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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 3625 - (show annotations) (download)
Thu Oct 3 13:37:25 2019 UTC (4 years, 5 months ago) by schoenebeck
File size: 94235 byte(s)
* gig format extension: Added support for different LFO wave forms
  (currently either sine [default], triangle, saw or square).

* gig format extension: Added support for LFO phase displacement
  (0°..360°).

* gig format extension: Added support for flipping LFO polarity on LFO 3
  (in the original gig format this was only available for LFO 1 and LFO 2).

* Bumped version (2.1.1.svn22).

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 - Added public API C++ class "LFO", which is a cluster class encapsulating
14 all the sampler's LFO implementations to be used by 3rd party applications
15 (e.g. by Gigedit).
16 - Added int math square LFO implementation.
17 - Added int math saw LFO implementation.
18 - Added numeric complex nr sine LFO implementation.
19 - Marked class LFOTriangleDiHarmonic as deprecated
20 (will be removed in future).
21 - Added LFOAll.h which includes all LFO implementation's header files.
22 - Refactored LFO class names and their file names (PulseLFO -> LFOPulse,
23 LFOSawIntMath -> LFOSawIntMathNew, SawLFO -> LFOSawIntMathOld,
24 SineLFO -> LFOSineBuiltinFn, LFOSine -> LFOSineNumericComplexNr,
25 SquareLFO -> LFOSquarePulse and separated the latter to its own header
26 file).
27 - Renamed type LFOSigned -> LFOTriangleSigned.
28 - Renamed type LFOUnsigned -> LFOTriangleUnsigned.
29
30 * Real-time instrument scripts:
31 - Added method ScriptVM::setExitResultEnabled() which allows to
32 explicitly enable the built-in exit() function to optionally accept
33 one function argument; the value of the passed exit() function
34 argument will then become available by calling
35 VMExecContext::exitResult() after script execution.
36 - 64 bit support for NKSP integer scripts variables (declare $foo).
37 - Variable names, function names and preprocessor condition names must start
38 with a regular character (a-z or A-Z); starting them with a digit or
39 underscore is not allowed.
40 - NKSP parser fix: equal comparison operator "=" and not equal comparison
41 operator "#" must only accept integer operands.
42 - NKSP language: Implemented support for standard units like Hertz, seconds,
43 Bel including support for metric unit prefixes; so one can now e.g.
44 conveniently use numbers in scripts like "5us" meaning "5 microseconds",
45 or e.g. "12kHz" meaning "12 kilo Hertz", or e.g. "-14mdB" meaning
46 "minus 14 Millidecibel", or e.g. "28c" meaning "28 cents" (for tuning).
47 - NKSP language: Introduced "final" operator "!" which is specifically
48 intended for synthesis parameter values to denote that the synthesis
49 parameter value is intended to be the "final" value for that synthesis
50 parameter that should explicitly be used by the engine and thus causing
51 the sampler engine to ignore all other modulation sources for the same
52 synthesis parameter (like e.g. LFO, EG); by simply prefixing a value,
53 variable or formula with this new "!" operator the expression is marked as
54 being "final".
55 - NKSP script editor API: Added support for detecting standard unit tokens
56 and their potential metric prefix token.
57 - NKSP language: Added support for NKSP real number literals and
58 arithmetic operations on them (e.g. "(3.9 + 2.9) / 12.3 - 42.0").
59 - NKSP language: Added support for NKSP real number (floating point) script
60 variables (declare ~foo).
61 - NKSP language: Added support for NKSP real number (floating point) array
62 script variables (declare ?foo[]).
63 - Built-in script function "message()" accepts now real number argument as
64 well.
65 - Added built-in script function "real_to_int()" and its short hand form
66 "int()" for casting from real number to integer in NKSP scripts.
67 - Added built-in script function "int_to_real()" and its short hand form
68 "real()" for casting from integer to real number in NKSP scripts.
69 - Allow built-in exit() function to potentially accept real number type
70 argument as well.
71 - Built-in script functions may have a different return type depending on
72 the arguments passed to the function.
73 - Built-in script function "abs()" optionally accepts and returns real
74 number.
75 - Built-in script functions "min()" and "max()" optionally accept real
76 number arguments and return real number as result in that case.
77 - NKSP VM API: Allow units and 'final'ness to be returned as result from
78 built-in functions (added methods VMFunction::returnUnitType() and
79 VMFunction::returnsFinal() for that purpose which must be implemented by
80 built-in function implementations).
81 - NKSP language: Allow metric unit prefixes of numeric scalar and array
82 variables to be changed freely at runtime (unlike unit types like Hz etc.
83 which are still sticky, parse-time features of variables which cannot be
84 changed at runtime for the intentional sake of determinism).
85 - NKSP language: 'final' values are prohibited for array variables for now
86 (attempt causes a parsers error).
87 - NKSP language: expressions with unit types (e.g. Hz) are prohibited for
88 conditions of runtime control structures like if(), while(), select()
89 (attempt causes a parser error).
90 - NKSP VM API: Allow built-in functions to perform their own, individual
91 parse time checks of arguments going to be passed to the function at
92 runtime (added method VMFunction::checkArgs() for that purpose).
93 - NKSP language: raise parser warning if only one operand of binary
94 operators (like logical 'or' comparison) contain a 'final' value (because
95 it would always yield in a 'final' result in such cases).
96 - NKSP language: Allow comparison (=, #, <, >, <=, >=) of values with
97 different metric unit prefixes, which will behave as expected (e.g.
98 result of expression '1000us < 2ms' is true).
99 - NKSP language: Allow adding values with different metric unit prefixes
100 (e.g. result of expression '100Hz + 5kHz' is '5100Hz').
101 - NKSP language: Allow subtracting values with different metric unit
102 prefixes (e.g. result of expression '1ms - 20us' is '980us').
103 - NKSP language: Allow multiplying with any metric unit prefixes
104 (e.g. result of expression '2k * 3ms' is '6s'), however multiplications
105 with unit types on both sides (e.g. '2s * 2s') is still prohibited since
106 we don't have any considerable practical use for a term like '4s^2'
107 (hence any attempt multiplying two unit types still causes parser error).
108 - NKSP language: Allow dividing by any metric unit prefixes and allow
109 division of same unit type on both sides (e.g. expression '8kHz / 1000Hz'
110 yields in unit free result '8'). So this is now a way to cast units away
111 e.g. for passing the result to other expressions, certain function calls
112 or variables which are not accepting any units (or that specific unit).
113 - NKSP language: integer arrays and real number arrays can now be converted
114 to strings (e.g. for dumping their content with message() calls for
115 script debugging purposes).
116 - NKSP language: expressions and variables with units are now correctly
117 casted to strings (e.g. with message() calls).
118 - NKSP language: comparing real numbers for equalness (e.g. '~foo = 3.1') or
119 unequalness (e.g. '~foo # 3.1') is now less strict and takes the expected
120 floating point tolerances into account.
121 - NKSP VM API: Added methods VMScalarNumberExpr::evalCastInt() and
122 VMScalarNumberExpr::evalCastReal().
123 - NKSP VM API: Added base class 'VMNumberArrayExpr' for classes
124 'VMIntArrayExpr' and 'VMRealArrayExpr'.
125 - NKSP VM API: replaced all unitPrefix() (parse time) methods by
126 unitFactor() (runtime) methods.
127 - Built-in function "exit()" supports now returning units and 'final'ness
128 for test cases.
129 - The following built-in functions support now units as well: "abs()",
130 "random()", "inc()", "dec()", "in_range()", "min()", "max()",
131 "real_to_int()", "int()", "int_to_real()" and "real()".
132 - Built-in functions "array_equal()", "search()" and "sort()" support now
133 real number arrays (correctly) as well.
134 - Added individual parse time checks of arguments to be passed to built-in
135 functions "random()", "inc()", "dec()", "in_range()", "min()", "max()",
136 "array_equal()" and "search()" specific for their individual purposes.
137 - NKSP VM refactoring: Renamed all methods, functions and classes matching
138 pattern *ScalarNumber* to simply *Number* (that is i.e. classes
139 VMScalarNumberExpr -> VMNumberExpr, ScalarNumberExpr -> NumberExpr,
140 ScalarNumberVariable -> NumberVariable, ScalarNumberBinaryOp ->
141 NumberBinaryOp, VMScalarNumberResultFunction -> VMNumberResultFunction,
142 method VMExpr::asScalarNumberExpr() -> VMExpr::asNumber(), function
143 isScalarNumber() -> isNumber()).
144 - NKSP VM API: Added 4 overridden methods to class VMNumberExpr:
145 evalCastInt(MetricPrefix_t), evalCastInt(MetricPrefix_t,MetricPrefix_t),
146 evalCastReal(MetricPrefix_t), evalCastReal(MetricPrefix_t,MetricPrefix_t)
147 as convenient methods for automatically converting values to expected
148 metric value basis.
149 - Built-in function "wait()" accepts now both integers and real numbers as
150 argument.
151 - NKSP VM API cleanup: Get rid of legacy method
152 VMFunction::argType(vmint iArg) which was already superseded by its new
153 replacement VMFunction::acceptsArgType(vmint iArg, ExprType_t type).
154 - NKSP parser: if wrong argument type was passed to a built-in function and
155 that built-in function accepts more than one data type for the argument,
156 then show all supported data types as parser error message.
157 - Built-in function "play_note()" accepts now real numbers and seconds as
158 unit type as well for its 3rd and 4th function arguments.
159 - The following built-in functions accept now real numbers as well for their
160 2nd function argument: "change_vol()", "change_tune()", "change_cutoff()",
161 "change_attack()", "change_decay()", "change_release()",
162 "change_sustain()", "change_cutoff_attack()", "change_cutoff_decay()",
163 "change_cutoff_sustain()", "change_cutoff_release()",
164 "change_amp_lfo_freq()", "change_cutoff_lfo_freq()",
165 "change_pitch_lfo_freq()", "change_vol_time()", "change_tune_time()",
166 "change_pan_time()", "fade_in()", "fade_out()", "change_play_pos()".
167 - Fixed built-in function "change_play_pos()" not having accepted metric
168 prefixes at all.
169 - Fixed the following built-in functions having misinterpreted values given
170 with unit type (for their 2nd argument) as if they were relative values
171 (that is as if they were passed without a unit type): "change_attack()",
172 "change_decay()", "change_release()", "change_cutoff_attack()",
173 "change_cutoff_decay()", "change_cutoff_release()".
174 - Fixed the following built-in functions having applied completely wrong
175 'final' values: "change_sustain()", "change_cutoff_sustain()" (since the
176 respective EGs being their modulation sink assume uint data type with
177 value range 0..1000 instead of 0.0..1.0.
178 - Added individual parse-time checks of function arguments for the following
179 built-in functions: "play_note()", "note_off()", "set_event_mark()",
180 "delete_event_mark()", "by_marks()", "change_cutoff()", "change_attack()",
181 "change_decay()", "change_release()", "change_cutoff_attack()",
182 "change_cutoff_decay()", "change_cutoff_release()",
183 "change_amp_lfo_freq()", "change_cutoff_lfo_freq()",
184 "change_pitch_lfo_freq()", "change_vol_time()", "change_tune_time()" and
185 "change_pan_time()".
186 - Don't abort function call if unit type was used and at the same time
187 'final' operator was omitted for the primary value argument of the
188 following built-in functions: "change_cutoff()", "change_attack()",
189 "change_decay()", "change_release()", "change_cutoff_attack()",
190 "change_cutoff_decay()", "change_cutoff_release()",
191 "change_amp_lfo_freq()", "change_cutoff_lfo_freq()",
192 "change_pitch_lfo_freq()", "change_vol_time()", "change_tune_time()",
193 "change_pan_time()", instead imply 'final'ness at runtime and raise an
194 appropriate parser warning at parse time.
