ViewVC logotype

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 3646 - (show annotations) (download)
Sun Dec 8 23:17:34 2019 UTC (46 hours, 30 minutes ago) by schoenebeck
File size: 94511 byte(s)
* gig format extension: Support for additional filter types:
  lowpass 1/2/4/6-pole, highpass 1/2/4/6-pole, bandpass 2-pole,
  bandreject 2-pole.

* Bumped version (2.1.1.svn24).

1 Version SVN trunk (?)
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 - Using now cubic interpolation as resampling algorithm by default; and
30 linear interpolation is deprecated now.
32 * Real-time instrument scripts:
33 - Added method ScriptVM::setExitResultEnabled() which allows to
34 explicitly enable the built-in exit() function to optionally accept
35 one function argument; the value of the passed exit() function
36 argument will then become available by calling
37 VMExecContext::exitResult() after script execution.
38 - 64 bit support for NKSP integer scripts variables (declare $foo).
39 - Variable names, function names and preprocessor condition names must start
40 with a regular character (a-z or A-Z); starting them with a digit or
41 underscore is not allowed.
42 - NKSP parser fix: equal comparison operator "=" and not equal comparison
43 operator "#" must only accept integer operands.
44 - NKSP language: Implemented support for standard units like Hertz, seconds,
45 Bel including support for metric unit prefixes; so one can now e.g.
46 conveniently use numbers in scripts like "5us" meaning "5 microseconds",
47 or e.g. "12kHz" meaning "12 kilo Hertz", or e.g. "-14mdB" meaning
48 "minus 14 Millidecibel", or e.g. "28c" meaning "28 cents" (for tuning).
49 - NKSP language: Introduced "final" operator "!" which is specifically
50 intended for synthesis parameter values to denote that the synthesis
51 parameter value is intended to be the "final" value for that synthesis
52 parameter that should explicitly be used by the engine and thus causing
53 the sampler engine to ignore all other modulation sources for the same
54 synthesis parameter (like e.g. LFO, EG); by simply prefixing a value,
55 variable or formula with this new "!" operator the expression is marked as
56 being "final".
57 - NKSP script editor API: Added support for detecting standard unit tokens
58 and their potential metric prefix token.
59 - NKSP language: Added support for NKSP real number literals and
60 arithmetic operations on them (e.g. "(3.9 + 2.9) / 12.3 - 42.0").
61 - NKSP language: Added support for NKSP real number (floating point) script
62 variables (declare ~foo).
63 - NKSP language: Added support for NKSP real number (floating point) array
64 script variables (declare ?foo[]).
65 - Built-in script function "message()" accepts now real number argument as
66 well.
67 - Added built-in script function "real_to_int()" and its short hand form
68 "int()" for casting from real number to integer in NKSP scripts.
69 - Added built-in script function "int_to_real()" and its short hand form
70 "real()" for casting from integer to real number in NKSP scripts.
71 - Allow built-in exit() function to potentially accept real number type
72 argument as well.
73 - Built-in script functions may have a different return type depending on
74 the arguments passed to the function.
75 - Built-in script function "abs()" optionally accepts and returns real
76 number.
77 - Built-in script functions "min()" and "max()" optionally accept real
78 number arguments and return real number as result in that case.
79 - NKSP VM API: Allow units and 'final'ness to be returned as result from
80 built-in functions (added methods VMFunction::returnUnitType() and
81 VMFunction::returnsFinal() for that purpose which must be implemented by
82 built-in function implementations).
83 - NKSP language: Allow metric unit prefixes of numeric scalar and array
84 variables to be changed freely at runtime (unlike unit types like Hz etc.
85 which are still sticky, parse-time features of variables which cannot be
86 changed at runtime for the intentional sake of determinism).
87 - NKSP language: 'final' values are prohibited for array variables for now
88 (attempt causes a parsers error).
89 - NKSP language: expressions with unit types (e.g. Hz) are prohibited for
90 conditions of runtime control structures like if(), while(), select()
91 (attempt causes a parser error).
92 - NKSP VM API: Allow built-in functions to perform their own, individual
93 parse time checks of arguments going to be passed to the function at
94 runtime (added method VMFunction::checkArgs() for that purpose).
95 - NKSP language: raise parser warning if only one operand of binary
96 operators (like logical 'or' comparison) contain a 'final' value (because
97 it would always yield in a 'final' result in such cases).
98 - NKSP language: Allow comparison (=, #, <, >, <=, >=) of values with
99 different metric unit prefixes, which will behave as expected (e.g.
100 result of expression '1000us < 2ms' is true).
101 - NKSP language: Allow adding values with different metric unit prefixes
102 (e.g. result of expression '100Hz + 5kHz' is '5100Hz').
103 - NKSP language: Allow subtracting values with different metric unit
104 prefixes (e.g. result of expression '1ms - 20us' is '980us').
105 - NKSP language: Allow multiplying with any metric unit prefixes
106 (e.g. result of expression '2k * 3ms' is '6s'), however multiplications
107 with unit types on both sides (e.g. '2s * 2s') is still prohibited since
108 we don't have any considerable practical use for a term like '4s^2'
109 (hence any attempt multiplying two unit types still causes parser error).
110 - NKSP language: Allow dividing by any metric unit prefixes and allow
111 division of same unit type on both sides (e.g. expression '8kHz / 1000Hz'
112 yields in unit free result '8'). So this is now a way to cast units away
113 e.g. for passing the result to other expressions, certain function calls
114 or variables which are not accepting any units (or that specific unit).
115 - NKSP language: integer arrays and real number arrays can now be converted
116 to strings (e.g. for dumping their content with message() calls for
117 script debugging purposes).
118 - NKSP language: expressions and variables with units are now correctly
119 casted to strings (e.g. with message() calls).
120 - NKSP language: comparing real numbers for equalness (e.g. '~foo = 3.1') or
121 unequalness (e.g. '~foo # 3.1') is now less strict and takes the expected
122 floating point tolerances into account.
123 - NKSP VM API: Added methods VMScalarNumberExpr::evalCastInt() and
124 VMScalarNumberExpr::evalCastReal().
125 - NKSP VM API: Added base class 'VMNumberArrayExpr' for classes
126 'VMIntArrayExpr' and 'VMRealArrayExpr'.
127 - NKSP VM API: replaced all unitPrefix() (parse time) methods by
128 unitFactor() (runtime) methods.
129 - Built-in function "exit()" supports now returning units and 'final'ness
130 for test cases.
131 - The following built-in functions support now units as well: "abs()",
132 "random()", "inc()", "dec()", "in_range()", "min()", "max()",
133 "real_to_int()", "int()", "int_to_real()" and "real()".
134 - Built-in functions "array_equal()", "search()" and "sort()" support now
135 real number arrays (correctly) as well.
136 - Added individual parse time checks of arguments to be passed to built-in
137 functions "random()", "inc()", "dec()", "in_range()", "min()", "max()",
138 "array_equal()" and "search()" specific for their individual purposes.
139 - NKSP VM refactoring: Renamed all methods, functions and classes matching
140 pattern *ScalarNumber* to simply *Number* (that is i.e. classes
141 VMScalarNumberExpr -> VMNumberExpr, ScalarNumberExpr -> NumberExpr,
142 ScalarNumberVariable -> NumberVariable, ScalarNumberBinaryOp ->
143 NumberBinaryOp, VMScalarNumberResultFunction -> VMNumberResultFunction,
144 method VMExpr::asScalarNumberExpr() -> VMExpr::asNumber(), function
145 isScalarNumber() -> isNumber()).
146 - NKSP VM API: Added 4 overridden methods to class VMNumberExpr:
147 evalCastInt(MetricPrefix_t), evalCastInt(MetricPrefix_t,MetricPrefix_t),
148 evalCastReal(MetricPrefix_t), evalCastReal(MetricPrefix_t,MetricPrefix_t)
149 as convenient methods for automatically converting values to expected
150 metric value basis.
151 - Built-in function "wait()" accepts now both integers and real numbers as
152 argument.
153 - NKSP VM API cleanup: Get rid of legacy method
154 VMFunction::argType(vmint iArg) which was already superseded by its new
155 replacement VMFunction::acceptsArgType(vmint iArg, ExprType_t type).
156 - NKSP parser: if wrong argument type was passed to a built-in function and
157 that built-in function accepts more than one data type for the argument,
158 then show all supported data types as parser error message.
159 - Built-in function "play_note()" accepts now real numbers and seconds as
160 unit type as well for its 3rd and 4th function arguments.
161 - The following built-in functions accept now real numbers as well for their
162 2nd function argument: "change_vol()", "change_tune()", "change_cutoff()",
163 "change_attack()", "change_decay()", "change_release()",
164 "change_sustain()", "change_cutoff_attack()", "change_cutoff_decay()",
165 "change_cutoff_sustain()", "change_cutoff_release()",
166 "change_amp_lfo_freq()", "change_cutoff_lfo_freq()",
167 "change_pitch_lfo_freq()", "change_vol_time()", "change_tune_time()",
168 "change_pan_time()", "fade_in()", "fade_out()", "change_play_pos()".
169 - Fixed built-in function "change_play_pos()" not having accepted metric
170 prefixes at all.
171 - Fixed the following built-in functions having misinterpreted values given
172 with unit type (for their 2nd argument) as if they were relative values
173 (that is as if they were passed without a unit type): "change_attack()",
174 "change_decay()", "change_release()", "change_cutoff_attack()",
175 "change_cutoff_decay()", "change_cutoff_release()".
176 - Fixed the following built-in functions having applied completely wrong
177 'final' values: "change_sustain()", "change_cutoff_sustain()" (since the
178 respective EGs being their modulation sink assume uint data type with
179 value range 0..1000 instead of 0.0..1.0.
180 - Added individual parse-time checks of function arguments for the following
181 built-in functions: "play_note()", "note_off()", "set_event_mark()",
182 "delete_event_mark()", "by_marks()", "change_cutoff()", "change_attack()",
183 "change_decay()", "change_release()", "change_cutoff_attack()",
184 "change_cutoff_decay()", "change_cutoff_release()",
185 "change_amp_lfo_freq()", "change_cutoff_lfo_freq()",
186 "change_pitch_lfo_freq()", "change_vol_time()", "change_tune_time()" and
187 "change_pan_time()".
