,
691 lfoN_volume_stepccX, lfoN_pitch_stepccX, lfoN_pan_stepccX,
692 lfoN_cutoff_stepccX, lfoN_resonance_stepccX
693 - implemented opcodes eq1_freq, eq2_freq, eq3_freq,
694 eq1_freqccN, eq2_freqccN, eq3_freqccN, eq1_bw, eq2_bw, eq3_bw,
695 eq1_bwccN, eq2_bwccN, eq3_bwccN, eq1_gain, eq2_gain, eq3_gain,
696 eq1_gainccN, eq2_gainccN, eq3_gainccN
697 - implemented opcodes delay, delay_onccN, delay_random,
698 delay_samples, delay_samples_onccN
699 - implemented opcodes egN_eq1gain, egN_eq2gain, egN_eq3gain,
700 egN_eq1gain_onccX, egN_eq2gain_onccX, egN_eq3gain_onccX, egN_eq1freq,
701 egN_eq2freq, egN_eq3freq, egN_eq1freq_onccX, egN_eq2freq_onccX,
702 egN_eq3freq_onccX, egN_eq1bw, egN_eq2bw, egN_eq3bw, egN_eq1bw_onccX,
703 egN_eq2bw_onccX, egN_eq3bw_onccX, lfoN_eq1gain, lfoN_eq2gain,
704 lfoN_eq3gain, lfoN_eq1gain_onccX, lfoN_eq2gain_onccX, lfoN_eq3gain_onccX,
705 lfoN_eq1gain_smoothccX, lfoN_eq2gain_smoothccX, lfoN_eq3gain_smoothccX,
706 lfoN_eq1gain_stepccX, lfoN_eq2gain_stepccX, lfoN_eq3gain_stepccX,
707 lfoN_eq1freq, lfoN_eq2freq, lfoN_eq3freq, lfoN_eq1freq_onccX,
708 lfoN_eq2freq_onccX, lfoN_eq3freq_onccX, lfoN_eq1freq_smoothccX,
709 lfoN_eq2freq_smoothccX, lfoN_eq3freq_smoothccX, lfoN_eq1freq_stepccX,
710 lfoN_eq2freq_stepccX, lfoN_eq3freq_stepccX, lfoN_eq1bw, lfoN_eq2bw,
711 lfoN_eq3bw, lfoN_eq1bw_onccX, lfoN_eq2bw_onccX, lfoN_eq3bw_onccX,
712 lfoN_eq1bw_smoothccX, lfoN_eq2bw_smoothccX, lfoN_eq3bw_smoothccX,
713 lfoN_eq1bw_stepccX, lfoN_eq2bw_stepccX, lfoN_eq3bw_stepccX
714 - implemented opcodes eq1_vel2freq, eq2_vel2freq,
715 eq3_vel2freq, eq1_vel2gain, eq2_vel2gain, eq3_vel2gain
716 - sfz parser: allow double spaces in sample filenames
717 - sfz parser: allow absolute paths for sample filenames
718 - use linear decay and release for filter and pitch EG
719 - bugfix: only the first amp_veltrack definition in a file was
720 used
721 - bugfix: looping was disabled if loop_start was set to 0
722 - allow regions with end=-1 to turn off other regions using the
723 group and off_by opcodes (#168)
724 - made end=0 play the whole sample
725 - fixed support for lochan and hichan opcodes (#155)
726 - fixed crash when using lochan/hichan opcodes (#187)
727 - sfz parser: allow -200 to 200 for pan_oncc opcode (#182)
728 - added FLAC support (#191)
729 - sfz parser bugfix: lines starting with whitespace were ignored
730 - added amplitude opcode
731 - added support for "#include" instruction
732 (modified patch which was originally posted by Sergey on LS mailing list)
733 - bugfix: generation of velocity curves etc should not be done
734 after each "#include", only after the main file is parsed
735 - bugfix: line numbers in error messages were wrong after "#include"
736 - added support for float and 32 bit sample files
737
738 * SoundFont format engine:
739 - Initial implementation (not usable yet)
740 - RT-safeness: avoid malloc in audio thread
741 - fixed a bug that could cause voice stealing to fail
742 - fine-tuned amplitude EG (by switching from gig to sfz EG)
743 - initial implementation of Vibrato LFO and Modulation LFO
744 - initial implementation of cutoff filter
745 - use linear decay and release for filter and pitch EG
746
747 * Host plugins (VST, AU, LV2, DSSI):
748 - AU bugfix: failed to destroy its audio/MIDI devices
749 - Listen to all interfaces on Mac OS X (INADDR_ANY)
750 - VST bugfix: If the host called resume() before and after
751 changing sample rate or block size, the number of channels was
752 incorrectly set to two. This caused silence in Cubase 5.
753 - save engine type (gig, sfz or sf2) in plugin state
754 - VST: when opening Fantasia, look for both 32 and 64 bit Java on
755 64 bit Windows
756 - AU: changed number of output channels from one stereo to 16
757 stereo
758 - VST: made it possible to build the VST plugin for Mac
759 - AU: link AU plugin dynamically if --disable-shared isn't
760 specified
761 - LV2 "state" extension support (patch by David Robillard)
762 - VST bugfix: instrument loading hang and crashed the host when
763 the plugin was loaded a second time (#174)
764 - plugin bugfix: instrument loading hang when the plugin was
765 loaded a second time (this time it's for Linux and Mac, previous
766 similar fix was for Windows)
767 - thread safety fixes for the instrument loading thread
768 - LV2: use the new lv2 package if present
769 - VST: try to open Fantasia automatically on Linux and Mac too (on
770 Linux, the Fantasia jar should be placed in <prefix>/share/java)
771 - VST: fixed crashes on Linux Ardour and EnergyXT
772 - DSSI bugfix: it wasn't possible to change engine type. The MIDI
773 port and audio channel routing for DSSI plugins are now visible.
774 - LV2: use urid and atom extensions instead of deprecated uri-map
775 and event
776 - LV2: lv2 package 1.0 is now required to build the LV2 plugin
777 - LV2: changed number of output channels to 16 stereo, just like
778 the VST and AU plugins
779 - LV2: fixed save/restore of SFZ state (patch by David Robillard)
780 - LV2: made LV2 plugin buildable on Windows and Mac
781 - VST: implemented retrieval and switching of programs using the
782 sampler's internal MIDI instrument mapping system
783
784 * Instrument editor interface:
785 - Changed instrument editor plugin interface, providing additional
786 informations like the EngineChannel for which the instrument editor was
787 spawned for. This allows the instrument editors to interact more actively
788 with the sampler.