195 - Added built-in real number functions "round()", "ceil()", "floor()",
196 "sqrt()", "log()", "log2()", "log10()", "exp()", "pow()", "sin()",
197 "cos()", "tan()", "asin()", "acos()", "atan()".
198 - Added built-in script real number constant "~NI_MATH_PI".
199 - Added built-in script real number constant "~NI_MATH_E".
200 - NKSP language: Allow unary '+' operator.
201
202 * test cases:
203 - Fixed compiler errors in test cases.
204 - Updated README for how to compile & run test cases.
205 - Updated test case
206 MutexTest::testDoubleLockStillBlocksConcurrentThread() to latest
207 expected behaviour of the Mutex class implementation (recursive
208 mutex type).
209 - Added test cases for NKSP core language aspects and core built-in
210 functions.
211 - Fixed thread tests segfaulting on Linux.
212 - NKSP: Added real number test cases for built-in functions exit(),
213 int_to_real(), real(), real_to_int() and int(), as well as for the
214 plus, minus and negate language operators.
215 - Added massive amount of NKSP test cases for standard measuring units and
216 'final' operator usage cases.
217 - Added NKSP test cases for (floating point tolerance aware) real number
218 equalness / unequalness comparison.
219 - Added NKSP int array and real array tests for value assignment and
220 initialization of arrays.
221 - Added NKSP test cases for built-in functions "round()", "ceil()",
222 "floor()", "sqrt()", "log()", "log2()", "log10()", "exp()", "pow()",
223 "sin()", "cos()", "tan()", "asin()", "acos()", "atan()".
224 - Added NKSP test cases for unary '+' operator.
225
226 * GigaStudio/Gigasampler format engine:
227 - LFOTriangleIntMath and LFOTriangleIntAbsMath: Fixed FlipPhase=true
228 behaviour for start_level_mid.
229 - Changed LFO start levels: LFO1 and LFO2 both to mid, LFO3 to max. start
230 level (see discussion "GigaStudio LFO compatibility" on mailing list
231 from 2019-09-26 for details).
232 - Changed default wave form for all 3 LFOs to sine (instead of triangle;
233 see discussion "GigaStudio LFO compatibility" on mailing list from
234 2019-09-26 for details).
235 - Format extension: Added support for different LFO wave forms (currently
236 either sine [default], triangle, saw or square).
237 - Format extension: Added support for LFO phase displacement (0��..360��).
238 - Format extension: Added support for flipping LFO polarity on LFO 3
239 (in the original gig format this was only available for LFO 1 and LFO 2).
240
241 * SFZ format engine:
242 - Fixed support for regions with loccN/hiccN conditions on more than one
243 MIDI controller.
244
245 * Benchmarks:
246 - Fixed benchmarks/triang.cpp falsely having favoured "int math abs"
247 algorithm (since result of 2nd run was not accumulated).
248 - Added benchmark for saw wave (benchmarks/saw.cpp).
249 - Added benchmark for sine wave (benchmarks/sine.cpp).
250 - Added benchmark for square wave (benchmarks/square.cpp).
251 - Increased amount of benchmarks runs by factor 6 to achieve benchmark times
252 which are large enough on modern systems.
253
254 Version 2.1.1 (27 Jul 2019)
255
256 * Real-time instrument scripts:
257 - Fixed behavior of built-in NKSP functions change_sustain(),
258 change_cutoff_attack(), change_cutoff_decay(), change_cutoff_sustain()
259 and change_cutoff_release().
260
261 * general changes:
262 - Only play release trigger samples on sustain pedal up if this behaviour
263 was explicitly requested by the instrument (otherwise only on note-off).
264 - Fixed compiler warnings.
265 - Fixed compilation error when cross-compiling to Mac.
266 - FX Sends: Provide more useful error messages on routing problems
267 (see bug #169).
268 - LSCP doc: Be more clear describing the two distinct approaches
269 of using external vs. internal effects (see bug #169).
270 - "optional" class: Fixed comparison operators.
271
272 * Gigasampler/GigaStudio format engine:
273 - Format extension: If requested by instrument then don't play release
274 trigger sample on note-off events.
275
276 * SFZ format engine:
277 - Fixed memory leak when releasing samples
278 (fixes bug #307, patch by Jacek Roszkowski)
279 - Fixed potential crash when a sample is shared by more than one region
280 (fixes bug #308, patch by Jacek Roszkowski).
281 - Opcode 'sample': Added support for built-in sample '*silence'
282 (fixes bug #310, patch by Jacek Roszkowski).
283
284 Version 2.1.0 (25 Nov 2017)
285
286 * SFZ format engine:
287 - added support for <global>, <master> and #define (patch by Alby M)
288 - Removed code duplication in SFZ file loading code.
289 - Added support for sfz extension opcode 'script' which may be used to
290 load real-time instrument script file (NKSP script language).
291 - Implemented opcode set_ccN (initial patch by Giovanni Senatore).
292 - Fixed unintended volume fade-in of voices under certain conditions.
293 - sfz parser: allow missing space between header and opcode
294
295 * Gigasampler/GigaStudio format engine:
296 - Fixed clicks and pumping noise with Lowpass Turbo filter on very low
297 cutoff settings.
298 - Got rid of resembling an ancient GSt misbehavior which did not pitch at
299 all if an up-pitch of more than 40 semi tones was requested (I don't
300 think there is any stock gig sound that requires this behavior to
301 resemble its original sound).
302 - Added support for controlling whether the individual EGADSR stages may
303 be aborted (as LinuxSampler extension to the original GigaStudio 4
304 format).
305
306 * general changes:
307 - fixed printf type errors (mostly in debug messages)
308 - use unique_ptr instead of auto_ptr when building with C++11
309 - Added RTAVLTree class which is a real-time safe ordered multi-map, thus
310 allowing to sort data efficiently in real-time safe manner.
311 - RTList class: added methods for moving/inserting elements to arbitrary
312 position within a list.
313 - RTList class: added method fromPtr() for creating an Iterator object from
314 a raw element pointer.
315 - Attempt to partly fix resetting engine channels vs. resetting engine, an
316 overall cleanup of the Reset*(), ConnectAudioDevice(),
317 DisconnectAudioDevice() API methods would still be desirable though,
318 because the current situation is still inconsistent and error prone.
319 - Active voices are now internally grouped to "Note" objects, instead of
320 being directly assigned to a keyboard key. This allows more fine graded
321 processing of voices, which is i.e. required for certain instrument
322 script features.
323 - Fix: Release trigger voices were not spawned on sustain pedal up
324 (CC #64) events.
325 - Fix: Release trigger voices can now also distinguish correctly between
326 note-off and sustain pedal up events.
327 - All engines: Increased ramp speed of volume smoother and pan smoother
328 (while slow rate of crossfade smoother is preserved) to allow quick
329 volume and pan changes by instrument scripts for instance.
330 - gig/sf2/sfz: Fixed aftertouch channel pressure events being processed
331 properly and thus were ignored for certain uses.
332 - Added new C++ API method
333 VirtualMidiDevice::SendChannelPressureToSampler().
334 - windows, 32-bit: fixed potential crashes by making sure the stack in
335 sub threads is 16-byte aligned
336 - fixed numerous compiler warnings
337 - Fixed invalid (note-on) event ID being assigned to new Note objects.
338 - Revised fundamental C++ classes "Thread", "Mutex" and "Condition" which
339 fixes potential undefined behavior.
340 - Fixed Note object leak when triggering notes on keys which did not
341 have a valid sample mapped (fixes bug #252).
342 - Fixed compilation errors when compiling with CONFIG_DEVMODE enabled.
343 - linuxsampler binary fix: option --create-instruments-db ignored
344 subsequent optional argument due to glibc's implementation oddity
345 which expects a "=" sign, but no space between them.
346
347 * packaging changes:
348 - removed unnecessary dependency to libuuid
349 (originated by libgig's usage of it)
350 - Automake: set environment variable GCC_COLORS=auto to allow GCC to
351 auto detect whether it (sh/c)ould output its messages in color.
352 - Debian: Fixed packaging error about invalid "Source-Version"
353 substitution variable.
354 - Debian: Raised Debian compatibility level to Debian 9 "Stretch".
355 - Debian: Added build dependency to libsqlite3-dev for building
356 linuxsampler with instruments DB support.
357
358 * Real-time instrument scripts:
359 - Implemented scheduler for delayed MIDI events and for suspended scripts.
360 - Built-in script function "wait()": implemented support for function's
361 "duration-us" argument, thus scripts using this function are now
362 correctly resumed after the requested amount of microseconds.
363 - Built-in script function "play_note()": implemented support for
364 function's "duration-us" argument, thus notes triggered with this
365 argument are now correctly released after the requested amount of
366 microseconds.
367 - Fix: script events were not cleared when engine channel was reset,
368 potentially causing undefined behavior.
369 - Fixed crash which happened when trying to reference an undeclared
370 variable.
371 - Built-in script function "play_note()": Added support for passing
372 special value -1 for "duration-us" argument, which will cause the
373 triggered note to be released once the original note was released.
374 - Instrument script classes now exported with the liblinuxsampler C++ API.
375 - Added new API method ScriptVM::syntaxHighlighting() which provides
376 a convenient syntax highlighting backend for external instrument
377 script editor applications.
378 - Added new C++ API class "ScriptVMFactory".
379 - Extended parser issues to provide not only first line and first
380 column, but also last line and last column of issue (thus marking
381 the precise span of the issue within the source code).
382 - Reload script automatically after being modified by an instrument
383 editor.
384 - NKSP language grammar correction: allow empty event handler bodies
385 like "on note end on".
386 - Implemented built-in script function "change_vol()".
387 - Implemented built-in script function "change_tune()".
388 - Implemented built-in script function "change_pan()".
389 - Implemented built-in script function "change_cutoff()".
390 - Implemented built-in script function "change_reso()".
391 - Implemented built-in script function "event_status()".
392 - Added built-in script constants "$EVENT_STATUS_INACTIVE" and
393 "$EVENT_STATUS_NOTE_QUEUE" both for being used as flags for
394 "event_status()" function.
395 - NKSP language: Added support for bitwise operators ".or.", ".and."
396 and ".not.".
397 - NKSP language scanner: Fixed IDs matching to require at least one
398 character (i.e. when matching function names or variable names).
399 - NKSP language scanner: disabled unusued rules.
400 - Fixed behavior of play_note() and note_off() functions which must
401 be distinguished engine internally from "real" MIDI note on/off
402 events in order to avoid misbehaviors like hanging notes.
403 - Implemented built-in script variable "$KSP_TIMER".
404 - Implemented built-in script variable "$NKSP_REAL_TIMER".
405 - Implemented built-in script variable "$NKSP_PERF_TIMER".
406 - Implemented built-in script variable "$ENGINE_UPTIME".
407 - Implemented built-in script function "inc()".
408 - Implemented built-in script function "dec()".
409 - NKSP language fix: division expressions were evaluated too often.
410 - NKSP language fix: string concatenation operator was right
411 associative instead of left (to right).
412 - NKSP language correction: allow empty statements in entire language
413 (i.e. "if end if").
414 - Implemented built-in script function "stop_wait()".
415 - Implemented built-in script variable "$NI_CALLBACK_ID".
416 - Implemented built-in script variable "$NI_CALLBACK_TYPE".
417 - Implemented built-in script variable "$NKSP_IGNORE_WAIT".
418 - Added support for read-only built-in variables (respectively
419 handled by the script parser).
420 - Added built-in script constant "$NI_CB_TYPE_INIT".
421 - Added built-in script constant "$NI_CB_TYPE_NOTE".
422 - Added built-in script constant "$NI_CB_TYPE_RELEASE".
423 - Added built-in script constant "$NI_CB_TYPE_CONTROLLER".
424 - NKSP Language: Added support for user defined script functions.