188 - Don't abort function call if unit type was used and at the same time
189 'final' operator was omitted for the primary value argument of the
190 following built-in functions: "change_cutoff()", "change_attack()",
191 "change_decay()", "change_release()", "change_cutoff_attack()",
192 "change_cutoff_decay()", "change_cutoff_release()",
193 "change_amp_lfo_freq()", "change_cutoff_lfo_freq()",
194 "change_pitch_lfo_freq()", "change_vol_time()", "change_tune_time()",
195 "change_pan_time()", instead imply 'final'ness at runtime and raise an
196 appropriate parser warning at parse time.
197 - Added built-in real number functions "round()", "ceil()", "floor()",
198 "sqrt()", "log()", "log2()", "log10()", "exp()", "pow()", "sin()",
199 "cos()", "tan()", "asin()", "acos()", "atan()".
200 - Added built-in script real number constant "~NI_MATH_PI".
201 - Added built-in script real number constant "~NI_MATH_E".
202 - NKSP language: Allow unary '+' operator.
204 * test cases:
205 - Fixed compiler errors in test cases.
206 - Updated README for how to compile & run test cases.
207 - Updated test case
208 MutexTest::testDoubleLockStillBlocksConcurrentThread() to latest
209 expected behaviour of the Mutex class implementation (recursive
210 mutex type).
211 - Added test cases for NKSP core language aspects and core built-in
212 functions.
213 - Fixed thread tests segfaulting on Linux.
214 - NKSP: Added real number test cases for built-in functions exit(),
215 int_to_real(), real(), real_to_int() and int(), as well as for the
216 plus, minus and negate language operators.
217 - Added massive amount of NKSP test cases for standard measuring units and
218 'final' operator usage cases.
219 - Added NKSP test cases for (floating point tolerance aware) real number
220 equalness / unequalness comparison.
221 - Added NKSP int array and real array tests for value assignment and
222 initialization of arrays.
223 - Added NKSP test cases for built-in functions "round()", "ceil()",
224 "floor()", "sqrt()", "log()", "log2()", "log10()", "exp()", "pow()",
225 "sin()", "cos()", "tan()", "asin()", "acos()", "atan()".
226 - Added NKSP test cases for unary '+' operator.
228 * GigaStudio/Gigasampler format engine:
229 - LFOTriangleIntMath and LFOTriangleIntAbsMath: Fixed FlipPhase=true
230 behaviour for start_level_mid.
231 - Changed LFO start levels: LFO1 and LFO2 both to mid, LFO3 to max. start
232 level (see discussion "GigaStudio LFO compatibility" on mailing list
233 from 2019-09-26 for details).
234 - Changed default wave form for all 3 LFOs to sine (instead of triangle;
235 see discussion "GigaStudio LFO compatibility" on mailing list from
236 2019-09-26 for details).
237 - Format extension: Added support for different LFO wave forms (currently
238 either sine [default], triangle, saw or square).
239 - Format extension: Added support for LFO phase displacement (0��..360��).
240 - Format extension: Added support for flipping LFO polarity on LFO 3
241 (in the original gig format this was only available for LFO 1 and LFO 2).
242 - Format extension: Support for additional filter types:
243 lowpass 1/2/4/6-pole, highpass 1/2/4/6-pole, bandpass 2-pole,
244 bandreject 2-pole.
246 * SFZ format engine:
247 - Fixed support for regions with loccN/hiccN conditions on more than one
248 MIDI controller.
250 * Benchmarks:
251 - Fixed benchmarks/triang.cpp falsely having favoured "int math abs"
252 algorithm (since result of 2nd run was not accumulated).
253 - Added benchmark for saw wave (benchmarks/saw.cpp).
254 - Added benchmark for sine wave (benchmarks/sine.cpp).
255 - Added benchmark for square wave (benchmarks/square.cpp).
256 - Increased amount of benchmarks runs by factor 6 to achieve benchmark times
257 which are large enough on modern systems.
259 Version 2.1.1 (27 Jul 2019)
261 * Real-time instrument scripts:
262 - Fixed behavior of built-in NKSP functions change_sustain(),
263 change_cutoff_attack(), change_cutoff_decay(), change_cutoff_sustain()
264 and change_cutoff_release().
266 * general changes:
267 - Only play release trigger samples on sustain pedal up if this behaviour
268 was explicitly requested by the instrument (otherwise only on note-off).
269 - Fixed compiler warnings.
270 - Fixed compilation error when cross-compiling to Mac.
271 - FX Sends: Provide more useful error messages on routing problems
272 (see bug #169).
273 - LSCP doc: Be more clear describing the two distinct approaches
274 of using external vs. internal effects (see bug #169).
275 - "optional" class: Fixed comparison operators.
277 * Gigasampler/GigaStudio format engine:
278 - Format extension: If requested by instrument then don't play release
279 trigger sample on note-off events.
281 * SFZ format engine:
282 - Fixed memory leak when releasing samples
283 (fixes bug #307, patch by Jacek Roszkowski)
284 - Fixed potential crash when a sample is shared by more than one region
285 (fixes bug #308, patch by Jacek Roszkowski).
286 - Opcode 'sample': Added support for built-in sample '*silence'
287 (fixes bug #310, patch by Jacek Roszkowski).
289 Version 2.1.0 (25 Nov 2017)
291 * SFZ format engine:
292 - added support for <global>, <master> and #define (patch by Alby M)
293 - Removed code duplication in SFZ file loading code.
294 - Added support for sfz extension opcode 'script' which may be used to
295 load real-time instrument script file (NKSP script language).
296 - Implemented opcode set_ccN (initial patch by Giovanni Senatore).
297 - Fixed unintended volume fade-in of voices under certain conditions.
298 - sfz parser: allow missing space between header and opcode
300 * Gigasampler/GigaStudio format engine:
301 - Fixed clicks and pumping noise with Lowpass Turbo filter on very low
302 cutoff settings.
303 - Got rid of resembling an ancient GSt misbehavior which did not pitch at
304 all if an up-pitch of more than 40 semi tones was requested (I don't
305 think there is any stock gig sound that requires this behavior to
306 resemble its original sound).
307 - Added support for controlling whether the individual EGADSR stages may
308 be aborted (as LinuxSampler extension to the original GigaStudio 4
309 format).
311 * general changes:
312 - fixed printf type errors (mostly in debug messages)
313 - use unique_ptr instead of auto_ptr when building with C++11
314 - Added RTAVLTree class which is a real-time safe ordered multi-map, thus
315 allowing to sort data efficiently in real-time safe manner.
316 - RTList class: added methods for moving/inserting elements to arbitrary
317 position within a list.
318 - RTList class: added method fromPtr() for creating an Iterator object from
319 a raw element pointer.
320 - Attempt to partly fix resetting engine channels vs. resetting engine, an
321 overall cleanup of the Reset*(), ConnectAudioDevice(),
322 DisconnectAudioDevice() API methods would still be desirable though,
323 because the current situation is still inconsistent and error prone.
324 - Active voices are now internally grouped to "Note" objects, instead of
325 being directly assigned to a keyboard key. This allows more fine graded
326 processing of voices, which is i.e. required for certain instrument
327 script features.
328 - Fix: Release trigger voices were not spawned on sustain pedal up
329 (CC #64) events.
330 - Fix: Release trigger voices can now also distinguish correctly between
331 note-off and sustain pedal up events.
332 - All engines: Increased ramp speed of volume smoother and pan smoother
333 (while slow rate of crossfade smoother is preserved) to allow quick
334 volume and pan changes by instrument scripts for instance.
335 - gig/sf2/sfz: Fixed aftertouch channel pressure events being processed
336 properly and thus were ignored for certain uses.
337 - Added new C++ API method
338 VirtualMidiDevice::SendChannelPressureToSampler().
339 - windows, 32-bit: fixed potential crashes by making sure the stack in
340 sub threads is 16-byte aligned
341 - fixed numerous compiler warnings
342 - Fixed invalid (note-on) event ID being assigned to new Note objects.
343 - Revised fundamental C++ classes "Thread", "Mutex" and "Condition" which
344 fixes potential undefined behavior.
345 - Fixed Note object leak when triggering notes on keys which did not
346 have a valid sample mapped (fixes bug #252).
347 - Fixed compilation errors when compiling with CONFIG_DEVMODE enabled.
348 - linuxsampler binary fix: option --create-instruments-db ignored
349 subsequent optional argument due to glibc's implementation oddity
350 which expects a "=" sign, but no space between them.
352 * packaging changes:
353 - removed unnecessary dependency to libuuid
354 (originated by libgig's usage of it)
355 - Automake: set environment variable GCC_COLORS=auto to allow GCC to
356 auto detect whether it (sh/c)ould output its messages in color.
357 - Debian: Fixed packaging error about invalid "Source-Version"
358 substitution variable.
359 - Debian: Raised Debian compatibility level to Debian 9 "Stretch".
360 - Debian: Added build dependency to libsqlite3-dev for building
361 linuxsampler with instruments DB support.
363 * Real-time instrument scripts:
364 - Implemented scheduler for delayed MIDI events and for suspended scripts.
365 - Built-in script function "wait()": implemented support for function's
366 "duration-us" argument, thus scripts using this function are now
367 correctly resumed after the requested amount of microseconds.
368 - Built-in script function "play_note()": implemented support for
369 function's "duration-us" argument, thus notes triggered with this
370 argument are now correctly released after the requested amount of
371 microseconds.
372 - Fix: script events were not cleared when engine channel was reset,
373 potentially causing undefined behavior.
374 - Fixed crash which happened when trying to reference an undeclared
375 variable.
376 - Built-in script function "play_note()": Added support for passing
377 special value -1 for "duration-us" argument, which will cause the
378 triggered note to be released once the original note was released.
379 - Instrument script classes now exported with the liblinuxsampler C++ API.
380 - Added new API method ScriptVM::syntaxHighlighting() which provides
381 a convenient syntax highlighting backend for external instrument
382 script editor applications.
383 - Added new C++ API class "ScriptVMFactory".