789
790 * MIDI driver:
791 - ALSA MIDI driver supports now "NAME" device parameter, for overriding
792 the ALSA sequencer client name
793 - removed limit of maximum amount of MIDI ports per MIDI device, since
794 there is no reason for this limit
795 - MME: fixed memory handling bug found with cppcheck
796 - MME: removed compiler warning
797 - CoreMIDI: implemented driver specific port parameter "CORE_MIDI_BINDINGS",
798 which allows to retrieve the list of CoreMIDI clients / ports and to connect
799 to them a la JACK, via the usual sampler APIs
800 - CoreMIDI: added driver specific port parameter "AUTO_BIND", if enabled
801 the driver will automatically connect to other CoreMIDI clients' ports
802 (e.g. external MIDI devices being attached to the Mac)
803 - added support for MIDI note on velocity filter
804 - CoreMIDI: fixed memory deallocation error
805 - Fixed variable underflow in VirtualMidiDevice, which caused graphical
806 virtual keyboards in frontends / instrument editors being stuck.
807 - Bugfix in VirtualMidiDevice: process note on with velocity 0 as note off.
808 - Implemented missing handling of MIDI "running status".
809 - CoreMIDI fix: a MIDIPacket can contain more than one event per packet.
810 - MME bugfix: driver wasn't closed properly
811 - CoreMIDI: automatically connect to all input sources by default (driver
812 parameter "AUTO_BIND").
813 - CoreMIDI: fixed auto bind feature to CoreMIDI ports that go online
814 - CoreMIDI: fixed minor memory leak
815
816 * audio driver:
817 - ASIO driver fixes for newer gcc versions (fix from PortAudio)
818 - JACK audio: react on sample rate changes.
819 - JACK audio: react on buffer size changes.
820 - JACK audio: jack_port_get_buffer() was cached and called outside
821 RT context.
822 - ASIO driver: removed compiler warnings
823 - CoreAudio: fixed minor error handling bug
824 - ASIO driver: be more verbose when no ASIO card could be found (fixes #203)
825 - JACK audio: return the JACK server's current sample rate as default value
826 for audio device parameter "SAMPLERATE" (fixes #166).
827
828 * LSCP server:
829 - added support for sending MIDI CC messages via LSCP command
830 "SEND CHANNEL MIDI_DATA CC <sampler-chan> <ctrl> <val>"
831 - added LSCP command "GET AVAILABLE_EFFECTS"
832 - added LSCP command "LIST AVAILABLE_EFFECTS"
833 - added LSCP command "GET EFFECT INFO <effect-index>"
834 - added LSCP command "CREATE EFFECT_INSTANCE <effect-index>"
835 - added LSCP command
836 "CREATE EFFECT_INSTANCE <effect-system> <module> <effect-name>"
837 - added LSCP command "DESTROY EFFECT_INSTANCE <effect-instance>"
838 - added LSCP command "GET EFFECT_INSTANCES"
839 - added LSCP command "LIST EFFECT_INSTANCES"
840 - added LSCP command "GET EFFECT_INSTANCE INFO <effect-instance>"
841 - added LSCP command
842 "GET EFFECT_INSTANCE_INPUT_CONTROL INFO <effect-instance> <input-control>"
843 - added LSCP command "SET EFFECT_INSTANCE_INPUT_CONTROL VALUE
844 <effect-instance> <input-control> <value>"
845 - added LSCP command "GET SEND_EFFECT_CHAINS <audio-device>"
846 - added LSCP command "LIST SEND_EFFECT_CHAINS <audio-device>"
847 - added LSCP command "ADD SEND_EFFECT_CHAIN <audio-device>"
848 - added LSCP command
849 "REMOVE SEND_EFFECT_CHAIN <audio-device> <effect-chain>"
850 - added LSCP command
851 "GET SEND_EFFECT_CHAIN INFO <audio-device> <effect-chain>"
852 - added LSCP command "APPEND SEND_EFFECT_CHAIN EFFECT <audio-device>
853 <effect-chain> <effect-instance>"
854 - added LSCP command "INSERT SEND_EFFECT_CHAIN EFFECT <audio-device>
855 <effect-chain> <effect-chain-pos> <effect-instance>"
856 - added LSCP command "REMOVE SEND_EFFECT_CHAIN EFFECT <audio-device>
857 <effect-chain> <chain-pos>"
858 - added LSCP command "SET FX_SEND EFFECT <sampler_channel>
859 <fx_send_id> <effect_chain> <chain_pos>"
860 - added LSCP command "REMOVE FX_SEND EFFECT <sampler_channel> <fx_send_id>"
861 - added LSCP commands "SUBSCRIBE EFFECT_INSTANCE_COUNT",
862 "SUBSCRIBE EFFECT_INSTANCE_INFO", "SUBSCRIBE SEND_EFFECT_CHAIN_COUNT",
863 "SUBSCRIBE SEND_EFFECT_CHAIN_INFO"
864 - provide comprehensive error messages on LSCP syntax errors
865 (suggesting expected next non-terminal symbols)
866 - Fixed client connection not being closed after network errors.
867
868 * LSCP shell:
869 - Added support for auto correction of obvious and trivial syntax mistakes.
870 - Added support for auto completion by tab key.
871 - Show currently available auto completion while typing.
872 - Added support for browsing command history with up / down keys.
873 - Show all possible next symbols immediately right to the current command
874 line while typing (no double tab required for this feature, as it would
875 be the case in other shells).
876 - Added support for moving cursor left/right with arrow keys.
877 - Added support for built-in LSCP reference documentation, which will
878 automatically show the relevant LSCP reference section on screen as soon
879 as one specific LSCP command was detected while typing on the command
880 line.
881
882 * Real-time instrument scripts:
883 - Implemented built-in script array variable %CC.
884 - Implemented built-in script int variable $CC_NUM.
885 - Implemented built-in script int variable $EVENT_NOTE.