425 - Implemented built-in script function "change_attack()".
426 - Implemented built-in script function "change_decay()".
427 - Implemented built-in script function "change_release()".
428 - Fixed all change_*() built-in script functions to apply their
429 synthesis parameter changes immediately in case the respective note
430 was triggered at the same time, instead of scheduling the parameter
431 change, especially because it would cause some parameter types's
432 changes either to be ramped (i.e. change_vol()) or other types even
433 to have not effect at all (i.e. change_attack()).
434 - Implemented built-in script function "sh_left()".
435 - Implemented built-in script function "sh_right()".
436 - Implemented built-in script function "min()".
437 - Implemented built-in script function "max()".
438 - NKSP Fix: "init" event handler was not always executed when a script
439 was loaded.
440 - NKSP built-in wait() script function: abort script execution if
441 a negative or zero wait time was passed as argument (since this is
442 a common indication of a bug either of the script or even of the
443 engine, which could lead to RT instability or even worse).
444 - ScriptVM: Implemented automatic suspension of RT safety
445 threatening scripts.
446 - Provide more user friendly error messages on syntax errors.
447 - Fixed NKSP parser warning "Not a statement" when assigning an
448 initializer list to an array variable.
449 - Implemented built-in script array variable "%ALL_EVENTS".
450 - Implemented built-in script function "in_range()".
451 - Implemented built-in script function "change_amp_lfo_depth()".
452 - Implemented built-in script function "change_amp_lfo_freq()".
453 - Implemented built-in script function "change_pitch_lfo_depth()".
454 - Implemented built-in script function "change_pitch_lfo_freq()".
455 - Implemented built-in script function "change_vol_time()".
456 - Implemented built-in script function "change_tune_time()".
457 - Implemented built-in script function "fade_in()".
458 - Implemented built-in script function "fade_out()".
459 - Fixed acceptance of wrong data type of parameters passed to built-in
460 script functions "change_vol()", "change_tune()", "change_pan()",
461 "change_cutoff()", "change_reso()", "change_attack()", "change_decay()",
462 "change_release()", "change_amp_lfo_depth()", "change_amp_lfo_freq()",
463 "change_pitch_lfo_depth()" and "change_pitch_lfo_freq()".
464 - Added built-in script function "get_event_par()" and implemented some
465 of its possible parameter selections.
466 - Added built-in script function "set_event_par()" and implemented some
467 of its possible parameter selections.
468 - Fixed a bunch of scheduler time related bugs.
469 - Fixed polyphonic variables not being reset to zero after usage.
470 - Built-in "ignore_event()" function: argument is now optional, like with
471 built-in function "ignore_controller()".
472 - Implemented built-in script function "change_velo()".
473 - Implemented built-in script function "change_note()".
474 - Adjusted behavior of "change_vol()" and "change_tune()" to a more
475 intuitive behavior if used in combination with "change_vol_time()" or
476 "change_tune_time()" respectively: now tuning/volume changes are only
477 assigned (without delay) immediately to a new note if the respective
478 timing function has not been called before, otherwise the volume/tuning
479 changes are automatically faded (before, only the event's time stamp was
480 relevant).
481 - Implemented built-in script function "array_equal()".
482 - Implemented built-in script function "search()".
483 - Implemented built-in script function "sort()".
484 - NKSP Fix: Never suspend "init" event handlers.
485 - Implemented built-in script function "same_region()" (currently only
486 available for gig format engine).
487 - Added built-in script constant "$NKSP_LINEAR".
488 - Added built-in script constant "$NKSP_EASE_IN_EASE_OUT".
489 - Implemented built-in script function "change_vol_curve()".
490 - Implemented built-in script function "change_tune_curve()".
491 - built-in "play_note()" function now supports a sample playback start
492 offset with argument 3, where special value -1 means to use the regular
493 sample offset as defined by the instrument file.
494 - Built-in array variable %KEY_DOWN[] is now a read-only variable.
495 - Built-in variable $EVENT_NOTE is now a read-only variable.
496 - Built-in variable $EVENT_VELOCITY is now a read-only variable.
497 - built-in "play_note()" function now accepts -2 for its fourth argument
498 (note duration) which means the life time of the note shall be sticked
499 to the requested note number of argument 1.
500 - Fix: built-in "play_note()" function now returns 0 as result value if
501 -1 was passed for its fourth argument (note duration) and the respective
502 parent note is already gone.
503 - Implemented built-in script function "change_play_pos()".
504 - NKSP language: Added support for user declared const array variables.
505 - NKSP language: Raise parser warning if array variable is accessed with
506 an index that exceeds the array's size.
507 - NKSP language fix: Unknown characters were not handled correctly.
508 - NKSP language: Added support for "synchronized .. end synchronized"
509 code blocks.
510 - Implemented built-in script function "abort()" which allows to abort
511 another script handler by passing its callback ID.
512 - Fixed potential memory access bug and potential undefined behavior of
513 "init" event handlers.
514 - Print a time stamp along to each call of built-in function "message()".
515 - ScriptVM API: Added VMParserContext::preprocessorComments() which allows
516 to retrieve all code blocks filtered out by the preprocessor.
517 - Added built-in script function "fork()".
518 - Added built-in array variable %NKSP_CALLBACK_CHILD_ID[].
519 - Added built-in variable $NKSP_CALLBACK_PARENT_ID.
520 - Fixed potential crash when accessing dynamic built-in array variables.
521 - Added built-in script function "callback_status()".
522 - Added built-in constant $CALLBACK_STATUS_TERMINATED.
523 - Added built-in constant $CALLBACK_STATUS_QUEUE.
524 - Added built-in constant $CALLBACK_STATUS_RUNNING.
525 - Removed max. value limitation of built-in functions "change_attack()",
526 "change_decay()" and "change_release()" to i.e. allow passing 2000000
527 for doubling the respective time.
528 - NKSP script editor syntax highlighting API: Fixed app termination due
529 to a lexer start condition stack underrun.
530 - NKSP preprocessor: Fixed wrong behavior on nested USE_CODE_IF() and
531 USE_CODE_IF_NOT() preprocessor statements.
532 - NKSP: Added built-in preprocessor condition NKSP_NO_MESSAGE, which
533 can be set to disable all subsequent built-in "message()" function calls
534 on preprocessor level.
535 - Implemented built-in script function "change_sustain()".
536 - NKSP script editor syntax highlighting API: catch all fatal lexer errors,
537 to avoid the editor app to crash on ill-formed text input.
538 - Added built-in script function "change_pan_time()".
539 - Added built-in script function "change_pan_curve()".
540 - Added built-in script function "change_cutoff_attack()".
541 - Added built-in script function "change_cutoff_decay()".
542 - Added built-in script function "change_cutoff_sustain()".
543 - Added built-in script function "change_cutoff_release()".
544 - Added built-in script function "change_cutoff_lfo_depth()".
545 - Added built-in script function "change_cutoff_lfo_freq()".
546
547 * Instruments DB:
548 - Fixed memory access bug of general DB access code which lead to
549 undefined behavior.
550 - Cleanup of instruments DB file creation and opening code.
551 - The instrument DB path of linuxsampler's --create-instruments-db argument
552 is now optional, if it is missing, then a default location is used.
553 - Added support for scanning SFZ (.sfz) files.
554 - Added support for scanning Sound Font (.sf2) files.
555 - Fixed undefined DB transaction behavior.
556
557 Version 2.0.0 (15 July 2015)
558
559 * packaging changes:
560 - fixed building with newer MinGW-w64
561 - Mac OS X: support the new dir for Core Audio SDK
562 - Mac OS X: fixed building outside source directory
563 - made sure all source files for hostplugins are included when
564 doing "make dist"
565 - removed empty directories left from the cvs to svn migration
566 - fixed compilation with gcc 4.6.1
567 - another "make dist" fix, for the LV2 plugin
568 - made --enable-pthread-testcancel default on Mac OS X
569 - minor Makefile fix for building lscpparser when build and source
570 directories are separate
571 - Mac OS X: made it possible to specify plugin installation dir to
572 configure
573 - Mac OS X: Makefile fix for the install-strip target
574 - fixed compilation with gcc 4.7
575 - fixed configure script error with old autoconf versions
576 - lsatomic.h: use gcc provided atomic functions if building with
577 gcc 4.7 and C++11
578 - modernized configure script
579 - fixed linkage error when building with
580 LDFLAGS="-Wl,--no-undefined" (#190)
581 - fixed compilation with Clang 3.2
582 - removed usage of deprecated Automake variable INCLUDES
583 - fixed building with C++11
584 - build fix: ChangeFlagRelaxed.h was missing in makefile
585 - build fix: libsndfile compiler flags were missing in some
586 makefiles
587 - fix for building with bison 3.0 (#202)
588 - Mac OS X: added temporary hack allowing to spawn gigedit as callback
589 on the process's main thread
590 - fixed build error on newer MinGW
591 - support building with older jack versions
592 - support building with spaces in vst sdk path
593 - enabled automake 'subdir-objects' option and moved external
594 source references (vst, au, asio) from makefiles to cpp files,
595 in order to get rid of warnings from automake 1.14
596
597 * general changes:
598 - Refactoring: moved the independent code from
599 the Gigasampler format engine to base classes
600 - command line option '--profile' is currently disabled, since the
601 respective profiling code is currently broken
602 - Introduced support for optional environment variable
603 "LINUXSAMPLER_PLUGIN_DIR" which allows to override the directory
604 where the sampler shall look for instrument editor plugins
605 (patch by Luis Garrido, slightly modified).
606 - implemented Roland GS NRPN 1ArrH which allows to set volume per note
607 - implemented Roland GS NRPN 1CrrH which allows to set pan per note
608 - implemented Roland GS NRPN 1DrrH which allows to set reverb send per
609 note (in this implementation of the sampler its simply hard routed to
610 the 1st effect send of the sampler channel, no matter what the actual
611 effect type is)
612 - implemented Roland GS NRPN 1ErrH which allows to set chorus send per
613 note (in this implementation of the sampler its simply hard routed to
614 the 2nd effect send of the sampler channel, no matter what the actual
615 effect type is)
616 - implemented support for internal LADSPA effects
617 - raised limit of program change queue from 100 to 512
618 (as suggested by Alex Stone)
619 - implemented sine LFO, pulse LFO and saw LFO
620 - experimental support for per voice equalization
621 - added command line option --exec-after-init
622 - Introduced new C++ API method:
623 EngineChannel::InstrumentFileName(int index)
624 allowing to retrieve the whole list of files used for the loaded
625 instrument on an engine channel (a.k.a. part). Some GigaStudio
626 instruments for example are splitted over several files like
627 "Foo.gig", "Foo.gx01", "Foo.gx02", ...
628 - Added new C++ API method Sampler::GetGlobalMaxVoices().
629 - Added new C++ API method Sampler::GetGlobalMaxStreams().
630 - Added new C++ API method Sampler::SetGlobalMaxVoices().
631 - Added new C++ API method Sampler::SetGlobalMaxStreams().
632 - Various "const" and "restrict" optimizations.
633 - all engines: add pan CC value to instrument pan parameter before
634 applying panning, instead of using two separate pan functions in
635 series (#182)
636 - added a lock guard class for exception safe mutex handling and
637 used it everywhere appropriate
638 - Immediately apply scale tuning changes to active voices.
639 - Exposed scale tuning to C++ API (along to the already existing standard
640 SysEx way).
641 - lsatomic.h fixes: seq_cst load and store were suboptimal for x86
642 and broken for ppc64. (Seq_cst loads and stores are actually not
643 used in LS, so the bug wasn't noticable.)
644 - lsatomic.h: added ARMv7 support
645 - Added support for multiple MIDI input ports per sampler channel (and
646 added various new C++ methods for this new feature / design change, old
647 C++ API methods for managing SamplerChannel's MIDI inputs are now marked
648 as deprecated but are still there and should provide full behavior
649 backward compatibility).