384 - Extended parser issues to provide not only first line and first
385 column, but also last line and last column of issue (thus marking
386 the precise span of the issue within the source code).
387 - Reload script automatically after being modified by an instrument
388 editor.
389 - NKSP language grammar correction: allow empty event handler bodies
390 like "on note end on".
391 - Implemented built-in script function "change_vol()".
392 - Implemented built-in script function "change_tune()".
393 - Implemented built-in script function "change_pan()".
394 - Implemented built-in script function "change_cutoff()".
395 - Implemented built-in script function "change_reso()".
396 - Implemented built-in script function "event_status()".
397 - Added built-in script constants "$EVENT_STATUS_INACTIVE" and
398 "$EVENT_STATUS_NOTE_QUEUE" both for being used as flags for
399 "event_status()" function.
400 - NKSP language: Added support for bitwise operators ".or.", ".and."
401 and ".not.".
402 - NKSP language scanner: Fixed IDs matching to require at least one
403 character (i.e. when matching function names or variable names).
404 - NKSP language scanner: disabled unusued rules.
405 - Fixed behavior of play_note() and note_off() functions which must
406 be distinguished engine internally from "real" MIDI note on/off
407 events in order to avoid misbehaviors like hanging notes.
408 - Implemented built-in script variable "$KSP_TIMER".
409 - Implemented built-in script variable "$NKSP_REAL_TIMER".
410 - Implemented built-in script variable "$NKSP_PERF_TIMER".
411 - Implemented built-in script variable "$ENGINE_UPTIME".
412 - Implemented built-in script function "inc()".
413 - Implemented built-in script function "dec()".
414 - NKSP language fix: division expressions were evaluated too often.
415 - NKSP language fix: string concatenation operator was right
416 associative instead of left (to right).
417 - NKSP language correction: allow empty statements in entire language
418 (i.e. "if end if").
419 - Implemented built-in script function "stop_wait()".
420 - Implemented built-in script variable "$NI_CALLBACK_ID".
421 - Implemented built-in script variable "$NI_CALLBACK_TYPE".
422 - Implemented built-in script variable "$NKSP_IGNORE_WAIT".
423 - Added support for read-only built-in variables (respectively
424 handled by the script parser).
425 - Added built-in script constant "$NI_CB_TYPE_INIT".
426 - Added built-in script constant "$NI_CB_TYPE_NOTE".
427 - Added built-in script constant "$NI_CB_TYPE_RELEASE".
428 - Added built-in script constant "$NI_CB_TYPE_CONTROLLER".
429 - NKSP Language: Added support for user defined script functions.
430 - Implemented built-in script function "change_attack()".
431 - Implemented built-in script function "change_decay()".
432 - Implemented built-in script function "change_release()".
433 - Fixed all change_*() built-in script functions to apply their
434 synthesis parameter changes immediately in case the respective note
435 was triggered at the same time, instead of scheduling the parameter
436 change, especially because it would cause some parameter types's
437 changes either to be ramped (i.e. change_vol()) or other types even
438 to have not effect at all (i.e. change_attack()).
439 - Implemented built-in script function "sh_left()".
440 - Implemented built-in script function "sh_right()".
441 - Implemented built-in script function "min()".
442 - Implemented built-in script function "max()".
443 - NKSP Fix: "init" event handler was not always executed when a script
444 was loaded.
445 - NKSP built-in wait() script function: abort script execution if
446 a negative or zero wait time was passed as argument (since this is
447 a common indication of a bug either of the script or even of the
448 engine, which could lead to RT instability or even worse).
449 - ScriptVM: Implemented automatic suspension of RT safety
450 threatening scripts.
451 - Provide more user friendly error messages on syntax errors.
452 - Fixed NKSP parser warning "Not a statement" when assigning an
453 initializer list to an array variable.
454 - Implemented built-in script array variable "%ALL_EVENTS".
455 - Implemented built-in script function "in_range()".
456 - Implemented built-in script function "change_amp_lfo_depth()".
457 - Implemented built-in script function "change_amp_lfo_freq()".
458 - Implemented built-in script function "change_pitch_lfo_depth()".
459 - Implemented built-in script function "change_pitch_lfo_freq()".
460 - Implemented built-in script function "change_vol_time()".
461 - Implemented built-in script function "change_tune_time()".
462 - Implemented built-in script function "fade_in()".
463 - Implemented built-in script function "fade_out()".
464 - Fixed acceptance of wrong data type of parameters passed to built-in
465 script functions "change_vol()", "change_tune()", "change_pan()",
466 "change_cutoff()", "change_reso()", "change_attack()", "change_decay()",
467 "change_release()", "change_amp_lfo_depth()", "change_amp_lfo_freq()",
468 "change_pitch_lfo_depth()" and "change_pitch_lfo_freq()".
469 - Added built-in script function "get_event_par()" and implemented some
470 of its possible parameter selections.
471 - Added built-in script function "set_event_par()" and implemented some
472 of its possible parameter selections.
473 - Fixed a bunch of scheduler time related bugs.
474 - Fixed polyphonic variables not being reset to zero after usage.
475 - Built-in "ignore_event()" function: argument is now optional, like with
476 built-in function "ignore_controller()".
477 - Implemented built-in script function "change_velo()".
478 - Implemented built-in script function "change_note()".
479 - Adjusted behavior of "change_vol()" and "change_tune()" to a more
480 intuitive behavior if used in combination with "change_vol_time()" or
481 "change_tune_time()" respectively: now tuning/volume changes are only
482 assigned (without delay) immediately to a new note if the respective
483 timing function has not been called before, otherwise the volume/tuning
484 changes are automatically faded (before, only the event's time stamp was
485 relevant).
486 - Implemented built-in script function "array_equal()".
487 - Implemented built-in script function "search()".
488 - Implemented built-in script function "sort()".
489 - NKSP Fix: Never suspend "init" event handlers.
490 - Implemented built-in script function "same_region()" (currently only
491 available for gig format engine).
492 - Added built-in script constant "$NKSP_LINEAR".
493 - Added built-in script constant "$NKSP_EASE_IN_EASE_OUT".
494 - Implemented built-in script function "change_vol_curve()".
495 - Implemented built-in script function "change_tune_curve()".
496 - built-in "play_note()" function now supports a sample playback start
497 offset with argument 3, where special value -1 means to use the regular
498 sample offset as defined by the instrument file.
499 - Built-in array variable %KEY_DOWN[] is now a read-only variable.
500 - Built-in variable $EVENT_NOTE is now a read-only variable.
501 - Built-in variable $EVENT_VELOCITY is now a read-only variable.
502 - built-in "play_note()" function now accepts -2 for its fourth argument
503 (note duration) which means the life time of the note shall be sticked
504 to the requested note number of argument 1.
505 - Fix: built-in "play_note()" function now returns 0 as result value if
506 -1 was passed for its fourth argument (note duration) and the respective
507 parent note is already gone.
508 - Implemented built-in script function "change_play_pos()".
509 - NKSP language: Added support for user declared const array variables.
510 - NKSP language: Raise parser warning if array variable is accessed with
511 an index that exceeds the array's size.
512 - NKSP language fix: Unknown characters were not handled correctly.
513 - NKSP language: Added support for "synchronized .. end synchronized"
514 code blocks.
515 - Implemented built-in script function "abort()" which allows to abort
516 another script handler by passing its callback ID.
517 - Fixed potential memory access bug and potential undefined behavior of
518 "init" event handlers.
519 - Print a time stamp along to each call of built-in function "message()".
520 - ScriptVM API: Added VMParserContext::preprocessorComments() which allows
521 to retrieve all code blocks filtered out by the preprocessor.
522 - Added built-in script function "fork()".
523 - Added built-in array variable %NKSP_CALLBACK_CHILD_ID[].
524 - Added built-in variable $NKSP_CALLBACK_PARENT_ID.
525 - Fixed potential crash when accessing dynamic built-in array variables.
526 - Added built-in script function "callback_status()".
527 - Added built-in constant $CALLBACK_STATUS_TERMINATED.
528 - Added built-in constant $CALLBACK_STATUS_QUEUE.
529 - Added built-in constant $CALLBACK_STATUS_RUNNING.
530 - Removed max. value limitation of built-in functions "change_attack()",
531 "change_decay()" and "change_release()" to i.e. allow passing 2000000
532 for doubling the respective time.
533 - NKSP script editor syntax highlighting API: Fixed app termination due
534 to a lexer start condition stack underrun.
535 - NKSP preprocessor: Fixed wrong behavior on nested USE_CODE_IF() and
536 USE_CODE_IF_NOT() preprocessor statements.
537 - NKSP: Added built-in preprocessor condition NKSP_NO_MESSAGE, which
538 can be set to disable all subsequent built-in "message()" function calls
539 on preprocessor level.
540 - Implemented built-in script function "change_sustain()".
541 - NKSP script editor syntax highlighting API: catch all fatal lexer errors,
542 to avoid the editor app to crash on ill-formed text input.
543 - Added built-in script function "change_pan_time()".
544 - Added built-in script function "change_pan_curve()".
545 - Added built-in script function "change_cutoff_attack()".
546 - Added built-in script function "change_cutoff_decay()".
547 - Added built-in script function "change_cutoff_sustain()".
548 - Added built-in script function "change_cutoff_release()".
549 - Added built-in script function "change_cutoff_lfo_depth()".
550 - Added built-in script function "change_cutoff_lfo_freq()".
552 * Instruments DB:
553 - Fixed memory access bug of general DB access code which lead to
554 undefined behavior.
555 - Cleanup of instruments DB file creation and opening code.
556 - The instrument DB path of linuxsampler's --create-instruments-db argument
557 is now optional, if it is missing, then a default location is used.
558 - Added support for scanning SFZ (.sfz) files.
559 - Added support for scanning Sound Font (.sf2) files.
560 - Fixed undefined DB transaction behavior.