886 - Implemented built-in script int variable $EVENT_VELOCITY.
887 - Implemented built-in script constant variable $VCC_MONO_AT.
888 - Implemented built-in script constant variable $VCC_PITCH_BEND.
889 - Implemented execution of script event handler "init".
890 - Implemented execution of script event handler "controller".
891 - Implemented execution of script event handler "note".
892 - Implemented execution of script event handler "release".
893 - Implemented built-in script function "play_note()" (only two of the
894 max. four function arguments are currently implemented yet though).
895 - Implemented built-in script int variable $EVENT_ID.
896 - Implemented built-in script function "ignore_event()"
897 - Implemented built-in script function "ignore_controller()" (may have one
898 or no argument).
899 - Implemented built-in script function "set_controller()".
900 - Added extended script VM for the Gigasampler/GigaStudio format sampler
901 engine, which extends the general instrument script VM with Giga format
902 specific variables and functions.
903 - Giga format scripts: added built-in script int constant variables
904 $GIG_DIM_CHANNEL, $GIG_DIM_LAYER, $GIG_DIM_VELOCITY, $GIG_DIM_AFTERTOUCH,
905 $GIG_DIM_RELEASE, $GIG_DIM_KEYBOARD, $GIG_DIM_ROUNDROBIN, $GIG_DIM_RANDOM,
906 $GIG_DIM_SMARTMIDI, $GIG_DIM_ROUNDROBINKEY, $GIG_DIM_MODWHEEL,
907 $GIG_DIM_BREATH, $GIG_DIM_FOOT, $GIG_DIM_PORTAMENTOTIME, $GIG_DIM_EFFECT1,
908 $GIG_DIM_EFFECT2, $GIG_DIM_GENPURPOSE1, $GIG_DIM_GENPURPOSE2,
909 $GIG_DIM_GENPURPOSE3, $GIG_DIM_GENPURPOSE4, $GIG_DIM_SUSTAIN,
910 $GIG_DIM_PORTAMENTO, $GIG_DIM_SOSTENUTO, $GIG_DIM_SOFT,
911 $GIG_DIM_GENPURPOSE5, $GIG_DIM_GENPURPOSE6, $GIG_DIM_GENPURPOSE7,
912 $GIG_DIM_GENPURPOSE8, $GIG_DIM_EFFECT1DEPTH, $GIG_DIM_EFFECT2DEPTH,
913 $GIG_DIM_EFFECT3DEPTH, $GIG_DIM_EFFECT4DEPTH, $GIG_DIM_EFFECT5DEPTH.
914 - Giga format scripts: Implemented built-in script function
915 "gig_set_dim_zone(event_id, dimension, zone)".
916 - Implemented built-in script int array variable %KEY_DOWN.
917 - Implemented built-in script function "abs()".
918 - Implemented built-in script function "random()".
919 - Implemented built-in script function "num_elements()".
920 - Implemented built-in script function "note_off()".
921 - Implemented built-in script function "set_event_mark()".
922 - Implemented built-in script function "delete_event_mark()".
923 - Implemented built-in script function "by_marks()".
924 - Added built-in script int const variables $MARK_1 to $MARK_28.
925 - Built-in script functions "ignore_event()", "note_off()" and
926 "gig_set_dim_zone()" now also accept an array of event IDs as argument
927 (i.e. return value of new script function "by_marks()").
928 - Pass/preserve polyphonic variable data from respective "note" event
929 handler to "release" event handler.
930 - Fixed crash when using built-in script function "by_marks()".
931
932 * Bug fixes:
933 - Fixed crash which may occur when MIDI key + transpose is out of range
934 - minor valgrind fixes
935 - fixed crash which occurred when changing an already deployed sampler
936 channel to a different engine type
937 - fixed crash when deleting a sampler channel or changing engine
938 type while an instrument load was in progress
939 - bugfix: playing a note while changing the instrument could cause
940 a crash, or give "Handing back unknown region" error messages
941 - bugfix: calling SET VOICES (which Fantasia does on start and
942 refresh) could cause instruments to be unloaded
943 - Mac OS X: fixed name collision of enum in EffectControl and
944 wrong name of destructor in AudioOutputDeviceCoreAudio.cpp
945 - Mac OS X: fixed hanging threads
946 - Windows: disabled the previous thread fix on non-Mac systems, as
947 it caused hanging threads on Windows
948 - Fixed possible crashes due to corrupted MIDI/audio device list
949 after MIDI/audio device creation failure
950 - When creating MIDI instrument map entries with "PERSISTENT" type, the
951 instruments were uselessly precached with zero samples, however it still
952 took the full preloading time and on 1st program change the respective
953 instrument was completely reloaded again.
954 - fixed handling of rapid bank select and program change messages
955 sent to the same sampler channel (patch from the Open Octave
956 project, slightly adjusted)
957 - fixed crash when trying to create an effect instance with controls
958 which min and/or max values depend on the sample rate
959 - fixed bug #162
960 - bugfix: LADSPA_PATH was not evaluated correctly when containing
961 multiple paths (#165)
962 - thread safety fixes for the instrument loading thread
963 - bugfix: instrument loading crashed for sfz and sf2 in Ardour
964 (#176)
965 - more thread safety fixes for the instrument loading thread
966 - sfz/sf2 engine: fixed crash when using small audio fragment size
967 - Mac OS X: fixed crash when unloading plugin on 10.7 and later
968 - Mac OS X: fixed process hang when unloading 32-bit plugin (bug
969 introduced in previous fix)
970 - fixed crash when a channel received a program change while
971 playing a note in a key group
972 - fixed erroneous error message piping in VoiceBase.h
973 (labelled "Disk stream not available in time")
974 - Update effects on sample rate & period size changes (to avoid
975 crashes and noise on such transitions).
976 - ignore missing LADSPA paths without ignoring valid LADSPA paths
977 (fixes #208)
978 - Fixed bug in LSCP grammar definition which caused a statement like
979 "GET SERVER INFOasdf\n" to be accepted as valid statement (was so far
980 practically irrelevant, however it caused problems with the new LSCP
981 shell's auto completion feature).
982 - Fixed MIDI program change messages being ignored if quickly executed
983 after each other (fixes #231).