650 - AbstractEngine::GSChecksum(): don't allocate memory on the stack (was
651 unsafe and caused compilation error with clang 2.x).
652 - Bugfix: only process the latest MIDI program change event.
653 - Introducing the LSCP shell, which provides convenient control of the
654 sampler from the command line by providing LSCP aware features.
655 - VirtualMidiDevice: Added support for program change.
656 - VirtualMidiDevice: Added support for bank select (MSB & LSB).
657 - VirtualMidiDevice: Added support for pitch bend.
658 - Aftertouch: extended API to explicitly handle channel pressure and
659 polyphonic key pressure events (so far polyphonic pressure was not
660 supported at all, and channel pressure was rerouted as CC128 but not
661 used so far).
662 - Added initial support for real-time instrument scripts. The script VM
663 code is shared by all sampler engine implemementations, however only the
664 gig file format currently provides support for storing instrument scripts
665 (as LinuxSampler extension to the original GigaStudio 4 format).
666
667 * Gigasampler format engine:
668 - implemented the "round robin keyboard" dimension
669 - fixed round robin and random dimensions for cases when number of
670 dimension zones is not a power of two
671 - made round robin use a counter for each region instead of each
672 key
673 - bugfix: pitch LFO controller "internal+aftertouch" was broken
674 - bugfix: filter keyboard tracking was broken
675 - filter performance fix (an unnecessary copy was made of the
676 filter parameters in each sub fragment)
677 - handle special case when pan parameter in gig file has max or
678 min value
679 - Exclusive Groups: don't ever stop voices of the same note,
680 doesn't sound naturally with a drumkit
681 - fixed EG1 modulation when attack or release is zero
682 - Fixed support for 'aftertouch' attenuation controller.
683 - Fixed crash that happened with velocity split sounds under certain
684 conditions (see also previous commit on libgig).
685 - fixed behaviour of filter LFO
686
687 * SFZ format engine:
688 - Initial implementation (not usable yet)
689 - added support for v2 multiple stage envelope generators
690 - added a fine-tuned v1 envelope generator instead of using the
691 one from the gig engine
692 - fixed memory leak and memory handling errors
693 - added support for trigger=first, trigger=legato and sw_previous
694 - allow non-numerical key values ("C#4" for example)
695 - "key" opcode now sets pitch_keycenter too
696 - fixed error when unloading instrument with same sample used by
697 multiple regions
698 - added some opcode aliases, like loopmode for loop_mode, to be
699 more compatible
700 - added support for trigger=release and rt_decay
701 - added support for off_mode=normal
702 - added support for random, seq_position, seq_length and volume
703 - added v1 LFO opcodes to parser (no support in engine yet)
704 - added support for amp_veltrack and amp_velcurve_N
705 - fine-tuned the default velocity curve
706 - added support for transpose
707 - fixed crash when using sw_down/up
708 - improved logic for sw_lokey/hikey/up/down/last
709 - added more v1 aliases to parser
710 - reduced memory usage for sfz data
711 - RT-safeness: avoid malloc in audio thread
712 - fixed a bug that could cause voice stealing to fail
713 - optimized sample lookup
714 - improved support for exclusive groups (group, off_by and
715 off_mode)
716 - added support for controller triggered regions (on_locc/on_hicc)
717 - added support for loop_mode=one_shot
718 - fixed playback of 16 bit wav files on big endian CPUs
719 - added support for Ogg Vorbis sample files
720 - use loop markers from sample file if loop_start and loop_end are
721 not set in sfz file
722 - implemeted filters. Filter types: lowpass, bandpass, bandreject
723 and highpass. 1, 2, 4 and 6 pole filters. Opcodes: fil_type,
724 cutoff, resonance, fil_veltrack, fil_keytrack, fil_keycenter,
725 cutoff_cc, cutoff_chanaft.
726 - bugfix: zero ampeg_sustain didn't work
727 - added support for velocity effect on amplifier envelope time
728 (ampeg_vel2attack, ampeg_vel2decay, ampeg_vel2sustain and
729 ampeg_vel2release)
730 - added support for EG hold (ampeg_hold)
731 - added support for sample offset (offset)
732 - use the newly introduced signal units model
733 - implemented opcodes lfoN_delay,
734 lfoN_freq, lfoN_pan, lfoN_cutoff, lfoN_resonance
735 - implemented opcodes ampeg_delay, ampeg_vel2delay
736 pitcheg_delay, pitcheg_start, pitcheg_attack, pitcheg_hold,
737 pitcheg_decay, pitcheg_sustain, pitcheg_release, pitcheg_vel2delay,
738 pitcheg_vel2attack, pitcheg_vel2hold, pitcheg_vel2decay,
739 pitcheg_vel2sustain, pitcheg_vel2release, pitcheg_depth
740 - implemented opcodes pitchlfo_delay, pitchlfo_freq,
741 pitchlfo_depth, fillfo_delay, fillfo_freq, fillfo_depth,
742 amplfo_delay, amplfo_freq, amplfo_depth
743 - implemented opcodes fileg_delay, fileg_start, fileg_attack,
744 fileg_hold, fileg_decay, fileg_sustain, fileg_release, fileg_vel2delay,
745 fileg_vel2attack, fileg_vel2hold, fileg_vel2decay, fileg_vel2sustain,
746 fileg_vel2release
747 - implemented opcode lfoN_wave
748 - implemented opcode pitchlfo_depthccN
749 - implemented opcodes lfoN_phase, lfoN_phase_onccX,
750 lfoN_pitch, lfoN_pitch_onccX
751 - implemented opcodes pitchlfo_fade,
752 fillfo_fade, amplfo_fade, lfoN_fade, lfoN_fade_onccX
753 - implemented opcodes fillfo_freqccN,
754 pitchlfo_freqccN, amplfo_freqccN, lfoN_freq_onccX
755 - implemented opcodes ampeg_delayccN, ampeg_startccN,
756 ampeg_attackccN, ampeg_holdccN, ampeg_decayccN, ampeg_sustainccN,
757 ampeg_releaseccN, egN_timeX_onccY, egN_levelX_onccY
758 - lfoN_* and egN_* opcodes defined in group sections
759 are now taken into account
760 - implemented curves
761 - implemented opcodes volume_onccN, volume_curveccN
762 - implemented opcode volume_smoothccN
763 - implemented opcodes fillfo_depthccN, amplfo_depthccN,
764 lfoN_volume, lfoN_volume_onccX, lfoN_volume_smoothccX,
765 lfoN_freq_smoothccX, lfoN_pitch_smoothccX, lfoN_pan_onccX,
766 lfoN_pan_smoothccX, lfoN_cutoff_onccX, lfoN_cutoff_smoothccX,
767 lfoN_resonance_onccX, lfoN_resonance_smoothccX, lfoN_delay_onccX
768 - implemented opcode end
769 - implemented opcodes egN_amplitude_onccX,
770 egN_volume, egN_volume_onccX, egN_cutoff, egN_cutoff_onccX,
771 egN_pitch, egN_pitch_onccX, egN_resonance, egN_resonance_onccX
772 - implemented opcodes xfin_lokey, xfin_hikey,
773 xfout_lokey, xfout_hikey, xf_keycurve, xfin_lovel, xfin_hivel,
774 xfout_lovel, xfout_hivel, xf_velcurve, xfin_loccN, xfin_hiccN,
775 xfout_loccN, xfout_hiccN, xf_cccurve
776 - implemented opcodes pan_onccN, pan_smoothccN,
777 pan_curveccN, egN_pan, egN_pan_curve, egN_pan_onccX,
778 egN_pan_curveccX, pitch_veltrack
779 - when failed to parse a sfz file
780 print the line number on which the error occurs
781 - use common pool of CC objects to minimize RAM usage
782 - implemented opcodes amplfo_delay_onccN,
783 amplfo_fade_onccN, fillfo_delay_onccN, fillfo_fade_onccN,
784 pitchlfo_delay_onccN, pitchlfo_fade_onccN
785 - implemented opcodes fileg_delay_onccN,
786 fileg_start_onccN, fileg_attack_onccN, fileg_hold_onccN,
787 fileg_decay_onccN, fileg_sustain_onccN, fileg_release_onccN,
788 fileg_depth_onccN, pitcheg_delay_onccN, pitcheg_start_onccN,
789 pitcheg_attack_onccN, pitcheg_hold_onccN, pitcheg_decay_onccN,
790 pitcheg_sustain_onccN, pitcheg_release_onccN, pitcheg_depth_onccN
791 - implemented automatic aliasing *ccN <-> *_onccN
792 - *lfo_freqccN wasn't working when the respective *lfo_freq
793 was not set or was set to zero
794 - lfoN_freq_onccX wasn't working when lfoN_freq
795 was not set or was set to zero
796 - implemented opcodes resonance_onccN, resonance_smoothccN,
797 resonance_curveccN, cutoff_smoothccN, cutoff_curveccN
798 - implemented opcodes fillfo_depthchanaft,
799 fillfo_freqchanaft, amplfo_depthchanaft, amplfo_freqchanaft,
800 pitchlfo_depthchanaft, pitchlfo_freqchanaft
801 - implemented opcodes pitch_onccN,
802 pitch_curveccN, pitch_smoothccN, pitch_stepccN
803 - implemented opcodes volume_stepccN, pan_stepccN,
804 cutoff_stepccN, resonance_stepccN, lfoN_freq_stepccX,
805 lfoN_volume_stepccX, lfoN_pitch_stepccX, lfoN_pan_stepccX,
806 lfoN_cutoff_stepccX, lfoN_resonance_stepccX
807 - implemented opcodes eq1_freq, eq2_freq, eq3_freq,
808 eq1_freqccN, eq2_freqccN, eq3_freqccN, eq1_bw, eq2_bw, eq3_bw,
809 eq1_bwccN, eq2_bwccN, eq3_bwccN, eq1_gain, eq2_gain, eq3_gain,
810 eq1_gainccN, eq2_gainccN, eq3_gainccN
811 - implemented opcodes delay, delay_onccN, delay_random,
812 delay_samples, delay_samples_onccN
813 - implemented opcodes egN_eq1gain, egN_eq2gain, egN_eq3gain,
814 egN_eq1gain_onccX, egN_eq2gain_onccX, egN_eq3gain_onccX, egN_eq1freq,
815 egN_eq2freq, egN_eq3freq, egN_eq1freq_onccX, egN_eq2freq_onccX,
816 egN_eq3freq_onccX, egN_eq1bw, egN_eq2bw, egN_eq3bw, egN_eq1bw_onccX,
817 egN_eq2bw_onccX, egN_eq3bw_onccX, lfoN_eq1gain, lfoN_eq2gain,
818 lfoN_eq3gain, lfoN_eq1gain_onccX, lfoN_eq2gain_onccX, lfoN_eq3gain_onccX,
819 lfoN_eq1gain_smoothccX, lfoN_eq2gain_smoothccX, lfoN_eq3gain_smoothccX,
820 lfoN_eq1gain_stepccX, lfoN_eq2gain_stepccX, lfoN_eq3gain_stepccX,
821 lfoN_eq1freq, lfoN_eq2freq, lfoN_eq3freq, lfoN_eq1freq_onccX,
822 lfoN_eq2freq_onccX, lfoN_eq3freq_onccX, lfoN_eq1freq_smoothccX,
823 lfoN_eq2freq_smoothccX, lfoN_eq3freq_smoothccX, lfoN_eq1freq_stepccX,
824 lfoN_eq2freq_stepccX, lfoN_eq3freq_stepccX, lfoN_eq1bw, lfoN_eq2bw,
825 lfoN_eq3bw, lfoN_eq1bw_onccX, lfoN_eq2bw_onccX, lfoN_eq3bw_onccX,
826 lfoN_eq1bw_smoothccX, lfoN_eq2bw_smoothccX, lfoN_eq3bw_smoothccX,
827 lfoN_eq1bw_stepccX, lfoN_eq2bw_stepccX, lfoN_eq3bw_stepccX
828 - implemented opcodes eq1_vel2freq, eq2_vel2freq,
829 eq3_vel2freq, eq1_vel2gain, eq2_vel2gain, eq3_vel2gain
830 - sfz parser: allow double spaces in sample filenames
831 - sfz parser: allow absolute paths for sample filenames
832 - use linear decay and release for filter and pitch EG
833 - bugfix: only the first amp_veltrack definition in a file was
834 used
835 - bugfix: looping was disabled if loop_start was set to 0
836 - allow regions with end=-1 to turn off other regions using the
837 group and off_by opcodes (#168)
838 - made end=0 play the whole sample
839 - fixed support for lochan and hichan opcodes (#155)
840 - fixed crash when using lochan/hichan opcodes (#187)
841 - sfz parser: allow -200 to 200 for pan_oncc opcode (#182)
842 - added FLAC support (#191)
843 - sfz parser bugfix: lines starting with whitespace were ignored
844 - added amplitude opcode
845 - added support for "#include" instruction
846 (modified patch which was originally posted by Sergey on LS mailing list)
847 - bugfix: generation of velocity curves etc should not be done
848 after each "#include", only after the main file is parsed
849 - bugfix: line numbers in error messages were wrong after "#include"
850 - added support for float and 32 bit sample files
851
852 * SoundFont format engine:
853 - Initial implementation (not usable yet)
854 - RT-safeness: avoid malloc in audio thread
855 - fixed a bug that could cause voice stealing to fail
856 - fine-tuned amplitude EG (by switching from gig to sfz EG)
857 - initial implementation of Vibrato LFO and Modulation LFO
858 - initial implementation of cutoff filter
859 - use linear decay and release for filter and pitch EG
860
861 * Host plugins (VST, AU, LV2, DSSI):
862 - AU bugfix: failed to destroy its audio/MIDI devices
863 - Listen to all interfaces on Mac OS X (INADDR_ANY)
864 - VST bugfix: If the host called resume() before and after
865 changing sample rate or block size, the number of channels was
866 incorrectly set to two. This caused silence in Cubase 5.