562 Version 2.0.0 (15 July 2015)
564 * packaging changes:
565 - fixed building with newer MinGW-w64
566 - Mac OS X: support the new dir for Core Audio SDK
567 - Mac OS X: fixed building outside source directory
568 - made sure all source files for hostplugins are included when
569 doing "make dist"
570 - removed empty directories left from the cvs to svn migration
571 - fixed compilation with gcc 4.6.1
572 - another "make dist" fix, for the LV2 plugin
573 - made --enable-pthread-testcancel default on Mac OS X
574 - minor Makefile fix for building lscpparser when build and source
575 directories are separate
576 - Mac OS X: made it possible to specify plugin installation dir to
577 configure
578 - Mac OS X: Makefile fix for the install-strip target
579 - fixed compilation with gcc 4.7
580 - fixed configure script error with old autoconf versions
581 - lsatomic.h: use gcc provided atomic functions if building with
582 gcc 4.7 and C++11
583 - modernized configure script
584 - fixed linkage error when building with
585 LDFLAGS="-Wl,--no-undefined" (#190)
586 - fixed compilation with Clang 3.2
587 - removed usage of deprecated Automake variable INCLUDES
588 - fixed building with C++11
589 - build fix: ChangeFlagRelaxed.h was missing in makefile
590 - build fix: libsndfile compiler flags were missing in some
591 makefiles
592 - fix for building with bison 3.0 (#202)
593 - Mac OS X: added temporary hack allowing to spawn gigedit as callback
594 on the process's main thread
595 - fixed build error on newer MinGW
596 - support building with older jack versions
597 - support building with spaces in vst sdk path
598 - enabled automake 'subdir-objects' option and moved external
599 source references (vst, au, asio) from makefiles to cpp files,
600 in order to get rid of warnings from automake 1.14
602 * general changes:
603 - Refactoring: moved the independent code from
604 the Gigasampler format engine to base classes
605 - command line option '--profile' is currently disabled, since the
606 respective profiling code is currently broken
607 - Introduced support for optional environment variable
608 "LINUXSAMPLER_PLUGIN_DIR" which allows to override the directory
609 where the sampler shall look for instrument editor plugins
610 (patch by Luis Garrido, slightly modified).
611 - implemented Roland GS NRPN 1ArrH which allows to set volume per note
612 - implemented Roland GS NRPN 1CrrH which allows to set pan per note
613 - implemented Roland GS NRPN 1DrrH which allows to set reverb send per
614 note (in this implementation of the sampler its simply hard routed to
615 the 1st effect send of the sampler channel, no matter what the actual
616 effect type is)
617 - implemented Roland GS NRPN 1ErrH which allows to set chorus send per
618 note (in this implementation of the sampler its simply hard routed to
619 the 2nd effect send of the sampler channel, no matter what the actual
620 effect type is)
621 - implemented support for internal LADSPA effects
622 - raised limit of program change queue from 100 to 512
623 (as suggested by Alex Stone)
624 - implemented sine LFO, pulse LFO and saw LFO
625 - experimental support for per voice equalization
626 - added command line option --exec-after-init
627 - Introduced new C++ API method:
628 EngineChannel::InstrumentFileName(int index)
629 allowing to retrieve the whole list of files used for the loaded
630 instrument on an engine channel (a.k.a. part). Some GigaStudio
631 instruments for example are splitted over several files like
632 "Foo.gig", "Foo.gx01", "Foo.gx02", ...
633 - Added new C++ API method Sampler::GetGlobalMaxVoices().
634 - Added new C++ API method Sampler::GetGlobalMaxStreams().
635 - Added new C++ API method Sampler::SetGlobalMaxVoices().
636 - Added new C++ API method Sampler::SetGlobalMaxStreams().
637 - Various "const" and "restrict" optimizations.
638 - all engines: add pan CC value to instrument pan parameter before
639 applying panning, instead of using two separate pan functions in
640 series (#182)
641 - added a lock guard class for exception safe mutex handling and
642 used it everywhere appropriate
643 - Immediately apply scale tuning changes to active voices.
644 - Exposed scale tuning to C++ API (along to the already existing standard
645 SysEx way).
646 - lsatomic.h fixes: seq_cst load and store were suboptimal for x86
647 and broken for ppc64. (Seq_cst loads and stores are actually not
648 used in LS, so the bug wasn't noticable.)
649 - lsatomic.h: added ARMv7 support
650 - Added support for multiple MIDI input ports per sampler channel (and
651 added various new C++ methods for this new feature / design change, old
652 C++ API methods for managing SamplerChannel's MIDI inputs are now marked
653 as deprecated but are still there and should provide full behavior
654 backward compatibility).
655 - AbstractEngine::GSChecksum(): don't allocate memory on the stack (was
656 unsafe and caused compilation error with clang 2.x).
657 - Bugfix: only process the latest MIDI program change event.
658 - Introducing the LSCP shell, which provides convenient control of the
659 sampler from the command line by providing LSCP aware features.
660 - VirtualMidiDevice: Added support for program change.
661 - VirtualMidiDevice: Added support for bank select (MSB & LSB).
662 - VirtualMidiDevice: Added support for pitch bend.
663 - Aftertouch: extended API to explicitly handle channel pressure and
664 polyphonic key pressure events (so far polyphonic pressure was not
665 supported at all, and channel pressure was rerouted as CC128 but not
666 used so far).
667 - Added initial support for real-time instrument scripts. The script VM
668 code is shared by all sampler engine implemementations, however only the
669 gig file format currently provides support for storing instrument scripts
670 (as LinuxSampler extension to the original GigaStudio 4 format).
672 * Gigasampler format engine:
673 - implemented the "round robin keyboard" dimension
674 - fixed round robin and random dimensions for cases when number of
675 dimension zones is not a power of two
676 - made round robin use a counter for each region instead of each
677 key
678 - bugfix: pitch LFO controller "internal+aftertouch" was broken
679 - bugfix: filter keyboard tracking was broken
680 - filter performance fix (an unnecessary copy was made of the
681 filter parameters in each sub fragment)
682 - handle special case when pan parameter in gig file has max or
683 min value
684 - Exclusive Groups: don't ever stop voices of the same note,
685 doesn't sound naturally with a drumkit
686 - fixed EG1 modulation when attack or release is zero
687 - Fixed support for 'aftertouch' attenuation controller.
688 - Fixed crash that happened with velocity split sounds under certain
689 conditions (see also previous commit on libgig).
690 - fixed behaviour of filter LFO
692 * SFZ format engine:
693 - Initial implementation (not usable yet)
694 - added support for v2 multiple stage envelope generators
695 - added a fine-tuned v1 envelope generator instead of using the
696 one from the gig engine
697 - fixed memory leak and memory handling errors
698 - added support for trigger=first, trigger=legato and sw_previous
699 - allow non-numerical key values ("C#4" for example)
700 - "key" opcode now sets pitch_keycenter too
701 - fixed error when unloading instrument with same sample used by
702 multiple regions
703 - added some opcode aliases, like loopmode for loop_mode, to be
704 more compatible
705 - added support for trigger=release and rt_decay
706 - added support for off_mode=normal
707 - added support for random, seq_position, seq_length and volume
708 - added v1 LFO opcodes to parser (no support in engine yet)
709 - added support for amp_veltrack and amp_velcurve_N
710 - fine-tuned the default velocity curve
711 - added support for transpose
712 - fixed crash when using sw_down/up
713 - improved logic for sw_lokey/hikey/up/down/last
714 - added more v1 aliases to parser
715 - reduced memory usage for sfz data
716 - RT-safeness: avoid malloc in audio thread
717 - fixed a bug that could cause voice stealing to fail
718 - optimized sample lookup
719 - improved support for exclusive groups (group, off_by and
720 off_mode)
721 - added support for controller triggered regions (on_locc/on_hicc)
722 - added support for loop_mode=one_shot
723 - fixed playback of 16 bit wav files on big endian CPUs
724 - added support for Ogg Vorbis sample files
725 - use loop markers from sample file if loop_start and loop_end are
726 not set in sfz file
727 - implemeted filters. Filter types: lowpass, bandpass, bandreject
728 and highpass. 1, 2, 4 and 6 pole filters. Opcodes: fil_type,
729 cutoff, resonance, fil_veltrack, fil_keytrack, fil_keycenter,
730 cutoff_cc, cutoff_chanaft.