984
985 Version 1.0.0 (31 July 2009)
986
987 * packaging changes:
988 - autoconf bugfix: the PKG_CONFIG variable wasn't initialized properly,
989 causing e.g. the libgig test to fail when
990 "./configure --disable-jack-driver" was used
991 (patch by Alexis Ballier)
992 - fixed compilation with gcc 4.3
993 - fixes for building on OS X (thanks to Ebrahim Mayat for testing)
994 - fixed configure so it detects x86_64 (#107)
995 - fixes for building with newer MinGW versions
996 - fix for building with bison 2.4 (#111)
997 - fixed building with libgig installed in a non-standard directory
998 - minor fix in configure for mmsystem.h detection on MinGW
999 - Windows: look for editor plugins and Fantasia using base
1000 directory of liblinuxsampler dll (look in the same directory and one
1001 directory above)
1002 - configure script fix: removed unconditional use of SSE
1003 - fixed building with sqlite installed in a non-standard directory
1004 - when cross-compiling, don't try to create instruments.db
1005 - fix for new mingw-w64 version, which has usleep
1006
1007 * general changes:
1008 - bugfix: on some POSIX systems instrument editor plugins refused to
1009 load as we used a non-portable Linux specific struct field
1010 (fixes bug #70, patch by Ronald Baljeu)
1011 - fixed endless loop which occured when loading an instrument editor
1012 plugin DLL on Windows systems failed
1013 - fixed memory leaks that occurred when liblinuxsampler was unloaded
1014 - optimized the SynchronizedConfig class so it doesn't wait
1015 unnecessarily long after an update
1016 - added support for notifying instrument editors on note-on / note-off
1017 events (e.g. to highlight the pressed keys on the virtual keyboard
1018 of gigedit)
1019 - added support for triggering notes by instrument editors (see above)
1020 - be verbose on DLL load errors (on Linux)
1021 - fixed di-harmonic triangle LFO implementation (this LFO implementation
1022 is only used by very few systems, most pick the int math
1023 implementation, which is usually faster)
1024 - fixes for audio drivers with varying buffer sizes
1025 - experimental support for running LinuxSampler as a DSSI, LV2 and
1026 VST plugin
1027 - notification events for stream/voice count statistics are now sent
1028 only when there are actual changes
1029 - added memory ordering constraints to improve stability on
1030 multi-core and multi-cpu systems
1031 - maximum voices and disk streams can now be altered at runtime
1032 - fixed CPU feature detection on x86_64 (maybe fixes #108)
1033 - automatic stacktrace mechanism is now turned off by default and can
1034 be switched on by command line option "--stacktrace" (the automatic
1035 stacktrace mechanism seems to be broken on most systems at the moment)
1036 - C++ API method InstrumentManager::LaunchInstrumentEditor() now returns
1037 a pointer to the launched InstrumentEditor object
1038 - added optional 3rd party user data parameter for following
1039 liblinuxsampler C++ API methods: InstrumentEditor::Main(),
1040 InstrumentEditor::Launch(),
1041 InstrumentManager::LaunchInstrumentEditor()
1042 - theoretical fix: made SynchronizedConfig follow C++0x memory
1043 model more strictly
1044 - fixes for using large audio device buffers
1045 - Windows: add the installation directory to the DLL search path
1046 when loading an editor plugin (solves problems with VST and
1047 gigedit on systems with other GTK versions installed)
1048 - updated linuxsampler man page
1049
1050 * audio driver:
1051 - removed the nonsense audio channel constraint (which was hard coded to
1052 max. 100 audio channels) for most audio drivers
1053 - JACK audio driver did not offer a device parameter "SAMPLERATE" as
1054 opposed to the LSCP specs
1055 - bugfix: the SAMPLERATE parameter of some drivers (e.g. JACK)
1056 reflected the wrong value
1057 - fixed a memory management bug in ASIO driver
1058 - Makefile fix: JACK_CFLAGS wasn't used
1059 - JACK: use jack_client_open instead of the deprecated
1060 jack_client_new
1061 - added (experimental) CoreAudio driver
1062 - applied old fixes to the ASIO driver that were included in the
1063 previous binary release but accidentally never committed to CVS
1064 (fixes #117)
1065 - fixes for ASIO on mingw-w64 (iasio wrapper is not needed on
1066 win64)
1067 - VST: added support for sample rate and buffer size changes
1068 - VST: close editor (Fantasia) when the VST is removed
1069 - VST: avoid opening Fantasia more than once for each VST instance
1070 - VST: export main function as "main" on Linux too (fix for
1071 energyXT)
1072 - VST: changed number of output channels from one stereo to 16
1073 stereo
1074 - added channel routing, fxsends and midi maps to the settings
1075 stored in the plugin state
1076 - performance optimization of AudioChannel::MixTo() and
1077 AudioChannel::CopyTo() methods using GCC vector exensions
1078 (if available)
1079 - ASIO fixes: avoid initializing the device twice, avoid throwing
1080 exception when getting parameters from a disconnected device
1081
1082 * MIDI driver:
1083 - added JACK MIDI driver
1084 - dispatch bank select as ordinary CC as well, the user might seriously
1085 want to (mis)use it for some purpose ("fixed" in all current MIDI
1086 input drivers: ALSA, CoreMIDI, JACK, MidiShare, MME)
1087 - bugfix: pitch bend wasn't working with JackMidi, VST, LV2, MME,
1088 CoreMidi or AU
1089 - fixed mingw-w64 compilation error in MME driver
1090 - made program change handling in MIDI thread real-time safe by
1091 moving the logic to a non-RT thread
1092 - fixed minor memory leak in ALSA driver
1093
1094 * instruments database:
1095 - avoid time consuming samples scanning when adding instruments
1096 to the instruments database
1097 - added support for handling lost files in the instruments database
1098 - Implemented option for adding instruments in separate directories
1099 in the instruments database
1100 (patch by Chris Cherrett & Andrew Williams, a bit adjusted)
1101 - work-around for missing fnmatch function on Windows to make
1102 instrument database compilable
1103 - added instrument database support on Windows
1104 works with both standalone mode and VST plugin.