867 - save engine type (gig, sfz or sf2) in plugin state
868 - VST: when opening Fantasia, look for both 32 and 64 bit Java on
869 64 bit Windows
870 - AU: changed number of output channels from one stereo to 16
871 stereo
872 - VST: made it possible to build the VST plugin for Mac
873 - AU: link AU plugin dynamically if --disable-shared isn't
874 specified
875 - LV2 "state" extension support (patch by David Robillard)
876 - VST bugfix: instrument loading hang and crashed the host when
877 the plugin was loaded a second time (#174)
878 - plugin bugfix: instrument loading hang when the plugin was
879 loaded a second time (this time it's for Linux and Mac, previous
880 similar fix was for Windows)
881 - thread safety fixes for the instrument loading thread
882 - LV2: use the new lv2 package if present
883 - VST: try to open Fantasia automatically on Linux and Mac too (on
884 Linux, the Fantasia jar should be placed in <prefix>/share/java)
885 - VST: fixed crashes on Linux Ardour and EnergyXT
886 - DSSI bugfix: it wasn't possible to change engine type. The MIDI
887 port and audio channel routing for DSSI plugins are now visible.
888 - LV2: use urid and atom extensions instead of deprecated uri-map
889 and event
890 - LV2: lv2 package 1.0 is now required to build the LV2 plugin
891 - LV2: changed number of output channels to 16 stereo, just like
892 the VST and AU plugins
893 - LV2: fixed save/restore of SFZ state (patch by David Robillard)
894 - LV2: made LV2 plugin buildable on Windows and Mac
895 - VST: implemented retrieval and switching of programs using the
896 sampler's internal MIDI instrument mapping system
897
898 * Instrument editor interface:
899 - Changed instrument editor plugin interface, providing additional
900 informations like the EngineChannel for which the instrument editor was
901 spawned for. This allows the instrument editors to interact more actively
902 with the sampler.
903
904 * MIDI driver:
905 - ALSA MIDI driver supports now "NAME" device parameter, for overriding
906 the ALSA sequencer client name
907 - removed limit of maximum amount of MIDI ports per MIDI device, since
908 there is no reason for this limit
909 - MME: fixed memory handling bug found with cppcheck
910 - MME: removed compiler warning
911 - CoreMIDI: implemented driver specific port parameter "CORE_MIDI_BINDINGS",
912 which allows to retrieve the list of CoreMIDI clients / ports and to connect
913 to them a la JACK, via the usual sampler APIs
914 - CoreMIDI: added driver specific port parameter "AUTO_BIND", if enabled
915 the driver will automatically connect to other CoreMIDI clients' ports
916 (e.g. external MIDI devices being attached to the Mac)
917 - added support for MIDI note on velocity filter
918 - CoreMIDI: fixed memory deallocation error
919 - Fixed variable underflow in VirtualMidiDevice, which caused graphical
920 virtual keyboards in frontends / instrument editors being stuck.
921 - Bugfix in VirtualMidiDevice: process note on with velocity 0 as note off.
922 - Implemented missing handling of MIDI "running status".
923 - CoreMIDI fix: a MIDIPacket can contain more than one event per packet.
924 - MME bugfix: driver wasn't closed properly
925 - CoreMIDI: automatically connect to all input sources by default (driver
926 parameter "AUTO_BIND").
927 - CoreMIDI: fixed auto bind feature to CoreMIDI ports that go online
928 - CoreMIDI: fixed minor memory leak
929
930 * audio driver:
931 - ASIO driver fixes for newer gcc versions (fix from PortAudio)
932 - JACK audio: react on sample rate changes.
933 - JACK audio: react on buffer size changes.
934 - JACK audio: jack_port_get_buffer() was cached and called outside
935 RT context.
936 - ASIO driver: removed compiler warnings
937 - CoreAudio: fixed minor error handling bug
938 - ASIO driver: be more verbose when no ASIO card could be found (fixes #203)
939 - JACK audio: return the JACK server's current sample rate as default value
940 for audio device parameter "SAMPLERATE" (fixes #166).
941
942 * LSCP server:
943 - added support for sending MIDI CC messages via LSCP command
944 "SEND CHANNEL MIDI_DATA CC <sampler-chan> <ctrl> <val>"
945 - added LSCP command "GET AVAILABLE_EFFECTS"
946 - added LSCP command "LIST AVAILABLE_EFFECTS"
947 - added LSCP command "GET EFFECT INFO <effect-index>"
948 - added LSCP command "CREATE EFFECT_INSTANCE <effect-index>"
949 - added LSCP command
950 "CREATE EFFECT_INSTANCE <effect-system> <module> <effect-name>"
951 - added LSCP command "DESTROY EFFECT_INSTANCE <effect-instance>"
952 - added LSCP command "GET EFFECT_INSTANCES"
953 - added LSCP command "LIST EFFECT_INSTANCES"
954 - added LSCP command "GET EFFECT_INSTANCE INFO <effect-instance>"
955 - added LSCP command
956 "GET EFFECT_INSTANCE_INPUT_CONTROL INFO <effect-instance> <input-control>"
957 - added LSCP command "SET EFFECT_INSTANCE_INPUT_CONTROL VALUE
958 <effect-instance> <input-control> <value>"
959 - added LSCP command "GET SEND_EFFECT_CHAINS <audio-device>"
960 - added LSCP command "LIST SEND_EFFECT_CHAINS <audio-device>"
961 - added LSCP command "ADD SEND_EFFECT_CHAIN <audio-device>"
962 - added LSCP command
963 "REMOVE SEND_EFFECT_CHAIN <audio-device> <effect-chain>"
964 - added LSCP command
965 "GET SEND_EFFECT_CHAIN INFO <audio-device> <effect-chain>"
966 - added LSCP command "APPEND SEND_EFFECT_CHAIN EFFECT <audio-device>
967 <effect-chain> <effect-instance>"
968 - added LSCP command "INSERT SEND_EFFECT_CHAIN EFFECT <audio-device>
969 <effect-chain> <effect-chain-pos> <effect-instance>"
970 - added LSCP command "REMOVE SEND_EFFECT_CHAIN EFFECT <audio-device>
971 <effect-chain> <chain-pos>"
972 - added LSCP command "SET FX_SEND EFFECT <sampler_channel>
973 <fx_send_id> <effect_chain> <chain_pos>"
974 - added LSCP command "REMOVE FX_SEND EFFECT <sampler_channel> <fx_send_id>"
975 - added LSCP commands "SUBSCRIBE EFFECT_INSTANCE_COUNT",
976 "SUBSCRIBE EFFECT_INSTANCE_INFO", "SUBSCRIBE SEND_EFFECT_CHAIN_COUNT",
977 "SUBSCRIBE SEND_EFFECT_CHAIN_INFO"
978 - provide comprehensive error messages on LSCP syntax errors
979 (suggesting expected next non-terminal symbols)
980 - Fixed client connection not being closed after network errors.
981
982 * LSCP shell:
983 - Added support for auto correction of obvious and trivial syntax mistakes.
984 - Added support for auto completion by tab key.
985 - Show currently available auto completion while typing.
986 - Added support for browsing command history with up / down keys.
987 - Show all possible next symbols immediately right to the current command
988 line while typing (no double tab required for this feature, as it would
989 be the case in other shells).
990 - Added support for moving cursor left/right with arrow keys.
991 - Added support for built-in LSCP reference documentation, which will
992 automatically show the relevant LSCP reference section on screen as soon
993 as one specific LSCP command was detected while typing on the command
994 line.
995
996 * Real-time instrument scripts:
997 - Implemented built-in script array variable %CC.
998 - Implemented built-in script int variable $CC_NUM.
999 - Implemented built-in script int variable $EVENT_NOTE.
1000 - Implemented built-in script int variable $EVENT_VELOCITY.
1001 - Implemented built-in script constant variable $VCC_MONO_AT.
1002 - Implemented built-in script constant variable $VCC_PITCH_BEND.
1003 - Implemented execution of script event handler "init".
1004 - Implemented execution of script event handler "controller".
1005 - Implemented execution of script event handler "note".
1006 - Implemented execution of script event handler "release".
1007 - Implemented built-in script function "play_note()" (only two of the
1008 max. four function arguments are currently implemented yet though).
1009 - Implemented built-in script int variable $EVENT_ID.
1010 - Implemented built-in script function "ignore_event()"
1011 - Implemented built-in script function "ignore_controller()" (may have one
1012 or no argument).
1013 - Implemented built-in script function "set_controller()".
1014 - Added extended script VM for the Gigasampler/GigaStudio format sampler
1015 engine, which extends the general instrument script VM with Giga format
1016 specific variables and functions.
1017 - Giga format scripts: added built-in script int constant variables
1018 $GIG_DIM_CHANNEL, $GIG_DIM_LAYER, $GIG_DIM_VELOCITY, $GIG_DIM_AFTERTOUCH,
1019 $GIG_DIM_RELEASE, $GIG_DIM_KEYBOARD, $GIG_DIM_ROUNDROBIN, $GIG_DIM_RANDOM,
1020 $GIG_DIM_SMARTMIDI, $GIG_DIM_ROUNDROBINKEY, $GIG_DIM_MODWHEEL,
1021 $GIG_DIM_BREATH, $GIG_DIM_FOOT, $GIG_DIM_PORTAMENTOTIME, $GIG_DIM_EFFECT1,
1022 $GIG_DIM_EFFECT2, $GIG_DIM_GENPURPOSE1, $GIG_DIM_GENPURPOSE2,
1023 $GIG_DIM_GENPURPOSE3, $GIG_DIM_GENPURPOSE4, $GIG_DIM_SUSTAIN,
1024 $GIG_DIM_PORTAMENTO, $GIG_DIM_SOSTENUTO, $GIG_DIM_SOFT,
1025 $GIG_DIM_GENPURPOSE5, $GIG_DIM_GENPURPOSE6, $GIG_DIM_GENPURPOSE7,
1026 $GIG_DIM_GENPURPOSE8, $GIG_DIM_EFFECT1DEPTH, $GIG_DIM_EFFECT2DEPTH,
1027 $GIG_DIM_EFFECT3DEPTH, $GIG_DIM_EFFECT4DEPTH, $GIG_DIM_EFFECT5DEPTH.