731 - bugfix: zero ampeg_sustain didn't work
732 - added support for velocity effect on amplifier envelope time
733 (ampeg_vel2attack, ampeg_vel2decay, ampeg_vel2sustain and
734 ampeg_vel2release)
735 - added support for EG hold (ampeg_hold)
736 - added support for sample offset (offset)
737 - use the newly introduced signal units model
738 - implemented opcodes lfoN_delay,
739 lfoN_freq, lfoN_pan, lfoN_cutoff, lfoN_resonance
740 - implemented opcodes ampeg_delay, ampeg_vel2delay
741 pitcheg_delay, pitcheg_start, pitcheg_attack, pitcheg_hold,
742 pitcheg_decay, pitcheg_sustain, pitcheg_release, pitcheg_vel2delay,
743 pitcheg_vel2attack, pitcheg_vel2hold, pitcheg_vel2decay,
744 pitcheg_vel2sustain, pitcheg_vel2release, pitcheg_depth
745 - implemented opcodes pitchlfo_delay, pitchlfo_freq,
746 pitchlfo_depth, fillfo_delay, fillfo_freq, fillfo_depth,
747 amplfo_delay, amplfo_freq, amplfo_depth
748 - implemented opcodes fileg_delay, fileg_start, fileg_attack,
749 fileg_hold, fileg_decay, fileg_sustain, fileg_release, fileg_vel2delay,
750 fileg_vel2attack, fileg_vel2hold, fileg_vel2decay, fileg_vel2sustain,
751 fileg_vel2release
752 - implemented opcode lfoN_wave
753 - implemented opcode pitchlfo_depthccN
754 - implemented opcodes lfoN_phase, lfoN_phase_onccX,
755 lfoN_pitch, lfoN_pitch_onccX
756 - implemented opcodes pitchlfo_fade,
757 fillfo_fade, amplfo_fade, lfoN_fade, lfoN_fade_onccX
758 - implemented opcodes fillfo_freqccN,
759 pitchlfo_freqccN, amplfo_freqccN, lfoN_freq_onccX
760 - implemented opcodes ampeg_delayccN, ampeg_startccN,
761 ampeg_attackccN, ampeg_holdccN, ampeg_decayccN, ampeg_sustainccN,
762 ampeg_releaseccN, egN_timeX_onccY, egN_levelX_onccY
763 - lfoN_* and egN_* opcodes defined in group sections
764 are now taken into account
765 - implemented curves
766 - implemented opcodes volume_onccN, volume_curveccN
767 - implemented opcode volume_smoothccN
768 - implemented opcodes fillfo_depthccN, amplfo_depthccN,
769 lfoN_volume, lfoN_volume_onccX, lfoN_volume_smoothccX,
770 lfoN_freq_smoothccX, lfoN_pitch_smoothccX, lfoN_pan_onccX,
771 lfoN_pan_smoothccX, lfoN_cutoff_onccX, lfoN_cutoff_smoothccX,
772 lfoN_resonance_onccX, lfoN_resonance_smoothccX, lfoN_delay_onccX
773 - implemented opcode end
774 - implemented opcodes egN_amplitude_onccX,
775 egN_volume, egN_volume_onccX, egN_cutoff, egN_cutoff_onccX,
776 egN_pitch, egN_pitch_onccX, egN_resonance, egN_resonance_onccX
777 - implemented opcodes xfin_lokey, xfin_hikey,
778 xfout_lokey, xfout_hikey, xf_keycurve, xfin_lovel, xfin_hivel,
779 xfout_lovel, xfout_hivel, xf_velcurve, xfin_loccN, xfin_hiccN,
780 xfout_loccN, xfout_hiccN, xf_cccurve
781 - implemented opcodes pan_onccN, pan_smoothccN,
782 pan_curveccN, egN_pan, egN_pan_curve, egN_pan_onccX,
783 egN_pan_curveccX, pitch_veltrack
784 - when failed to parse a sfz file
785 print the line number on which the error occurs
786 - use common pool of CC objects to minimize RAM usage
787 - implemented opcodes amplfo_delay_onccN,
788 amplfo_fade_onccN, fillfo_delay_onccN, fillfo_fade_onccN,
789 pitchlfo_delay_onccN, pitchlfo_fade_onccN
790 - implemented opcodes fileg_delay_onccN,
791 fileg_start_onccN, fileg_attack_onccN, fileg_hold_onccN,
792 fileg_decay_onccN, fileg_sustain_onccN, fileg_release_onccN,
793 fileg_depth_onccN, pitcheg_delay_onccN, pitcheg_start_onccN,
794 pitcheg_attack_onccN, pitcheg_hold_onccN, pitcheg_decay_onccN,
795 pitcheg_sustain_onccN, pitcheg_release_onccN, pitcheg_depth_onccN
796 - implemented automatic aliasing *ccN <-> *_onccN
797 - *lfo_freqccN wasn't working when the respective *lfo_freq
798 was not set or was set to zero
799 - lfoN_freq_onccX wasn't working when lfoN_freq
800 was not set or was set to zero
801 - implemented opcodes resonance_onccN, resonance_smoothccN,
802 resonance_curveccN, cutoff_smoothccN, cutoff_curveccN
803 - implemented opcodes fillfo_depthchanaft,
804 fillfo_freqchanaft, amplfo_depthchanaft, amplfo_freqchanaft,
805 pitchlfo_depthchanaft, pitchlfo_freqchanaft
806 - implemented opcodes pitch_onccN,
807 pitch_curveccN, pitch_smoothccN, pitch_stepccN
808 - implemented opcodes volume_stepccN, pan_stepccN,
809 cutoff_stepccN, resonance_stepccN, lfoN_freq_stepccX,
810 lfoN_volume_stepccX, lfoN_pitch_stepccX, lfoN_pan_stepccX,
811 lfoN_cutoff_stepccX, lfoN_resonance_stepccX
812 - implemented opcodes eq1_freq, eq2_freq, eq3_freq,
813 eq1_freqccN, eq2_freqccN, eq3_freqccN, eq1_bw, eq2_bw, eq3_bw,
814 eq1_bwccN, eq2_bwccN, eq3_bwccN, eq1_gain, eq2_gain, eq3_gain,
815 eq1_gainccN, eq2_gainccN, eq3_gainccN
816 - implemented opcodes delay, delay_onccN, delay_random,
817 delay_samples, delay_samples_onccN
818 - implemented opcodes egN_eq1gain, egN_eq2gain, egN_eq3gain,
819 egN_eq1gain_onccX, egN_eq2gain_onccX, egN_eq3gain_onccX, egN_eq1freq,
820 egN_eq2freq, egN_eq3freq, egN_eq1freq_onccX, egN_eq2freq_onccX,
821 egN_eq3freq_onccX, egN_eq1bw, egN_eq2bw, egN_eq3bw, egN_eq1bw_onccX,
822 egN_eq2bw_onccX, egN_eq3bw_onccX, lfoN_eq1gain, lfoN_eq2gain,
823 lfoN_eq3gain, lfoN_eq1gain_onccX, lfoN_eq2gain_onccX, lfoN_eq3gain_onccX,
824 lfoN_eq1gain_smoothccX, lfoN_eq2gain_smoothccX, lfoN_eq3gain_smoothccX,
825 lfoN_eq1gain_stepccX, lfoN_eq2gain_stepccX, lfoN_eq3gain_stepccX,
826 lfoN_eq1freq, lfoN_eq2freq, lfoN_eq3freq, lfoN_eq1freq_onccX,
827 lfoN_eq2freq_onccX, lfoN_eq3freq_onccX, lfoN_eq1freq_smoothccX,
828 lfoN_eq2freq_smoothccX, lfoN_eq3freq_smoothccX, lfoN_eq1freq_stepccX,
829 lfoN_eq2freq_stepccX, lfoN_eq3freq_stepccX, lfoN_eq1bw, lfoN_eq2bw,
830 lfoN_eq3bw, lfoN_eq1bw_onccX, lfoN_eq2bw_onccX, lfoN_eq3bw_onccX,
831 lfoN_eq1bw_smoothccX, lfoN_eq2bw_smoothccX, lfoN_eq3bw_smoothccX,
832 lfoN_eq1bw_stepccX, lfoN_eq2bw_stepccX, lfoN_eq3bw_stepccX
833 - implemented opcodes eq1_vel2freq, eq2_vel2freq,
834 eq3_vel2freq, eq1_vel2gain, eq2_vel2gain, eq3_vel2gain
835 - sfz parser: allow double spaces in sample filenames
836 - sfz parser: allow absolute paths for sample filenames
837 - use linear decay and release for filter and pitch EG
838 - bugfix: only the first amp_veltrack definition in a file was
839 used
840 - bugfix: looping was disabled if loop_start was set to 0
841 - allow regions with end=-1 to turn off other regions using the
842 group and off_by opcodes (#168)
843 - made end=0 play the whole sample
844 - fixed support for lochan and hichan opcodes (#155)
845 - fixed crash when using lochan/hichan opcodes (#187)
846 - sfz parser: allow -200 to 200 for pan_oncc opcode (#182)
847 - added FLAC support (#191)
848 - sfz parser bugfix: lines starting with whitespace were ignored
849 - added amplitude opcode
850 - added support for "#include" instruction
851 (modified patch which was originally posted by Sergey on LS mailing list)
852 - bugfix: generation of velocity curves etc should not be done
853 after each "#include", only after the main file is parsed
854 - bugfix: line numbers in error messages were wrong after "#include"
855 - added support for float and 32 bit sample files
857 * SoundFont format engine:
858 - Initial implementation (not usable yet)
859 - RT-safeness: avoid malloc in audio thread
860 - fixed a bug that could cause voice stealing to fail
861 - fine-tuned amplitude EG (by switching from gig to sfz EG)
862 - initial implementation of Vibrato LFO and Modulation LFO
863 - initial implementation of cutoff filter
864 - use linear decay and release for filter and pitch EG
866 * Host plugins (VST, AU, LV2, DSSI):
867 - AU bugfix: failed to destroy its audio/MIDI devices
868 - Listen to all interfaces on Mac OS X (INADDR_ANY)
869 - VST bugfix: If the host called resume() before and after
870 changing sample rate or block size, the number of channels was
871 incorrectly set to two. This caused silence in Cubase 5.
872 - save engine type (gig, sfz or sf2) in plugin state
873 - VST: when opening Fantasia, look for both 32 and 64 bit Java on
874 64 bit Windows
875 - AU: changed number of output channels from one stereo to 16
876 stereo
877 - VST: made it possible to build the VST plugin for Mac
878 - AU: link AU plugin dynamically if --disable-shared isn't
879 specified
880 - LV2 "state" extension support (patch by David Robillard)
881 - VST bugfix: instrument loading hang and crashed the host when
882 the plugin was loaded a second time (#174)
883 - plugin bugfix: instrument loading hang when the plugin was
884 loaded a second time (this time it's for Linux and Mac, previous
885 similar fix was for Windows)
886 - thread safety fixes for the instrument loading thread
887 - LV2: use the new lv2 package if present
888 - VST: try to open Fantasia automatically on Linux and Mac too (on
889 Linux, the Fantasia jar should be placed in <prefix>/share/java)
890 - VST: fixed crashes on Linux Ardour and EnergyXT
891 - DSSI bugfix: it wasn't possible to change engine type. The MIDI
892 port and audio channel routing for DSSI plugins are now visible.
893 - LV2: use urid and atom extensions instead of deprecated uri-map
894 and event
895 - LV2: lv2 package 1.0 is now required to build the LV2 plugin
896 - LV2: changed number of output channels to 16 stereo, just like
897 the VST and AU plugins
898 - LV2: fixed save/restore of SFZ state (patch by David Robillard)
899 - LV2: made LV2 plugin buildable on Windows and Mac
900 - VST: implemented retrieval and switching of programs using the
901 sampler's internal MIDI instrument mapping system
903 * Instrument editor interface:
904 - Changed instrument editor plugin interface, providing additional
905 informations like the EngineChannel for which the instrument editor was
906 spawned for. This allows the instrument editors to interact more actively
907 with the sampler.