1105 the instruments DB file is located in
1106 %USERPROFILE%\.linuxsampler\instruments.db which allows different
1107 databases for each windows user
1108 if no DB is present it automatically creates the .linuxsampler subdir
1109 and then creates an empty DB
1110 - fixed recursive import, which was broken on Windows
1111
1112 * Gigasampler format engine:
1113 - fixed a memory leak that could happen when a channel was deleted
1114 while notes were playing
1115 - made it possible to load an instrument even if the audio thread
1116 isn't running
1117 - added partial support for the "Controller Triggered" MIDI rule,
1118 enough for piano gigs with pedal noise samples
1119 - minor fix: only mark FX sends as being modified if really the
1120 respective FX send MIDI controller was used
1121 - added support for GM global device volume SysEx message
1122 (can be selected at compile time to either apply globally to the
1123 whole sampler [default] or only to the sampler channels that are
1124 connected to the respective MIDI input port the SysEx message arrived
1125 on)
1126 - bugfix: notes triggered at position 0 in the audio buffer were
1127 sometimes wrongly killed in the same buffer, causing no sound to
1128 be played
1129 - added support for chromatic / drumkit mode Roland GS Sysex message
1130 which is usally used to switch a part between chromatic sounds and
1131 drumkit sounds (as opposed to their common default setting of MIDI
1132 part 10 being a drumkit part and all other ones chromatic parts), in
1133 LS however one can switch between the first 16 MIDI instrument maps
1134 defined for the sampler
1135 - made it possible to create multiple sample channels even if the
1136 audio thread isn't running
1137 - fixed crash when removing channel with active voices (#116)
1138 - bugfix: on sample reference changes (instrument editor), only
1139 un-cache the respective sample if it's really not used by any
1140 sampler engine anymore
1141 - re-cache samples in case they were changed by an instrument editor,
1142 e.g. when a sample was added while playing (#82)
1143 - fixed hanging notes which occured when note-off event had the exact
1144 same time stamp as the voice's note-on event and occured both in the
1145 same audio fragment cycle (fixes bug #112)
1146 - added support for the "fine tune" and "pitch bend range"
1147 instrument-level gig parameters
1148 - fixed minor artifacts in pitch bend handling
1149 - added support for GS Reset SysEx message
1150 - allow gig files to use unlimited downward pitch shifting
1151 - added a limit check for upward pitch shifting
1152 - bugfix: sometimes, when playing a note twice fast, the second
1153 note was silent
1154 - fixed crash happening when a pitch bend event arrived at the
1155 same time a new instrument was loading
1156
1157 * LSCP server:
1158 - added new LSCP event "CHANNEL_MIDI" which can be used by frontends to
1159 react on MIDI data arriving on certain sampler channels (so far only
1160 Note-On and Note-Off events are sent via this LSCP event)
1161 - added new LSCP event "DEVICE_MIDI" which can be used by frontends to
1162 react on MIDI data arriving on certain MIDI input devices (so far only
1163 Note-On and Note-Off events are sent via this LSCP event)
1164 - added new LSCP commands: FIND LOST DB_INSTRUMENT_FILES and
1165 SET DB_INSTRUMENT FILE_PATH
1166 - added new LSCP command: SEND CHANNEL MIDI_DATA which can be used by
1167 frontends to send MIDI messages to specific sampler channel
1168 - added two additional fields to GET FILE INSTRUMENT INFO command -
1169 KEY_BINDINGS and KEYSWITCH_BINDINGS
1170 - bugfix: the bank number provided by MIDI_INSTRUMENT_INFO notifications
1171 was incorrect
1172 - Optimized the retrieval of the MIDI instrument mappings
1173 - added new LSCP command "SET VOICES" to globally alter the maximum
1174 amount of voices, added new LSCP event "GLOBAL_INFO:VOICES" which will
1175 be triggered respectively
1176 - added new LSCP command "SET STREAMS" to globally alter the maximum
1177 amount of disk streams, added new LSCP event "GLOBAL_INFO:STREAMS"
1178 which will be triggered respectively
1179 - bugfix: retry if "select" returns EINTR (this fixes a crash when
1180 a gigedit file dialog is opened)
1181 - close all connections when LSCPServer is deleted
1182 - hard close of all sockets on exit
1183 - bugfix: SET CHANNEL MIDI_INPUT_TYPE didn't work with the MME
1184 driver
1185
1186 * Bug fixes:
1187 - fixed a crash which occurs when removing a sampler channel waiting
1188 to start instrument loading after another channel
1189 - fixed a crash which occurs when removing a sampler channel with
1190 instrument loading in progress (bug #113)
1191 - fixed termination caused by uncaught exception when adding MIDI
1192 instrument with PERSISTENT load mode
1193 - fixed possible iterator invalidations when resetting the sampler
1194 - fixed memory leaks when issuing the following LSCP commands:
1195 GET AUDIO_OUTPUT_DRIVER INFO
1196 GET AUDIO_OUTPUT_DRIVER_PARAMETER INFO
1197 GET MIDI_INPUT_DRIVER INFO
1198 GET MIDI_INPUT_DRIVER_PARAMETER INFO
1199 - fixed possible compilation error when sqlite is not present
1200 - fixed orphaned pointers when setting maximum voices limit (bug #118)
1201 - fixed crash when changing the audio output device of a sampler
1202 channel with loaded instrument and start playing notes
1203 - fixed endless loop in Engine::SuspendAll() (bug #120)
1204 - fixed a low-level atomic load function that was broken on 64-bit
1205 PowerPC, which probably could cause crashes on that platform
1206 - fixed a memory management error which could cause a crash when a
1207 plugin was unloaded
1208 - bugfix: two private structs had the same name, which could cause
1209 problems if the linker chose the wrong constructor
1210 - fixed low-level ConditionServer usage bug that caused lockups on
1211 Windows
1212
1213
1214 Version 0.5.1 (6 December 2007)
1215
1216 * packaging changes:
1217 - added autoconf checks for pthread library
1218 - added autoconf check for pthread bug found on certain NPTL-enabled
1219 glibc versions (see Gentoo bug report #194076)
1220 - added autoconf checks for MS Windows
1221
1222 * general changes:
1223 - bugfix: the thread used by an editor plugin didn't die when the
1224 editor closed
1225 - bugfix: Ringbuffer.h: fill_write_space_with_null() did not zero
1226 out all the space. operator--() did not apply size_mask after
1227 decrementing the read_ptr. DEFAULT_WRAP_ELEMENTS set to 0 in
1228 order to avoid problems with the _NonVolatileReader functions.