1028 - Giga format scripts: Implemented built-in script function
1029 "gig_set_dim_zone(event_id, dimension, zone)".
1030 - Implemented built-in script int array variable %KEY_DOWN.
1031 - Implemented built-in script function "abs()".
1032 - Implemented built-in script function "random()".
1033 - Implemented built-in script function "num_elements()".
1034 - Implemented built-in script function "note_off()".
1035 - Implemented built-in script function "set_event_mark()".
1036 - Implemented built-in script function "delete_event_mark()".
1037 - Implemented built-in script function "by_marks()".
1038 - Added built-in script int const variables $MARK_1 to $MARK_28.
1039 - Built-in script functions "ignore_event()", "note_off()" and
1040 "gig_set_dim_zone()" now also accept an array of event IDs as argument
1041 (i.e. return value of new script function "by_marks()").
1042 - Pass/preserve polyphonic variable data from respective "note" event
1043 handler to "release" event handler.
1044 - Fixed crash when using built-in script function "by_marks()".
1045
1046 * Bug fixes:
1047 - Fixed crash which may occur when MIDI key + transpose is out of range
1048 - minor valgrind fixes
1049 - fixed crash which occurred when changing an already deployed sampler
1050 channel to a different engine type
1051 - fixed crash when deleting a sampler channel or changing engine
1052 type while an instrument load was in progress
1053 - bugfix: playing a note while changing the instrument could cause
1054 a crash, or give "Handing back unknown region" error messages
1055 - bugfix: calling SET VOICES (which Fantasia does on start and
1056 refresh) could cause instruments to be unloaded
1057 - Mac OS X: fixed name collision of enum in EffectControl and
1058 wrong name of destructor in AudioOutputDeviceCoreAudio.cpp
1059 - Mac OS X: fixed hanging threads
1060 - Windows: disabled the previous thread fix on non-Mac systems, as
1061 it caused hanging threads on Windows
1062 - Fixed possible crashes due to corrupted MIDI/audio device list
1063 after MIDI/audio device creation failure
1064 - When creating MIDI instrument map entries with "PERSISTENT" type, the
1065 instruments were uselessly precached with zero samples, however it still
1066 took the full preloading time and on 1st program change the respective
1067 instrument was completely reloaded again.
1068 - fixed handling of rapid bank select and program change messages
1069 sent to the same sampler channel (patch from the Open Octave
1070 project, slightly adjusted)
1071 - fixed crash when trying to create an effect instance with controls
1072 which min and/or max values depend on the sample rate
1073 - fixed bug #162
1074 - bugfix: LADSPA_PATH was not evaluated correctly when containing
1075 multiple paths (#165)
1076 - thread safety fixes for the instrument loading thread
1077 - bugfix: instrument loading crashed for sfz and sf2 in Ardour
1078 (#176)
1079 - more thread safety fixes for the instrument loading thread
1080 - sfz/sf2 engine: fixed crash when using small audio fragment size
1081 - Mac OS X: fixed crash when unloading plugin on 10.7 and later
1082 - Mac OS X: fixed process hang when unloading 32-bit plugin (bug
1083 introduced in previous fix)
1084 - fixed crash when a channel received a program change while
1085 playing a note in a key group
1086 - fixed erroneous error message piping in VoiceBase.h
1087 (labelled "Disk stream not available in time")
1088 - Update effects on sample rate & period size changes (to avoid
1089 crashes and noise on such transitions).
1090 - ignore missing LADSPA paths without ignoring valid LADSPA paths
1091 (fixes #208)
1092 - Fixed bug in LSCP grammar definition which caused a statement like
1093 "GET SERVER INFOasdf\n" to be accepted as valid statement (was so far
1094 practically irrelevant, however it caused problems with the new LSCP
1095 shell's auto completion feature).
1096 - Fixed MIDI program change messages being ignored if quickly executed
1097 after each other (fixes #231).
1098
1099 Version 1.0.0 (31 July 2009)
1100
1101 * packaging changes:
1102 - autoconf bugfix: the PKG_CONFIG variable wasn't initialized properly,
1103 causing e.g. the libgig test to fail when
1104 "./configure --disable-jack-driver" was used
1105 (patch by Alexis Ballier)
1106 - fixed compilation with gcc 4.3
1107 - fixes for building on OS X (thanks to Ebrahim Mayat for testing)
1108 - fixed configure so it detects x86_64 (#107)
1109 - fixes for building with newer MinGW versions
1110 - fix for building with bison 2.4 (#111)
1111 - fixed building with libgig installed in a non-standard directory
1112 - minor fix in configure for mmsystem.h detection on MinGW
1113 - Windows: look for editor plugins and Fantasia using base
1114 directory of liblinuxsampler dll (look in the same directory and one
1115 directory above)
1116 - configure script fix: removed unconditional use of SSE
1117 - fixed building with sqlite installed in a non-standard directory
1118 - when cross-compiling, don't try to create instruments.db
1119 - fix for new mingw-w64 version, which has usleep
1120
1121 * general changes:
1122 - bugfix: on some POSIX systems instrument editor plugins refused to
1123 load as we used a non-portable Linux specific struct field
1124 (fixes bug #70, patch by Ronald Baljeu)
1125 - fixed endless loop which occured when loading an instrument editor
1126 plugin DLL on Windows systems failed
1127 - fixed memory leaks that occurred when liblinuxsampler was unloaded
1128 - optimized the SynchronizedConfig class so it doesn't wait
1129 unnecessarily long after an update
1130 - added support for notifying instrument editors on note-on / note-off
1131 events (e.g. to highlight the pressed keys on the virtual keyboard
1132 of gigedit)
1133 - added support for triggering notes by instrument editors (see above)
1134 - be verbose on DLL load errors (on Linux)
1135 - fixed di-harmonic triangle LFO implementation (this LFO implementation
1136 is only used by very few systems, most pick the int math
1137 implementation, which is usually faster)
1138 - fixes for audio drivers with varying buffer sizes
1139 - experimental support for running LinuxSampler as a DSSI, LV2 and
1140 VST plugin
1141 - notification events for stream/voice count statistics are now sent
1142 only when there are actual changes
1143 - added memory ordering constraints to improve stability on
1144 multi-core and multi-cpu systems
1145 - maximum voices and disk streams can now be altered at runtime
1146 - fixed CPU feature detection on x86_64 (maybe fixes #108)
1147 - automatic stacktrace mechanism is now turned off by default and can
1148 be switched on by command line option "--stacktrace" (the automatic
1149 stacktrace mechanism seems to be broken on most systems at the moment)
1150 - C++ API method InstrumentManager::LaunchInstrumentEditor() now returns
1151 a pointer to the launched InstrumentEditor object
1152 - added optional 3rd party user data parameter for following
1153 liblinuxsampler C++ API methods: InstrumentEditor::Main(),
1154 InstrumentEditor::Launch(),
1155 InstrumentManager::LaunchInstrumentEditor()
1156 - theoretical fix: made SynchronizedConfig follow C++0x memory
1157 model more strictly
1158 - fixes for using large audio device buffers
1159 - Windows: add the installation directory to the DLL search path
1160 when loading an editor plugin (solves problems with VST and
1161 gigedit on systems with other GTK versions installed)
1162 - updated linuxsampler man page
1163
1164 * audio driver:
1165 - removed the nonsense audio channel constraint (which was hard coded to
1166 max. 100 audio channels) for most audio drivers
1167 - JACK audio driver did not offer a device parameter "SAMPLERATE" as
1168 opposed to the LSCP specs
1169 - bugfix: the SAMPLERATE parameter of some drivers (e.g. JACK)
1170 reflected the wrong value
1171 - fixed a memory management bug in ASIO driver
1172 - Makefile fix: JACK_CFLAGS wasn't used
1173 - JACK: use jack_client_open instead of the deprecated
1174 jack_client_new
1175 - added (experimental) CoreAudio driver
1176 - applied old fixes to the ASIO driver that were included in the
1177 previous binary release but accidentally never committed to CVS
1178 (fixes #117)
1179 - fixes for ASIO on mingw-w64 (iasio wrapper is not needed on
1180 win64)
1181 - VST: added support for sample rate and buffer size changes
1182 - VST: close editor (Fantasia) when the VST is removed
1183 - VST: avoid opening Fantasia more than once for each VST instance
1184 - VST: export main function as "main" on Linux too (fix for
1185 energyXT)
1186 - VST: changed number of output channels from one stereo to 16
1187 stereo
1188 - added channel routing, fxsends and midi maps to the settings
1189 stored in the plugin state
1190 - performance optimization of AudioChannel::MixTo() and
1191 AudioChannel::CopyTo() methods using GCC vector exensions
1192 (if available)
1193 - ASIO fixes: avoid initializing the device twice, avoid throwing
1194 exception when getting parameters from a disconnected device
1195
1196 * MIDI driver:
1197 - added JACK MIDI driver
1198 - dispatch bank select as ordinary CC as well, the user might seriously
1199 want to (mis)use it for some purpose ("fixed" in all current MIDI
1200 input drivers: ALSA, CoreMIDI, JACK, MidiShare, MME)
1201 - bugfix: pitch bend wasn't working with JackMidi, VST, LV2, MME,
1202 CoreMidi or AU
1203 - fixed mingw-w64 compilation error in MME driver
1204 - made program change handling in MIDI thread real-time safe by
1205 moving the logic to a non-RT thread
1206 - fixed minor memory leak in ALSA driver
1207
1208 * instruments database:
1209 - avoid time consuming samples scanning when adding instruments
1210 to the instruments database
1211 - added support for handling lost files in the instruments database
1212 - Implemented option for adding instruments in separate directories
1213 in the instruments database
1214 (patch by Chris Cherrett & Andrew Williams, a bit adjusted)
1215 - work-around for missing fnmatch function on Windows to make
1216 instrument database compilable
1217 - added instrument database support on Windows
1218 works with both standalone mode and VST plugin.