909 * MIDI driver:
910 - ALSA MIDI driver supports now "NAME" device parameter, for overriding
911 the ALSA sequencer client name
912 - removed limit of maximum amount of MIDI ports per MIDI device, since
913 there is no reason for this limit
914 - MME: fixed memory handling bug found with cppcheck
915 - MME: removed compiler warning
916 - CoreMIDI: implemented driver specific port parameter "CORE_MIDI_BINDINGS",
917 which allows to retrieve the list of CoreMIDI clients / ports and to connect
918 to them a la JACK, via the usual sampler APIs
919 - CoreMIDI: added driver specific port parameter "AUTO_BIND", if enabled
920 the driver will automatically connect to other CoreMIDI clients' ports
921 (e.g. external MIDI devices being attached to the Mac)
922 - added support for MIDI note on velocity filter
923 - CoreMIDI: fixed memory deallocation error
924 - Fixed variable underflow in VirtualMidiDevice, which caused graphical
925 virtual keyboards in frontends / instrument editors being stuck.
926 - Bugfix in VirtualMidiDevice: process note on with velocity 0 as note off.
927 - Implemented missing handling of MIDI "running status".
928 - CoreMIDI fix: a MIDIPacket can contain more than one event per packet.
929 - MME bugfix: driver wasn't closed properly
930 - CoreMIDI: automatically connect to all input sources by default (driver
931 parameter "AUTO_BIND").
932 - CoreMIDI: fixed auto bind feature to CoreMIDI ports that go online
933 - CoreMIDI: fixed minor memory leak
935 * audio driver:
936 - ASIO driver fixes for newer gcc versions (fix from PortAudio)
937 - JACK audio: react on sample rate changes.
938 - JACK audio: react on buffer size changes.
939 - JACK audio: jack_port_get_buffer() was cached and called outside
940 RT context.
941 - ASIO driver: removed compiler warnings
942 - CoreAudio: fixed minor error handling bug
943 - ASIO driver: be more verbose when no ASIO card could be found (fixes #203)
944 - JACK audio: return the JACK server's current sample rate as default value
945 for audio device parameter "SAMPLERATE" (fixes #166).
947 * LSCP server:
948 - added support for sending MIDI CC messages via LSCP command
949 "SEND CHANNEL MIDI_DATA CC <sampler-chan> <ctrl> <val>"
950 - added LSCP command "GET AVAILABLE_EFFECTS"
951 - added LSCP command "LIST AVAILABLE_EFFECTS"
952 - added LSCP command "GET EFFECT INFO <effect-index>"
953 - added LSCP command "CREATE EFFECT_INSTANCE <effect-index>"
954 - added LSCP command
955 "CREATE EFFECT_INSTANCE <effect-system> <module> <effect-name>"
956 - added LSCP command "DESTROY EFFECT_INSTANCE <effect-instance>"
957 - added LSCP command "GET EFFECT_INSTANCES"
958 - added LSCP command "LIST EFFECT_INSTANCES"
959 - added LSCP command "GET EFFECT_INSTANCE INFO <effect-instance>"
960 - added LSCP command
961 "GET EFFECT_INSTANCE_INPUT_CONTROL INFO <effect-instance> <input-control>"
963 <effect-instance> <input-control> <value>"
964 - added LSCP command "GET SEND_EFFECT_CHAINS <audio-device>"
965 - added LSCP command "LIST SEND_EFFECT_CHAINS <audio-device>"
966 - added LSCP command "ADD SEND_EFFECT_CHAIN <audio-device>"
967 - added LSCP command
968 "REMOVE SEND_EFFECT_CHAIN <audio-device> <effect-chain>"
969 - added LSCP command
970 "GET SEND_EFFECT_CHAIN INFO <audio-device> <effect-chain>"
971 - added LSCP command "APPEND SEND_EFFECT_CHAIN EFFECT <audio-device>
972 <effect-chain> <effect-instance>"
973 - added LSCP command "INSERT SEND_EFFECT_CHAIN EFFECT <audio-device>
974 <effect-chain> <effect-chain-pos> <effect-instance>"
975 - added LSCP command "REMOVE SEND_EFFECT_CHAIN EFFECT <audio-device>
976 <effect-chain> <chain-pos>"
977 - added LSCP command "SET FX_SEND EFFECT <sampler_channel>
978 <fx_send_id> <effect_chain> <chain_pos>"
979 - added LSCP command "REMOVE FX_SEND EFFECT <sampler_channel> <fx_send_id>"
983 - provide comprehensive error messages on LSCP syntax errors
984 (suggesting expected next non-terminal symbols)
985 - Fixed client connection not being closed after network errors.
987 * LSCP shell:
988 - Added support for auto correction of obvious and trivial syntax mistakes.
989 - Added support for auto completion by tab key.
990 - Show currently available auto completion while typing.
991 - Added support for browsing command history with up / down keys.
992 - Show all possible next symbols immediately right to the current command
993 line while typing (no double tab required for this feature, as it would
994 be the case in other shells).
995 - Added support for moving cursor left/right with arrow keys.
996 - Added support for built-in LSCP reference documentation, which will
997 automatically show the relevant LSCP reference section on screen as soon
998 as one specific LSCP command was detected while typing on the command
999 line.
1001 * Real-time instrument scripts:
1002 - Implemented built-in script array variable %CC.
1003 - Implemented built-in script int variable $CC_NUM.
1004 - Implemented built-in script int variable $EVENT_NOTE.
1005 - Implemented built-in script int variable $EVENT_VELOCITY.
1006 - Implemented built-in script constant variable $VCC_MONO_AT.
1007 - Implemented built-in script constant variable $VCC_PITCH_BEND.
1008 - Implemented execution of script event handler "init".
1009 - Implemented execution of script event handler "controller".
1010 - Implemented execution of script event handler "note".
1011 - Implemented execution of script event handler "release".
1012 - Implemented built-in script function "play_note()" (only two of the
1013 max. four function arguments are currently implemented yet though).
1014 - Implemented built-in script int variable $EVENT_ID.
1015 - Implemented built-in script function "ignore_event()"
1016 - Implemented built-in script function "ignore_controller()" (may have one
1017 or no argument).
1018 - Implemented built-in script function "set_controller()".
1019 - Added extended script VM for the Gigasampler/GigaStudio format sampler
1020 engine, which extends the general instrument script VM with Giga format
1021 specific variables and functions.
1022 - Giga format scripts: added built-in script int constant variables
1033 - Giga format scripts: Implemented built-in script function
1034 "gig_set_dim_zone(event_id, dimension, zone)".
1035 - Implemented built-in script int array variable %KEY_DOWN.
1036 - Implemented built-in script function "abs()".
1037 - Implemented built-in script function "random()".
1038 - Implemented built-in script function "num_elements()".
1039 - Implemented built-in script function "note_off()".
1040 - Implemented built-in script function "set_event_mark()".
1041 - Implemented built-in script function "delete_event_mark()".
1042 - Implemented built-in script function "by_marks()".
1043 - Added built-in script int const variables $MARK_1 to $MARK_28.
1044 - Built-in script functions "ignore_event()", "note_off()" and
1045 "gig_set_dim_zone()" now also accept an array of event IDs as argument
1046 (i.e. return value of new script function "by_marks()").
1047 - Pass/preserve polyphonic variable data from respective "note" event
1048 handler to "release" event handler.
1049 - Fixed crash when using built-in script function "by_marks()".
1051 * Bug fixes:
1052 - Fixed crash which may occur when MIDI key + transpose is out of range
1053 - minor valgrind fixes
1054 - fixed crash which occurred when changing an already deployed sampler
1055 channel to a different engine type
1056 - fixed crash when deleting a sampler channel or changing engine
1057 type while an instrument load was in progress
1058 - bugfix: playing a note while changing the instrument could cause
1059 a crash, or give "Handing back unknown region" error messages
1060 - bugfix: calling SET VOICES (which Fantasia does on start and
1061 refresh) could cause instruments to be unloaded
1062 - Mac OS X: fixed name collision of enum in EffectControl and
1063 wrong name of destructor in AudioOutputDeviceCoreAudio.cpp
1064 - Mac OS X: fixed hanging threads
1065 - Windows: disabled the previous thread fix on non-Mac systems, as
1066 it caused hanging threads on Windows
1067 - Fixed possible crashes due to corrupted MIDI/audio device list
1068 after MIDI/audio device creation failure
1069 - When creating MIDI instrument map entries with "PERSISTENT" type, the
1070 instruments were uselessly precached with zero samples, however it still
1071 took the full preloading time and on 1st program change the respective
1072 instrument was completely reloaded again.
1073 - fixed handling of rapid bank select and program change messages
1074 sent to the same sampler channel (patch from the Open Octave
1075 project, slightly adjusted)
1076 - fixed crash when trying to create an effect instance with controls
1077 which min and/or max values depend on the sample rate
1078 - fixed bug #162
1079 - bugfix: LADSPA_PATH was not evaluated correctly when containing
1080 multiple paths (#165)
1081 - thread safety fixes for the instrument loading thread
1082 - bugfix: instrument loading crashed for sfz and sf2 in Ardour
1083 (#176)
1084 - more thread safety fixes for the instrument loading thread
1085 - sfz/sf2 engine: fixed crash when using small audio fragment size
1086 - Mac OS X: fixed crash when unloading plugin on 10.7 and later
1087 - Mac OS X: fixed process hang when unloading 32-bit plugin (bug
1088 introduced in previous fix)
1089 - fixed crash when a channel received a program change while
1090 playing a note in a key group
1091 - fixed erroneous error message piping in VoiceBase.h
1092 (labelled "Disk stream not available in time")
1093 - Update effects on sample rate & period size changes (to avoid
1094 crashes and noise on such transitions).
1095 - ignore missing LADSPA paths without ignoring valid LADSPA paths
1096 (fixes #208)
1097 - Fixed bug in LSCP grammar definition which caused a statement like
1098 "GET SERVER INFOasdf\n" to be accepted as valid statement (was so far
1099 practically irrelevant, however it caused problems with the new LSCP
1100 shell's auto completion feature).
1101 - Fixed MIDI program change messages being ignored if quickly executed
1102 after each other (fixes #231).