1229 - bugfix: Stream.h: added a missing divide by BytesPerSample in
1230 GetWriteSpace(). Since this function is currently only used in
1231 the stream's qsort() compare function, it didn't trigger any
1232 bugs.
1233 - Resampler.h, Synthesizer.h: cubic interpolation now works in
1234 24bit mode too. Faster method to read 24bit words on little
1235 endian machines (x86): replaced 3 byte reads + shifts with a 1
1236 unaligned 32bit read and shift
1237 - experimental support for MS Windows (MIDI input via MME, AUDIO
1238 output via ASIO)
1239 - made handling of SIGINT signal (Ctrl-C) a bit more robust
1240 - support for monitoring the total number of active disk streams
1241 (new LSCP commands: GET TOTAL_STREAM_COUNT,
1242 SUBSCRIBE TOTAL_STREAM_COUNT, UNSUBSCRIBE TOTAL_STREAM_COUNT)
1243
1244 * AUDIO driver:
1245 - added Windows ASIO low latency audio driver
1246
1247 * MIDI driver:
1248 - added MME Windows MIDI driver
1249
1250 * LSCP server:
1251 - added support for Windows style path / filenames, however with
1252 forward slash path separators instead of backslash
1253 (i.e. "C:/foo/bar.gig")
1254 - allow naughty liblscp to send non-string device parameters within
1255 apostrophes as well
1256 - added new LSCP commands: "GET FILE INSTRUMENTS <file>",
1257 "LIST FILE INSTRUMENTS <file>" and
1258 "GET FILE INSTRUMENT INFO <file> <index>" for retrieving informations
1259 about an arbitrary instrument file on the system where the sampler is
1260 running on
1261
1262 Version 0.5.0 (15 October 2007)
1263
1264 * packaging changes:
1265 - config.h is not going to be installed along with liblinuxsampler's
1266 API header files anymore
1267 - only the API relevant header (and source) files will be exported to
1268 the Doxygen API documentation (explicitly listed in Doxyfile.in)
1269 - added completely new XCode project files for Mac OSX which is now
1270 capable to execute our autoconf environment, thus no need anymore
1271 to maintain the compile time configuration file (osx/version.h) for
1272 OSX manually (patch by Toshi Nagata)
1273 - fixed buggy boolean --enable-foo / --disable-foo configure script
1274 parameters
1275 - global.h now only covers global definitions that are needed for the
1276 C++ API header files, all implementation internal global definitions
1277 are now in global_private.h
1278 - atomic.h is not exposed to the C++ API anymore
1279 - no need to include config.h anymore for using LS's API header files
1280 - fixed warnings in API doc generation
1281
1282 * general changes:
1283 - replaced the old, confusing MIDI program change mechanism by a
1284 flexible MIDI instrument mapper which allows to map arbitrary
1285 (MIDI bank, MIDI program) pairs with arbitrary
1286 (engine type, instrument file, file index) triplets which will be
1287 loaded on the respective channel when such MIDI program change
1288 messages arrive, beside that, each entry allows to define a life-time
1289 strategy for the instrument, so the user can define whether the
1290 instrument should i.e. be permanently loaded (to be able to switch
1291 quickly among sounds) or i.e. loaded just on demand when the
1292 respective program change arrives (to safe RAM space), as well as a
1293 global volume factor for each entry, so the user can adjust the volume
1294 dynamic of his mapped instrument collection without having to modify
1295 the instrument files, also one can manage arbitrary amount of such
1296 MIDI instrument maps and assign each sampler channel individually
1297 a certain map, so that i.e. separation of normal instruments and
1298 drumkits is possible
1299 - new notification events for tracking changes to audio/MIDI devices,
1300 MIDI instrument maps, MIDI instruments, FX sends, global volume.
1301 - sampler was limited to load max. 200 instruments in the background
1302 due to a constant size RingBuffer FIFO which is now replaced by a
1303 dynamic (unlimited) size std::list FIFO
1304 - added FX sends, these allow to route audio signals to arbitrary audio
1305 output channels for being processed by external effect processors
1306 (i.e. jack-rack), the send levels are controllable via arbitrary MIDI
1307 controllers
1308 - global (sampler wide) volume can now be controlled at runtime
1309 - Implemented new, improved notification system
1310 - fixed compilation errors regarding OSX
1311 (patch by Toshi Nagata)
1312 - implemented instruments database
1313 - added support for escape sequences to the instruments database
1314 - added highly experimental support for on-the-fly instrument editing
1315 within the sampler's process (by using instrument editor plugins),
1316 you'll notice the new "Registered instrument editors:" message on
1317 startup, added a new LSCP command:
1318 "EDIT CHANNEL INSTRUMENT <sampler-channel>"
1319 to spawn a matching instrument editor for the instrument on the
1320 given sampler channel, the plugin path can be overridden at compile
1321 time with ./configure --enable-plugin-dir=/some/dir
1322 - added experimental code for synchronizing instrument editors hosted
1323 in the sampler's process to safely edit instruments while playing
1324 without a crash (hopefully) by either suspending single regions
1325 wherever possible or if unavoidable whole engine(s)
1326 - fixed several issues in fundamental "Thread" class: set scheduling
1327 policy and priority on thread level, set a minimum stack size for
1328 thread (TODO: a reasonable value yet to be tested), bugfix: non-RT
1329 threads simply inherited properties of starting thread instead of
1330 setting their own policy and priority
1331 - minor fix in our automatic stack trace mechanism on crashes, the
1332 main process did not wait for the stack trace process to finish
1333 its output
1334 - fixed some minor memory leaks
1335 - reenabled assembly features support, at the moment only for
1336 enabling a fast denormal FPU mode (x86 platforms supporting SSE2)
1337 - minor assembly fix in x86 features detection (don't use the PIC
1338 register, to avoid relocations in the text segment at runtime)
1339 - POSIX callback functions of Thread.h are hidden
1340
1341 * MIDI driver:
1342 - dispatch bank select (MSB and LSB) messages
1343
1344 * audio driver:
1345 - the ALSA audio output driver parameters now reflect the correct
1346 parameter value ranges for the respective selected sound card
1347 (patch by Till Wimmer, a bit fixed and extended)
1348
1349 * Gigasampler format engine:
1350 - if a filter is used and EG2 finishes before EG1, let the voice
1351 die when EG2 has finished (this fixes a problem with clicks and
1352 voice starvation for some gigs)
1353 - playback is no longer disabled during instrument loading
1354 - all notes playing on a channel that changes its instrument keep
1355 playing with the old instrument until they get a note off
1356 command
1357 - EG fix: a release value of zero could cause noises or crash
1358 - handle MIDI coarse tuning messages (MIDI RPN #0 MSB #2 LSB)
1359 - EG fine tuning: when attack is zero the EG starts at a level
1360 above max sustain level, which means that there is a decay phase
1361 even if sustain is 100%
1362 - more EG fixes: the level could sometimes go below zero and cause
1363 noises or crashes
1364 - minor fix of EGDecay (patch by Toshi Nagata)
1365 - fixed compiler error when --enable-override-filter-type was
1366 supplied to the configure script (fixes #46)
1367 - disk thread: queue sizes are now proportional to CONFIG_MAX_STREAMS
1368 instead of a fix value
1369 - behavior fix: on MIDI CC# 65 (portamento on / off), 126 (mono mode),
1370 127 (solo mode) only kill voices if the respective mode really
1371 changed
1372
1373 * LSCP server:
1374 - fixed compile time error for old Bison versions
1375 (i.e. v1.28 found on MacOS 10.4, patch by Toshi Nagata)
1376 - parser now supports extended ASCII character set
1377 (up to ASCII code 255, i.e. includes now umlauts and accents)
1378 - filename arguments in LSCP commands now allow to use escape
1379 sequences, that is directly literal as one of: \', \", \\, \n, \r,
1380 \f, \t, \v, or as octal ASCII code value like \132, or as hex ASCII
1381 code value like \xf2) (fixes bug #24)
1382 - the following LSCP commands now also support escape sequences for at
1383 least one of their text-based parameters (i.e. name, description):
1384 "ADD MIDI_INSTRUMENT_MAP", "MAP MIDI_INSTRUMENT",
1385 "SET MIDI_INSTRUMENT_MAP NAME", "SET FX_SEND NAME", "CREATE FX_SEND",
1386 "SET DB_INSTRUMENT_DIRECTORY NAME",
1387 "SET DB_INSTRUMENT_DIRECTORY DESCRIPTION", "SET DB_INSTRUMENT NAME",
1388 "SET DB_INSTRUMENT DESCRIPTION", "FIND DB_INSTRUMENTS",
1389 "FIND DB_INSTRUMENT_DIRECTORIES"
1390 - returns verbose syntax errors (line and column where syntax error
1391 occured, the unexpected character and the actually expected, possible
1392 character(s), the latter only if less than 5 possibilities)
1393 - made sure that LSCP syntax is not affected by gigedit locale
1394 settings
1395 - bugfix regarding strings parameter lists: all comma separated lists
1396 of strings were treated as being one string containing commas
1397 (fixes #57)
1398
1399 * Bug fixes:
1400 - fixed crash occurring on certain LSCP scripts (Bug 39)
1401 - another thread safety fix for lscp "load engine" and "set
1402 channel audio output device"
1403 - fixed a crash which occurs when reassigning the same engine
1404 on a sampler channel with connected MIDI device
1405 - fixed a crash which occurs when changing the number of ports of a MIDI
1406 device connected to a sampler channel to number less then or equal
1407 to the index of the port to which the sampler channel is connected.
1408 - The previous bindings were not been disconnected when altering
1409 the ALSA_SEQ_BINDINGS parameter. Introduced a NONE keyword for
1410 unsubscribing from all bindings (e.g. ALSA_SEQ_BINDINGS=NONE).
1411 - The active stream/voice count statistic was incorrect.
1412 - notification events were not been sent for some sampler
1413 channel changes
1414 - added default min and max values to restrict the number of allowed
1415 audio output channels and MIDI input ports
1416 - the connection to the PCM interface is now closed when destroying
1417 an audio output device
1418 - files with slash in their path or filename could not be loaded
1419
1420 * test cases:
1421 - updated and fixed (haven't been touched in a while)
1422
1423 Version 0.4.0 (24 November 2006)
1424
1425 * packaging changes:
1426 - changed deprecated copyright attribute to license; added ldconfig
1427 to post-(un)install steps; added devel package for liblinuxsampler;
1428 to linuxsampler.spec (RPM)
1429 - install necessary development header files for allowing 3rd party
1430 applications to link against liblinuxsampler
1431 - liblinuxsampler's API documentation can be generated with 'make docs'
1432 (Doxygen required)
1433 - added benchmark to automatically detect the best triangle LFO
1434 implementation (currently either an integer math solution or a
1435 di-harmonic approximation), automatic detection can be overriden
1436 with --enable-signed-triang-algo=x and --enable-unsigned-triang-algo=x
1437 configure script argument though (mandatory for cross-compilation)
1438 - do not automatically pick optimized gcc flags if the user already
1439 provided some on his own (as CXXFLAGS)
1440 - added compile time option to disable processing of All-Notes-Off MIDI
1441 messages
1442 - added compile time options to allow disabling the various audio and
1443 MIDI drivers
1444 - fixed automatic GCC CPU switch detection on PPC
1445 (patch by Ebrahim Mayat)
1446
1447 * Gigasampler format engine:
1448 - extensive synthesis optimization
1449 (reimplementation of EGs and LFO(s), removed synthesis parameter
1450 prerendering and the synthesis parameter matrix in general, splitting
1451 each audio fragment into subfragments now where each subfragment uses
1452 constant pitch and filter coefficients. The volume coefficient is
1453 linearly interpolated inside a subfragment, unless
1454 --disable-interpolate-volume is set.)
1455 - fine tuning of the EG modulation parameters
1456 - improved filter cutoff calculation by adding support for the
1457 following gig parameters: Cutoff freq (used when no cutoff
1458 controller is defined), Control invert, Minimum cutoff, Velocity
1459 curve and Velocity range. The keyboard tracking now scales
1460 cutoff frequency instead of resonance.
1461 - added support for gig parameter Resonance.