1219 the instruments DB file is located in
1220 %USERPROFILE%\.linuxsampler\instruments.db which allows different
1221 databases for each windows user
1222 if no DB is present it automatically creates the .linuxsampler subdir
1223 and then creates an empty DB
1224 - fixed recursive import, which was broken on Windows
1225
1226 * Gigasampler format engine:
1227 - fixed a memory leak that could happen when a channel was deleted
1228 while notes were playing
1229 - made it possible to load an instrument even if the audio thread
1230 isn't running
1231 - added partial support for the "Controller Triggered" MIDI rule,
1232 enough for piano gigs with pedal noise samples
1233 - minor fix: only mark FX sends as being modified if really the
1234 respective FX send MIDI controller was used
1235 - added support for GM global device volume SysEx message
1236 (can be selected at compile time to either apply globally to the
1237 whole sampler [default] or only to the sampler channels that are
1238 connected to the respective MIDI input port the SysEx message arrived
1239 on)
1240 - bugfix: notes triggered at position 0 in the audio buffer were
1241 sometimes wrongly killed in the same buffer, causing no sound to
1242 be played
1243 - added support for chromatic / drumkit mode Roland GS Sysex message
1244 which is usally used to switch a part between chromatic sounds and
1245 drumkit sounds (as opposed to their common default setting of MIDI
1246 part 10 being a drumkit part and all other ones chromatic parts), in
1247 LS however one can switch between the first 16 MIDI instrument maps
1248 defined for the sampler
1249 - made it possible to create multiple sample channels even if the
1250 audio thread isn't running
1251 - fixed crash when removing channel with active voices (#116)
1252 - bugfix: on sample reference changes (instrument editor), only
1253 un-cache the respective sample if it's really not used by any
1254 sampler engine anymore
1255 - re-cache samples in case they were changed by an instrument editor,
1256 e.g. when a sample was added while playing (#82)
1257 - fixed hanging notes which occured when note-off event had the exact
1258 same time stamp as the voice's note-on event and occured both in the
1259 same audio fragment cycle (fixes bug #112)
1260 - added support for the "fine tune" and "pitch bend range"
1261 instrument-level gig parameters
1262 - fixed minor artifacts in pitch bend handling
1263 - added support for GS Reset SysEx message
1264 - allow gig files to use unlimited downward pitch shifting
1265 - added a limit check for upward pitch shifting
1266 - bugfix: sometimes, when playing a note twice fast, the second
1267 note was silent
1268 - fixed crash happening when a pitch bend event arrived at the
1269 same time a new instrument was loading
1270
1271 * LSCP server:
1272 - added new LSCP event "CHANNEL_MIDI" which can be used by frontends to
1273 react on MIDI data arriving on certain sampler channels (so far only
1274 Note-On and Note-Off events are sent via this LSCP event)
1275 - added new LSCP event "DEVICE_MIDI" which can be used by frontends to
1276 react on MIDI data arriving on certain MIDI input devices (so far only
1277 Note-On and Note-Off events are sent via this LSCP event)
1278 - added new LSCP commands: FIND LOST DB_INSTRUMENT_FILES and
1279 SET DB_INSTRUMENT FILE_PATH
1280 - added new LSCP command: SEND CHANNEL MIDI_DATA which can be used by
1281 frontends to send MIDI messages to specific sampler channel
1282 - added two additional fields to GET FILE INSTRUMENT INFO command -
1283 KEY_BINDINGS and KEYSWITCH_BINDINGS
1284 - bugfix: the bank number provided by MIDI_INSTRUMENT_INFO notifications
1285 was incorrect
1286 - Optimized the retrieval of the MIDI instrument mappings
1287 - added new LSCP command "SET VOICES" to globally alter the maximum
1288 amount of voices, added new LSCP event "GLOBAL_INFO:VOICES" which will
1289 be triggered respectively
1290 - added new LSCP command "SET STREAMS" to globally alter the maximum
1291 amount of disk streams, added new LSCP event "GLOBAL_INFO:STREAMS"
1292 which will be triggered respectively
1293 - bugfix: retry if "select" returns EINTR (this fixes a crash when
1294 a gigedit file dialog is opened)
1295 - close all connections when LSCPServer is deleted
1296 - hard close of all sockets on exit
1297 - bugfix: SET CHANNEL MIDI_INPUT_TYPE didn't work with the MME
1298 driver
1299
1300 * Bug fixes:
1301 - fixed a crash which occurs when removing a sampler channel waiting
1302 to start instrument loading after another channel
1303 - fixed a crash which occurs when removing a sampler channel with
1304 instrument loading in progress (bug #113)
1305 - fixed termination caused by uncaught exception when adding MIDI
1306 instrument with PERSISTENT load mode
1307 - fixed possible iterator invalidations when resetting the sampler
1308 - fixed memory leaks when issuing the following LSCP commands:
1309 GET AUDIO_OUTPUT_DRIVER INFO
1310 GET AUDIO_OUTPUT_DRIVER_PARAMETER INFO
1311 GET MIDI_INPUT_DRIVER INFO
1312 GET MIDI_INPUT_DRIVER_PARAMETER INFO
1313 - fixed possible compilation error when sqlite is not present
1314 - fixed orphaned pointers when setting maximum voices limit (bug #118)
1315 - fixed crash when changing the audio output device of a sampler
1316 channel with loaded instrument and start playing notes
1317 - fixed endless loop in Engine::SuspendAll() (bug #120)
1318 - fixed a low-level atomic load function that was broken on 64-bit
1319 PowerPC, which probably could cause crashes on that platform
1320 - fixed a memory management error which could cause a crash when a
1321 plugin was unloaded
1322 - bugfix: two private structs had the same name, which could cause
1323 problems if the linker chose the wrong constructor
1324 - fixed low-level ConditionServer usage bug that caused lockups on
1325 Windows
1326
1327
1328 Version 0.5.1 (6 December 2007)
1329
1330 * packaging changes:
1331 - added autoconf checks for pthread library
1332 - added autoconf check for pthread bug found on certain NPTL-enabled
1333 glibc versions (see Gentoo bug report #194076)
1334 - added autoconf checks for MS Windows
1335
1336 * general changes:
1337 - bugfix: the thread used by an editor plugin didn't die when the
1338 editor closed
1339 - bugfix: Ringbuffer.h: fill_write_space_with_null() did not zero
1340 out all the space. operator--() did not apply size_mask after
1341 decrementing the read_ptr. DEFAULT_WRAP_ELEMENTS set to 0 in
1342 order to avoid problems with the _NonVolatileReader functions.
1343 - bugfix: Stream.h: added a missing divide by BytesPerSample in
1344 GetWriteSpace(). Since this function is currently only used in
1345 the stream's qsort() compare function, it didn't trigger any
1346 bugs.
1347 - Resampler.h, Synthesizer.h: cubic interpolation now works in
1348 24bit mode too. Faster method to read 24bit words on little
1349 endian machines (x86): replaced 3 byte reads + shifts with a 1
1350 unaligned 32bit read and shift
1351 - experimental support for MS Windows (MIDI input via MME, AUDIO
1352 output via ASIO)
1353 - made handling of SIGINT signal (Ctrl-C) a bit more robust
1354 - support for monitoring the total number of active disk streams
1355 (new LSCP commands: GET TOTAL_STREAM_COUNT,
1356 SUBSCRIBE TOTAL_STREAM_COUNT, UNSUBSCRIBE TOTAL_STREAM_COUNT)
1357
1358 * AUDIO driver:
1359 - added Windows ASIO low latency audio driver
1360
1361 * MIDI driver:
1362 - added MME Windows MIDI driver
1363
1364 * LSCP server:
1365 - added support for Windows style path / filenames, however with
1366 forward slash path separators instead of backslash
1367 (i.e. "C:/foo/bar.gig")
1368 - allow naughty liblscp to send non-string device parameters within
1369 apostrophes as well
1370 - added new LSCP commands: "GET FILE INSTRUMENTS <file>",
1371 "LIST FILE INSTRUMENTS <file>" and
1372 "GET FILE INSTRUMENT INFO <file> <index>" for retrieving informations
1373 about an arbitrary instrument file on the system where the sampler is
1374 running on
1375
1376 Version 0.5.0 (15 October 2007)
1377
1378 * packaging changes:
1379 - config.h is not going to be installed along with liblinuxsampler's
1380 API header files anymore
1381 - only the API relevant header (and source) files will be exported to
1382 the Doxygen API documentation (explicitly listed in Doxyfile.in)
1383 - added completely new XCode project files for Mac OSX which is now
1384 capable to execute our autoconf environment, thus no need anymore
1385 to maintain the compile time configuration file (osx/version.h) for
1386 OSX manually (patch by Toshi Nagata)
1387 - fixed buggy boolean --enable-foo / --disable-foo configure script
1388 parameters
1389 - global.h now only covers global definitions that are needed for the
1390 C++ API header files, all implementation internal global definitions
1391 are now in global_private.h
1392 - atomic.h is not exposed to the C++ API anymore
1393 - no need to include config.h anymore for using LS's API header files
1394 - fixed warnings in API doc generation
1395
1396 * general changes:
1397 - replaced the old, confusing MIDI program change mechanism by a
1398 flexible MIDI instrument mapper which allows to map arbitrary
1399 (MIDI bank, MIDI program) pairs with arbitrary
1400 (engine type, instrument file, file index) triplets which will be
1401 loaded on the respective channel when such MIDI program change
1402 messages arrive, beside that, each entry allows to define a life-time
1403 strategy for the instrument, so the user can define whether the
1404 instrument should i.e. be permanently loaded (to be able to switch
1405 quickly among sounds) or i.e. loaded just on demand when the
1406 respective program change arrives (to safe RAM space), as well as a
1407 global volume factor for each entry, so the user can adjust the volume
1408 dynamic of his mapped instrument collection without having to modify
1409 the instrument files, also one can manage arbitrary amount of such
1410 MIDI instrument maps and assign each sampler channel individually
1411 a certain map, so that i.e. separation of normal instruments and
1412 drumkits is possible
1413 - new notification events for tracking changes to audio/MIDI devices,
1414 MIDI instrument maps, MIDI instruments, FX sends, global volume.
1415 - sampler was limited to load max. 200 instruments in the background
1416 due to a constant size RingBuffer FIFO which is now replaced by a
1417 dynamic (unlimited) size std::list FIFO
1418 - added FX sends, these allow to route audio signals to arbitrary audio
1419 output channels for being processed by external effect processors
1420 (i.e. jack-rack), the send levels are controllable via arbitrary MIDI
1421 controllers
1422 - global (sampler wide) volume can now be controlled at runtime
1423 - Implemented new, improved notification system
1424 - fixed compilation errors regarding OSX
1425 (patch by Toshi Nagata)
1426 - implemented instruments database
1427 - added support for escape sequences to the instruments database
1428 - added highly experimental support for on-the-fly instrument editing
1429 within the sampler's process (by using instrument editor plugins),
1430 you'll notice the new "Registered instrument editors:" message on
1431 startup, added a new LSCP command:
1432 "EDIT CHANNEL INSTRUMENT <sampler-channel>"
1433 to spawn a matching instrument editor for the instrument on the
1434 given sampler channel, the plugin path can be overridden at compile
1435 time with ./configure --enable-plugin-dir=/some/dir
1436 - added experimental code for synchronizing instrument editors hosted
1437 in the sampler's process to safely edit instruments while playing
1438 without a crash (hopefully) by either suspending single regions
1439 wherever possible or if unavoidable whole engine(s)
1440 - fixed several issues in fundamental "Thread" class: set scheduling
1441 policy and priority on thread level, set a minimum stack size for
1442 thread (TODO: a reasonable value yet to be tested), bugfix: non-RT
1443 threads simply inherited properties of starting thread instead of
1444 setting their own policy and priority
1445 - minor fix in our automatic stack trace mechanism on crashes, the
1446 main process did not wait for the stack trace process to finish
1447 its output
1448 - fixed some minor memory leaks
1449 - reenabled assembly features support, at the moment only for
1450 enabling a fast denormal FPU mode (x86 platforms supporting SSE2)
1451 - minor assembly fix in x86 features detection (don't use the PIC
1452 register, to avoid relocations in the text segment at runtime)
1453 - POSIX callback functions of Thread.h are hidden
1454
1455 * MIDI driver:
1456 - dispatch bank select (MSB and LSB) messages
1457
1458 * audio driver:
1459 - the ALSA audio output driver parameters now reflect the correct
1460 parameter value ranges for the respective selected sound card
1461 (patch by Till Wimmer, a bit fixed and extended)
1462
1463 * Gigasampler format engine:
1464 - if a filter is used and EG2 finishes before EG1, let the voice
1465 die when EG2 has finished (this fixes a problem with clicks and
1466 voice starvation for some gigs)
1467 - playback is no longer disabled during instrument loading
1468 - all notes playing on a channel that changes its instrument keep
1469 playing with the old instrument until they get a note off
1470 command
1471 - EG fix: a release value of zero could cause noises or crash
1472 - handle MIDI coarse tuning messages (MIDI RPN #0 MSB #2 LSB)
1473 - EG fine tuning: when attack is zero the EG starts at a level
1474 above max sustain level, which means that there is a decay phase
1475 even if sustain is 100%
1476 - more EG fixes: the level could sometimes go below zero and cause
1477 noises or crashes
1478 - minor fix of EGDecay (patch by Toshi Nagata)
1479 - fixed compiler error when --enable-override-filter-type was
1480 supplied to the configure script (fixes #46)
1481 - disk thread: queue sizes are now proportional to CONFIG_MAX_STREAMS
1482 instead of a fix value
1483 - behavior fix: on MIDI CC# 65 (portamento on / off), 126 (mono mode),
1484 127 (solo mode) only kill voices if the respective mode really
1485 changed
1486
1487 * LSCP server:
1488 - fixed compile time error for old Bison versions
1489 (i.e. v1.28 found on MacOS 10.4, patch by Toshi Nagata)
1490 - parser now supports extended ASCII character set
1491 (up to ASCII code 255, i.e. includes now umlauts and accents)
1492 - filename arguments in LSCP commands now allow to use escape
1493 sequences, that is directly literal as one of: \', \", \\, \n, \r,
1494 \f, \t, \v, or as octal ASCII code value like \132, or as hex ASCII
1495 code value like \xf2) (fixes bug #24)
1496 - the following LSCP commands now also support escape sequences for at
1497 least one of their text-based parameters (i.e. name, description):
1498 "ADD MIDI_INSTRUMENT_MAP", "MAP MIDI_INSTRUMENT",
1499 "SET MIDI_INSTRUMENT_MAP NAME", "SET FX_SEND NAME", "CREATE FX_SEND",
1500 "SET DB_INSTRUMENT_DIRECTORY NAME",
1501 "SET DB_INSTRUMENT_DIRECTORY DESCRIPTION", "SET DB_INSTRUMENT NAME",
1502 "SET DB_INSTRUMENT DESCRIPTION", "FIND DB_INSTRUMENTS",
1503 "FIND DB_INSTRUMENT_DIRECTORIES"
1504 - returns verbose syntax errors (line and column where syntax error
1505 occured, the unexpected character and the actually expected, possible
1506 character(s), the latter only if less than 5 possibilities)
1507 - made sure that LSCP syntax is not affected by gigedit locale
1508 settings
1509 - bugfix regarding strings parameter lists: all comma separated lists
1510 of strings were treated as being one string containing commas
1511 (fixes #57)
1512
1513 * Bug fixes:
1514 - fixed crash occurring on certain LSCP scripts (Bug 39)
1515 - another thread safety fix for lscp "load engine" and "set
1516 channel audio output device"
1517 - fixed a crash which occurs when reassigning the same engine
1518 on a sampler channel with connected MIDI device
1519 - fixed a crash which occurs when changing the number of ports of a MIDI
1520 device connected to a sampler channel to number less then or equal
1521 to the index of the port to which the sampler channel is connected.
1522 - The previous bindings were not been disconnected when altering
1523 the ALSA_SEQ_BINDINGS parameter. Introduced a NONE keyword for
1524 unsubscribing from all bindings (e.g. ALSA_SEQ_BINDINGS=NONE).
1525 - The active stream/voice count statistic was incorrect.
1526 - notification events were not been sent for some sampler
1527 channel changes
1528 - added default min and max values to restrict the number of allowed
1529 audio output channels and MIDI input ports
1530 - the connection to the PCM interface is now closed when destroying
1531 an audio output device
1532 - files with slash in their path or filename could not be loaded
1533
1534 * test cases:
1535 - updated and fixed (haven't been touched in a while)
1536
1537 Version 0.4.0 (24 November 2006)
1538
1539 * packaging changes:
1540 - changed deprecated copyright attribute to license; added ldconfig
1541 to post-(un)install steps; added devel package for liblinuxsampler;
1542 to linuxsampler.spec (RPM)
1543 - install necessary development header files for allowing 3rd party
1544 applications to link against liblinuxsampler
1545 - liblinuxsampler's API documentation can be generated with 'make docs'
1546 (Doxygen required)
1547 - added benchmark to automatically detect the best triangle LFO
1548 implementation (currently either an integer math solution or a
1549 di-harmonic approximation), automatic detection can be overriden
1550 with --enable-signed-triang-algo=x and --enable-unsigned-triang-algo=x
1551 configure script argument though (mandatory for cross-compilation)
1552 - do not automatically pick optimized gcc flags if the user already
1553 provided some on his own (as CXXFLAGS)
1554 - added compile time option to disable processing of All-Notes-Off MIDI
1555 messages
1556 - added compile time options to allow disabling the various audio and
1557 MIDI drivers
1558 - fixed automatic GCC CPU switch detection on PPC
1559 (patch by Ebrahim Mayat)
1560
1561 * Gigasampler format engine:
1562 - extensive synthesis optimization
1563 (reimplementation of EGs and LFO(s), removed synthesis parameter
1564 prerendering and the synthesis parameter matrix in general, splitting
1565 each audio fragment into subfragments now where each subfragment uses
1566 constant pitch and filter coefficients. The volume coefficient is
1567 linearly interpolated inside a subfragment, unless
1568 --disable-interpolate-volume is set.)
1569 - fine tuning of the EG modulation parameters
1570 - improved filter cutoff calculation by adding support for the
1571 following gig parameters: Cutoff freq (used when no cutoff
1572 controller is defined), Control invert, Minimum cutoff, Velocity
1573 curve and Velocity range. The keyboard tracking now scales
1574 cutoff frequency instead of resonance.
1575 - added support for gig parameter Resonance.
1576 - fixed bug in sysex handling (patch by Juan Linietsky)
1577 - added global volume attenuation of -9 dB (0.35f) to prevent clipping
1578 which can be overridden with --enable-global-attenuation
1579 - EG fixes: made the length of "attack hold" stage more
1580 accurate. Release stage can now start before attack stage
1581 ends. Cancel release didn't work when sustain was zero. Attack
1582 time now has a minimal value to prevent clicks.
1583 - fixed pitch changes larger than one octave
1584 - fixed EG3 (pitch envelope) synthesis which was neutral all the time
1585 - implemented portamento mode and solo mode (a.k.a 'mono mode'):
1586 all modes can be altered via standard GM messages, that is CC5 for
1587 altering portamento time, CC65 for enabling / disabling portamento
1588 mode, CC126 for enabling solo mode and CC127 for disabling solo mode
1589 - fine tuning of the curves for volume (CC7), pan (CC10 and gig
1590 parameter) and crossfade
1591 - added support for the "attenuation controller threshold" gig
1592 parameter
1593 - added smoothing of volume changes caused by control change
1594 messages
1595 - sample loop parameters are now taken from the DimensionRegion
1596 instead of the wave chunk
1597 - fixed keyswitching for v3 gigs with a number of keyswitch splits
1598 not equal to a power of two
1599 - reimplementation of the filter algorithm. The new filter is more
1600 accurate and supports all gig filter types, including bandreject
1601 and lowpass turbo.
1602 - real support for 24 bit samples - samples are not truncated to
1603 16 bits anymore
1604 - support for aftertouch (channel pressure, not polyphonic
1605 aftertouch)
1606 - LFO1 behaviour fixed (dampening from max volume instead of
1607 amplifying from 0)
1608
1609 * LSCP server:
1610 - fixed application exit on broken pipe error (fixes bug #20)
1611 - fixed the notification messages delay due to lack of
1612 network activity (fixes bug #26)
1613 - fixed parser bug which occured on space(s) within device parameters
1614 (e.g. "SET AUDIO_OUTPUT_CHANNEL_PARAMETER 0 0 SOMEPARAM='foo bar'")
1615
1616 * audio driver:
1617 - added aRts audio output driver (by no means RT safe)
1618
1619 * MIDI driver:
1620 - fixed legacy sysex code which caused dispatching of MIDI SysEx
1621 messages several times instead of once
1622 - API extension for MIDI drivers which already supply exact time stamps
1623 for events (i.e. for offline rendering based MIDI drivers)
1624 - added checks for bad MIDI data
1625
1626 * linuxsampler application:
1627 - show available sampler engine types on startup
1628
1629 * general changes:
1630 - support for muting sampler channels and solo mode of the same, two new
1631 LSCP commands ("SET CHANNEL MUTE" and "SET CHANNEL SOLO") and two new
1632 fields ("MUTE" and "SOLO") for command "GET CHANNEL INFO" were
1633 introduced for this, the behavior is the same like on a mixer console
1634 (patch by Grigor Iliev, a bit adjusted). Also added configure option
1635 --enable-process-muted-channels which can be used to enable the
1636 processing of muted channels.
1637 - support for sostenuto pedal
1638 - support for monitoring the total number of active voices
1639 (new LSCP commands: GET TOTAL_VOICE_COUNT, GET TOTAL_VOICE_COUNT_MAX,
1640 SUBSCRIBE TOTAL_VOICE_COUNT, UNSUBSCRIBE TOTAL_VOICE_COUNT)
1641 - fixed some memory management errors
1642 - fixed some concurrency problems that could lead to crashes when
1643 LSCP commands were executed
1644 - fixed crash when instrument loading failed previously on the same
1645 sampler channel (fixes bug #36)
1646
1647 Version 0.3.3 (15 July 2005)
1648
1649 * packaging changes:
1650 - fixed compilation with gcc 4.0
1651
1652 * Gigasampler format engine:
1653 - LFO preprocessing was performed even though the respective LFO was
1654 disabled by the instrument patch (minor efficiency fix)
1655 - if period time of chosen audio device is too small (< MIN_RELEASE_TIME)
1656 for volume ramp downs in the same fragment (needed for current voice
1657 stealing implementation) then simply inform the user with a warning
1658 about possible click sounds and reduce the volume ramp down
1659 appropriately instead of cancelling the audio device connection
1660
1661 Version 0.3.2 (24 June 2005)
1662
1663 * packaging changes:
1664 - updated autotools build files to compile on OS X
1665 (conditional compilation of CoreMIDI and MidiShare drivers)
1666 - hand-crafted assembly optimization code can be disabled with
1667 './configure --disable-asm' (definitely not recommended)
1668 - fixed 'make dist' rule to include all necessary files
1669 - require automake (>= 1.5) for 'make -f Makefile.cvs'
1670 (due to 'dist-bzip2' automake option)
1671
1672 * Gigasampler format engine:
1673 - support for the gig parameters for "release velocity response" curves
1674 - fine tuning of the EGADSR envelope
1675 - volume of release triggered samples now depends on note-on velocity,
1676 note length and gig parameter "release trigger decay" instead of
1677 note-off velocity.
1678 - revised voice stealing
1679 (fixes crash and endless loop caused by voice stealing)
1680 - don't reset scale tuning on instrument or audio output device change
1681 - handle key group conflicts right at the beginning of each fragment
1682 instead of when the respective voice is actually launched
1683 (fixes undefined behavior if stolen voices belonged to a key group -
1684 this case was followed by a "killed voice survived" error message)
1685 - fixed minor issue with null/silence samples
1686 (those stole voices even though they don't need a voice at all which
1687 resulted in "voice stealing didn't work out" messages)
1688 - don't reset volume, pan, pitch and MIDI controller values on
1689 instrument or audio output device change
1690
1691 * LSCP server:
1692 - fixed some crashes (patch by Grigor Iliev, fixes #19)
1693 - fixed LSCP event "CHANNEL_INFO" notification
1694 (e.g. did not notify on volume changes or MIDI program change events)
1695
1696 * linuxsampler application:
1697 - added command line parameters --lscp-addr and --lscp-port to override
1698 default IP address and port of LSCP server
1699
1700 Version 0.3.1 (24 May 2005)
1701
1702 * initial release

  ViewVC Help
Powered by ViewVC