1104 Version 1.0.0 (31 July 2009)
1106 * packaging changes:
1107 - autoconf bugfix: the PKG_CONFIG variable wasn't initialized properly,
1108 causing e.g. the libgig test to fail when
1109 "./configure --disable-jack-driver" was used
1110 (patch by Alexis Ballier)
1111 - fixed compilation with gcc 4.3
1112 - fixes for building on OS X (thanks to Ebrahim Mayat for testing)
1113 - fixed configure so it detects x86_64 (#107)
1114 - fixes for building with newer MinGW versions
1115 - fix for building with bison 2.4 (#111)
1116 - fixed building with libgig installed in a non-standard directory
1117 - minor fix in configure for mmsystem.h detection on MinGW
1118 - Windows: look for editor plugins and Fantasia using base
1119 directory of liblinuxsampler dll (look in the same directory and one
1120 directory above)
1121 - configure script fix: removed unconditional use of SSE
1122 - fixed building with sqlite installed in a non-standard directory
1123 - when cross-compiling, don't try to create instruments.db
1124 - fix for new mingw-w64 version, which has usleep
1126 * general changes:
1127 - bugfix: on some POSIX systems instrument editor plugins refused to
1128 load as we used a non-portable Linux specific struct field
1129 (fixes bug #70, patch by Ronald Baljeu)
1130 - fixed endless loop which occured when loading an instrument editor
1131 plugin DLL on Windows systems failed
1132 - fixed memory leaks that occurred when liblinuxsampler was unloaded
1133 - optimized the SynchronizedConfig class so it doesn't wait
1134 unnecessarily long after an update
1135 - added support for notifying instrument editors on note-on / note-off
1136 events (e.g. to highlight the pressed keys on the virtual keyboard
1137 of gigedit)
1138 - added support for triggering notes by instrument editors (see above)
1139 - be verbose on DLL load errors (on Linux)
1140 - fixed di-harmonic triangle LFO implementation (this LFO implementation
1141 is only used by very few systems, most pick the int math
1142 implementation, which is usually faster)
1143 - fixes for audio drivers with varying buffer sizes
1144 - experimental support for running LinuxSampler as a DSSI, LV2 and
1145 VST plugin
1146 - notification events for stream/voice count statistics are now sent
1147 only when there are actual changes
1148 - added memory ordering constraints to improve stability on
1149 multi-core and multi-cpu systems
1150 - maximum voices and disk streams can now be altered at runtime
1151 - fixed CPU feature detection on x86_64 (maybe fixes #108)
1152 - automatic stacktrace mechanism is now turned off by default and can
1153 be switched on by command line option "--stacktrace" (the automatic
1154 stacktrace mechanism seems to be broken on most systems at the moment)
1155 - C++ API method InstrumentManager::LaunchInstrumentEditor() now returns
1156 a pointer to the launched InstrumentEditor object
1157 - added optional 3rd party user data parameter for following
1158 liblinuxsampler C++ API methods: InstrumentEditor::Main(),
1159 InstrumentEditor::Launch(),
1160 InstrumentManager::LaunchInstrumentEditor()
1161 - theoretical fix: made SynchronizedConfig follow C++0x memory
1162 model more strictly
1163 - fixes for using large audio device buffers
1164 - Windows: add the installation directory to the DLL search path
1165 when loading an editor plugin (solves problems with VST and
1166 gigedit on systems with other GTK versions installed)
1167 - updated linuxsampler man page
1169 * audio driver:
1170 - removed the nonsense audio channel constraint (which was hard coded to
1171 max. 100 audio channels) for most audio drivers
1172 - JACK audio driver did not offer a device parameter "SAMPLERATE" as
1173 opposed to the LSCP specs
1174 - bugfix: the SAMPLERATE parameter of some drivers (e.g. JACK)
1175 reflected the wrong value
1176 - fixed a memory management bug in ASIO driver
1177 - Makefile fix: JACK_CFLAGS wasn't used
1178 - JACK: use jack_client_open instead of the deprecated
1179 jack_client_new
1180 - added (experimental) CoreAudio driver
1181 - applied old fixes to the ASIO driver that were included in the
1182 previous binary release but accidentally never committed to CVS
1183 (fixes #117)
1184 - fixes for ASIO on mingw-w64 (iasio wrapper is not needed on
1185 win64)
1186 - VST: added support for sample rate and buffer size changes
1187 - VST: close editor (Fantasia) when the VST is removed
1188 - VST: avoid opening Fantasia more than once for each VST instance
1189 - VST: export main function as "main" on Linux too (fix for
1190 energyXT)
1191 - VST: changed number of output channels from one stereo to 16
1192 stereo
1193 - added channel routing, fxsends and midi maps to the settings
1194 stored in the plugin state
1195 - performance optimization of AudioChannel::MixTo() and
1196 AudioChannel::CopyTo() methods using GCC vector exensions
1197 (if available)
1198 - ASIO fixes: avoid initializing the device twice, avoid throwing
1199 exception when getting parameters from a disconnected device
1201 * MIDI driver:
1202 - added JACK MIDI driver
1203 - dispatch bank select as ordinary CC as well, the user might seriously
1204 want to (mis)use it for some purpose ("fixed" in all current MIDI
1205 input drivers: ALSA, CoreMIDI, JACK, MidiShare, MME)
1206 - bugfix: pitch bend wasn't working with JackMidi, VST, LV2, MME,
1207 CoreMidi or AU
1208 - fixed mingw-w64 compilation error in MME driver
1209 - made program change handling in MIDI thread real-time safe by
1210 moving the logic to a non-RT thread
1211 - fixed minor memory leak in ALSA driver
1213 * instruments database:
1214 - avoid time consuming samples scanning when adding instruments
1215 to the instruments database
1216 - added support for handling lost files in the instruments database
1217 - Implemented option for adding instruments in separate directories
1218 in the instruments database
1219 (patch by Chris Cherrett & Andrew Williams, a bit adjusted)
1220 - work-around for missing fnmatch function on Windows to make
1221 instrument database compilable
1222 - added instrument database support on Windows
1223 works with both standalone mode and VST plugin.
1224 the instruments DB file is located in
1225 %USERPROFILE%\.linuxsampler\instruments.db which allows different
1226 databases for each windows user
1227 if no DB is present it automatically creates the .linuxsampler subdir
1228 and then creates an empty DB
1229 - fixed recursive import, which was broken on Windows
1231 * Gigasampler format engine:
1232 - fixed a memory leak that could happen when a channel was deleted
1233 while notes were playing
1234 - made it possible to load an instrument even if the audio thread
1235 isn't running
1236 - added partial support for the "Controller Triggered" MIDI rule,
1237 enough for piano gigs with pedal noise samples
1238 - minor fix: only mark FX sends as being modified if really the
1239 respective FX send MIDI controller was used
1240 - added support for GM global device volume SysEx message
1241 (can be selected at compile time to either apply globally to the
1242 whole sampler [default] or only to the sampler channels that are
1243 connected to the respective MIDI input port the SysEx message arrived
1244 on)
1245 - bugfix: notes triggered at position 0 in the audio buffer were
1246 sometimes wrongly killed in the same buffer, causing no sound to
1247 be played
1248 - added support for chromatic / drumkit mode Roland GS Sysex message
1249 which is usally used to switch a part between chromatic sounds and
1250 drumkit sounds (as opposed to their common default setting of MIDI
1251 part 10 being a drumkit part and all other ones chromatic parts), in
1252 LS however one can switch between the first 16 MIDI instrument maps
1253 defined for the sampler
1254 - made it possible to create multiple sample channels even if the
1255 audio thread isn't running
1256 - fixed crash when removing channel with active voices (#116)
1257 - bugfix: on sample reference changes (instrument editor), only
1258 un-cache the respective sample if it's really not used by any
1259 sampler engine anymore
1260 - re-cache samples in case they were changed by an instrument editor,
1261 e.g. when a sample was added while playing (#82)
1262 - fixed hanging notes which occured when note-off event had the exact
1263 same time stamp as the voice's note-on event and occured both in the
1264 same audio fragment cycle (fixes bug #112)
1265 - added support for the "fine tune" and "pitch bend range"
1266 instrument-level gig parameters
1267 - fixed minor artifacts in pitch bend handling
1268 - added support for GS Reset SysEx message
1269 - allow gig files to use unlimited downward pitch shifting
1270 - added a limit check for upward pitch shifting
1271 - bugfix: sometimes, when playing a note twice fast, the second
1272 note was silent
1273 - fixed crash happening when a pitch bend event arrived at the
1274 same time a new instrument was loading
1276 * LSCP server:
1277 - added new LSCP event "CHANNEL_MIDI" which can be used by frontends to
1278 react on MIDI data arriving on certain sampler channels (so far only
1279 Note-On and Note-Off events are sent via this LSCP event)
1280 - added new LSCP event "DEVICE_MIDI" which can be used by frontends to
1281 react on MIDI data arriving on certain MIDI input devices (so far only
1282 Note-On and Note-Off events are sent via this LSCP event)
1283 - added new LSCP commands: FIND LOST DB_INSTRUMENT_FILES and
1285 - added new LSCP command: SEND CHANNEL MIDI_DATA which can be used by
1286 frontends to send MIDI messages to specific sampler channel
1287 - added two additional fields to GET FILE INSTRUMENT INFO command -
1289 - bugfix: the bank number provided by MIDI_INSTRUMENT_INFO notifications
1290 was incorrect
1291 - Optimized the retrieval of the MIDI instrument mappings
1292 - added new LSCP command "SET VOICES" to globally alter the maximum
1293 amount of voices, added new LSCP event "GLOBAL_INFO:VOICES" which will
1294 be triggered respectively
1295 - added new LSCP command "SET STREAMS" to globally alter the maximum
1296 amount of disk streams, added new LSCP event "GLOBAL_INFO:STREAMS"
1297 which will be triggered respectively
1298 - bugfix: retry if "select" returns EINTR (this fixes a crash when
1299 a gigedit file dialog is opened)
1300 - close all connections when LSCPServer is deleted
1301 - hard close of all sockets on exit
1302 - bugfix: SET CHANNEL MIDI_INPUT_TYPE didn't work with the MME
1303 driver
1305 * Bug fixes:
1306 - fixed a crash which occurs when removing a sampler channel waiting
1307 to start instrument loading after another channel
1308 - fixed a crash which occurs when removing a sampler channel with
1309 instrument loading in progress (bug #113)
1310 - fixed termination caused by uncaught exception when adding MIDI
1311 instrument with PERSISTENT load mode
1312 - fixed possible iterator invalidations when resetting the sampler
1313 - fixed memory leaks when issuing the following LSCP commands:
1318 - fixed possible compilation error when sqlite is not present
1319 - fixed orphaned pointers when setting maximum voices limit (bug #118)
1320 - fixed crash when changing the audio output device of a sampler
1321 channel with loaded instrument and start playing notes
1322 - fixed endless loop in Engine::SuspendAll() (bug #120)
1323 - fixed a low-level atomic load function that was broken on 64-bit
1324 PowerPC, which probably could cause crashes on that platform
1325 - fixed a memory management error which could cause a crash when a
1326 plugin was unloaded
1327 - bugfix: two private structs had the same name, which could cause
1328 problems if the linker chose the wrong constructor
1329 - fixed low-level ConditionServer usage bug that caused lockups on
1330 Windows
1333 Version 0.5.1 (6 December 2007)
1335 * packaging changes:
1336 - added autoconf checks for pthread library
1337 - added autoconf check for pthread bug found on certain NPTL-enabled
1338 glibc versions (see Gentoo bug report #194076)
1339 - added autoconf checks for MS Windows
1341 * general changes:
1342 - bugfix: the thread used by an editor plugin didn't die when the
1343 editor closed
1344 - bugfix: Ringbuffer.h: fill_write_space_with_null() did not zero
1345 out all the space. operator--() did not apply size_mask after
1346 decrementing the read_ptr. DEFAULT_WRAP_ELEMENTS set to 0 in
1347 order to avoid problems with the _NonVolatileReader functions.