1462 - fixed bug in sysex handling (patch by Juan Linietsky)
1463 - added global volume attenuation of -9 dB (0.35f) to prevent clipping
1464 which can be overridden with --enable-global-attenuation
1465 - EG fixes: made the length of "attack hold" stage more
1466 accurate. Release stage can now start before attack stage
1467 ends. Cancel release didn't work when sustain was zero. Attack
1468 time now has a minimal value to prevent clicks.
1469 - fixed pitch changes larger than one octave
1470 - fixed EG3 (pitch envelope) synthesis which was neutral all the time
1471 - implemented portamento mode and solo mode (a.k.a 'mono mode'):
1472 all modes can be altered via standard GM messages, that is CC5 for
1473 altering portamento time, CC65 for enabling / disabling portamento
1474 mode, CC126 for enabling solo mode and CC127 for disabling solo mode
1475 - fine tuning of the curves for volume (CC7), pan (CC10 and gig
1476 parameter) and crossfade
1477 - added support for the "attenuation controller threshold" gig
1478 parameter
1479 - added smoothing of volume changes caused by control change
1480 messages
1481 - sample loop parameters are now taken from the DimensionRegion
1482 instead of the wave chunk
1483 - fixed keyswitching for v3 gigs with a number of keyswitch splits
1484 not equal to a power of two
1485 - reimplementation of the filter algorithm. The new filter is more
1486 accurate and supports all gig filter types, including bandreject
1487 and lowpass turbo.
1488 - real support for 24 bit samples - samples are not truncated to
1489 16 bits anymore
1490 - support for aftertouch (channel pressure, not polyphonic
1491 aftertouch)
1492 - LFO1 behaviour fixed (dampening from max volume instead of
1493 amplifying from 0)
1494
1495 * LSCP server:
1496 - fixed application exit on broken pipe error (fixes bug #20)
1497 - fixed the notification messages delay due to lack of
1498 network activity (fixes bug #26)
1499 - fixed parser bug which occured on space(s) within device parameters
1500 (e.g. "SET AUDIO_OUTPUT_CHANNEL_PARAMETER 0 0 SOMEPARAM='foo bar'")
1501
1502 * audio driver:
1503 - added aRts audio output driver (by no means RT safe)
1504
1505 * MIDI driver:
1506 - fixed legacy sysex code which caused dispatching of MIDI SysEx
1507 messages several times instead of once
1508 - API extension for MIDI drivers which already supply exact time stamps
1509 for events (i.e. for offline rendering based MIDI drivers)
1510 - added checks for bad MIDI data
1511
1512 * linuxsampler application:
1513 - show available sampler engine types on startup
1514
1515 * general changes:
1516 - support for muting sampler channels and solo mode of the same, two new
1517 LSCP commands ("SET CHANNEL MUTE" and "SET CHANNEL SOLO") and two new
1518 fields ("MUTE" and "SOLO") for command "GET CHANNEL INFO" were
1519 introduced for this, the behavior is the same like on a mixer console
1520 (patch by Grigor Iliev, a bit adjusted). Also added configure option
1521 --enable-process-muted-channels which can be used to enable the
1522 processing of muted channels.
1523 - support for sostenuto pedal
1524 - support for monitoring the total number of active voices
1525 (new LSCP commands: GET TOTAL_VOICE_COUNT, GET TOTAL_VOICE_COUNT_MAX,
1526 SUBSCRIBE TOTAL_VOICE_COUNT, UNSUBSCRIBE TOTAL_VOICE_COUNT)
1527 - fixed some memory management errors
1528 - fixed some concurrency problems that could lead to crashes when
1529 LSCP commands were executed
1530 - fixed crash when instrument loading failed previously on the same
1531 sampler channel (fixes bug #36)
1532
1533 Version 0.3.3 (15 July 2005)
1534
1535 * packaging changes:
1536 - fixed compilation with gcc 4.0
1537
1538 * Gigasampler format engine:
1539 - LFO preprocessing was performed even though the respective LFO was
1540 disabled by the instrument patch (minor efficiency fix)
1541 - if period time of chosen audio device is too small (< MIN_RELEASE_TIME)
1542 for volume ramp downs in the same fragment (needed for current voice
1543 stealing implementation) then simply inform the user with a warning
1544 about possible click sounds and reduce the volume ramp down
1545 appropriately instead of cancelling the audio device connection
1546
1547 Version 0.3.2 (24 June 2005)
1548
1549 * packaging changes:
1550 - updated autotools build files to compile on OS X
1551 (conditional compilation of CoreMIDI and MidiShare drivers)
1552 - hand-crafted assembly optimization code can be disabled with
1553 './configure --disable-asm' (definitely not recommended)
1554 - fixed 'make dist' rule to include all necessary files
1555 - require automake (>= 1.5) for 'make -f Makefile.cvs'
1556 (due to 'dist-bzip2' automake option)
1557
1558 * Gigasampler format engine:
1559 - support for the gig parameters for "release velocity response" curves
1560 - fine tuning of the EGADSR envelope
1561 - volume of release triggered samples now depends on note-on velocity,
1562 note length and gig parameter "release trigger decay" instead of
1563 note-off velocity.
1564 - revised voice stealing
1565 (fixes crash and endless loop caused by voice stealing)
1566 - don't reset scale tuning on instrument or audio output device change
1567 - handle key group conflicts right at the beginning of each fragment
1568 instead of when the respective voice is actually launched
1569 (fixes undefined behavior if stolen voices belonged to a key group -
1570 this case was followed by a "killed voice survived" error message)
1571 - fixed minor issue with null/silence samples
1572 (those stole voices even though they don't need a voice at all which
1573 resulted in "voice stealing didn't work out" messages)
1574 - don't reset volume, pan, pitch and MIDI controller values on
1575 instrument or audio output device change
1576
1577 * LSCP server:
1578 - fixed some crashes (patch by Grigor Iliev, fixes #19)
1579 - fixed LSCP event "CHANNEL_INFO" notification
1580 (e.g. did not notify on volume changes or MIDI program change events)
1581
1582 * linuxsampler application:
1583 - added command line parameters --lscp-addr and --lscp-port to override
1584 default IP address and port of LSCP server
1585
1586 Version 0.3.1 (24 May 2005)
1587
1588 * initial release

  ViewVC Help
Powered by ViewVC