1348 - bugfix: Stream.h: added a missing divide by BytesPerSample in
1349 GetWriteSpace(). Since this function is currently only used in
1350 the stream's qsort() compare function, it didn't trigger any
1351 bugs.
1352 - Resampler.h, Synthesizer.h: cubic interpolation now works in
1353 24bit mode too. Faster method to read 24bit words on little
1354 endian machines (x86): replaced 3 byte reads + shifts with a 1
1355 unaligned 32bit read and shift
1356 - experimental support for MS Windows (MIDI input via MME, AUDIO
1357 output via ASIO)
1358 - made handling of SIGINT signal (Ctrl-C) a bit more robust
1359 - support for monitoring the total number of active disk streams
1360 (new LSCP commands: GET TOTAL_STREAM_COUNT,
1363 * AUDIO driver:
1364 - added Windows ASIO low latency audio driver
1366 * MIDI driver:
1367 - added MME Windows MIDI driver
1369 * LSCP server:
1370 - added support for Windows style path / filenames, however with
1371 forward slash path separators instead of backslash
1372 (i.e. "C:/foo/bar.gig")
1373 - allow naughty liblscp to send non-string device parameters within
1374 apostrophes as well
1375 - added new LSCP commands: "GET FILE INSTRUMENTS <file>",
1376 "LIST FILE INSTRUMENTS <file>" and
1377 "GET FILE INSTRUMENT INFO <file> <index>" for retrieving informations
1378 about an arbitrary instrument file on the system where the sampler is
1379 running on
1381 Version 0.5.0 (15 October 2007)
1383 * packaging changes:
1384 - config.h is not going to be installed along with liblinuxsampler's
1385 API header files anymore
1386 - only the API relevant header (and source) files will be exported to
1387 the Doxygen API documentation (explicitly listed in Doxyfile.in)
1388 - added completely new XCode project files for Mac OSX which is now
1389 capable to execute our autoconf environment, thus no need anymore
1390 to maintain the compile time configuration file (osx/version.h) for
1391 OSX manually (patch by Toshi Nagata)
1392 - fixed buggy boolean --enable-foo / --disable-foo configure script
1393 parameters
1394 - global.h now only covers global definitions that are needed for the
1395 C++ API header files, all implementation internal global definitions
1396 are now in global_private.h
1397 - atomic.h is not exposed to the C++ API anymore
1398 - no need to include config.h anymore for using LS's API header files
1399 - fixed warnings in API doc generation
1401 * general changes:
1402 - replaced the old, confusing MIDI program change mechanism by a
1403 flexible MIDI instrument mapper which allows to map arbitrary
1404 (MIDI bank, MIDI program) pairs with arbitrary
1405 (engine type, instrument file, file index) triplets which will be
1406 loaded on the respective channel when such MIDI program change
1407 messages arrive, beside that, each entry allows to define a life-time
1408 strategy for the instrument, so the user can define whether the
1409 instrument should i.e. be permanently loaded (to be able to switch
1410 quickly among sounds) or i.e. loaded just on demand when the
1411 respective program change arrives (to safe RAM space), as well as a
1412 global volume factor for each entry, so the user can adjust the volume
1413 dynamic of his mapped instrument collection without having to modify
1414 the instrument files, also one can manage arbitrary amount of such
1415 MIDI instrument maps and assign each sampler channel individually
1416 a certain map, so that i.e. separation of normal instruments and
1417 drumkits is possible
1418 - new notification events for tracking changes to audio/MIDI devices,
1419 MIDI instrument maps, MIDI instruments, FX sends, global volume.
1420 - sampler was limited to load max. 200 instruments in the background
1421 due to a constant size RingBuffer FIFO which is now replaced by a
1422 dynamic (unlimited) size std::list FIFO
1423 - added FX sends, these allow to route audio signals to arbitrary audio
1424 output channels for being processed by external effect processors
1425 (i.e. jack-rack), the send levels are controllable via arbitrary MIDI
1426 controllers
1427 - global (sampler wide) volume can now be controlled at runtime
1428 - Implemented new, improved notification system
1429 - fixed compilation errors regarding OSX
1430 (patch by Toshi Nagata)
1431 - implemented instruments database
1432 - added support for escape sequences to the instruments database
1433 - added highly experimental support for on-the-fly instrument editing
1434 within the sampler's process (by using instrument editor plugins),
1435 you'll notice the new "Registered instrument editors:" message on
1436 startup, added a new LSCP command:
1437 "EDIT CHANNEL INSTRUMENT <sampler-channel>"
1438 to spawn a matching instrument editor for the instrument on the
1439 given sampler channel, the plugin path can be overridden at compile
1440 time with ./configure --enable-plugin-dir=/some/dir
1441 - added experimental code for synchronizing instrument editors hosted
1442 in the sampler's process to safely edit instruments while playing
1443 without a crash (hopefully) by either suspending single regions
1444 wherever possible or if unavoidable whole engine(s)
1445 - fixed several issues in fundamental "Thread" class: set scheduling
1446 policy and priority on thread level, set a minimum stack size for
1447 thread (TODO: a reasonable value yet to be tested), bugfix: non-RT
1448 threads simply inherited properties of starting thread instead of
1449 setting their own policy and priority
1450 - minor fix in our automatic stack trace mechanism on crashes, the
1451 main process did not wait for the stack trace process to finish
1452 its output
1453 - fixed some minor memory leaks
1454 - reenabled assembly features support, at the moment only for
1455 enabling a fast denormal FPU mode (x86 platforms supporting SSE2)
1456 - minor assembly fix in x86 features detection (don't use the PIC
1457 register, to avoid relocations in the text segment at runtime)
1458 - POSIX callback functions of Thread.h are hidden
1460 * MIDI driver:
1461 - dispatch bank select (MSB and LSB) messages
1463 * audio driver:
1464 - the ALSA audio output driver parameters now reflect the correct
1465 parameter value ranges for the respective selected sound card
1466 (patch by Till Wimmer, a bit fixed and extended)
1468 * Gigasampler format engine:
1469 - if a filter is used and EG2 finishes before EG1, let the voice
1470 die when EG2 has finished (this fixes a problem with clicks and
1471 voice starvation for some gigs)
1472 - playback is no longer disabled during instrument loading
1473 - all notes playing on a channel that changes its instrument keep
1474 playing with the old instrument until they get a note off
1475 command
1476 - EG fix: a release value of zero could cause noises or crash
1477 - handle MIDI coarse tuning messages (MIDI RPN #0 MSB #2 LSB)
1478 - EG fine tuning: when attack is zero the EG starts at a level
1479 above max sustain level, which means that there is a decay phase
1480 even if sustain is 100%
1481 - more EG fixes: the level could sometimes go below zero and cause
1482 noises or crashes
1483 - minor fix of EGDecay (patch by Toshi Nagata)
1484 - fixed compiler error when --enable-override-filter-type was
1485 supplied to the configure script (fixes #46)
1486 - disk thread: queue sizes are now proportional to CONFIG_MAX_STREAMS
1487 instead of a fix value
1488 - behavior fix: on MIDI CC# 65 (portamento on / off), 126 (mono mode),
1489 127 (solo mode) only kill voices if the respective mode really
1490 changed
1492 * LSCP server:
1493 - fixed compile time error for old Bison versions
1494 (i.e. v1.28 found on MacOS 10.4, patch by Toshi Nagata)
1495 - parser now supports extended ASCII character set
1496 (up to ASCII code 255, i.e. includes now umlauts and accents)
1497 - filename arguments in LSCP commands now allow to use escape
1498 sequences, that is directly literal as one of: \', \", \\, \n, \r,
1499 \f, \t, \v, or as octal ASCII code value like \132, or as hex ASCII
1500 code value like \xf2) (fixes bug #24)
1501 - the following LSCP commands now also support escape sequences for at
1502 least one of their text-based parameters (i.e. name, description):
1509 - returns verbose syntax errors (line and column where syntax error
1510 occured, the unexpected character and the actually expected, possible
1511 character(s), the latter only if less than 5 possibilities)
1512 - made sure that LSCP syntax is not affected by gigedit locale
1513 settings
1514 - bugfix regarding strings parameter lists: all comma separated lists
1515 of strings were treated as being one string containing commas
1516 (fixes #57)
1518 * Bug fixes:
1519 - fixed crash occurring on certain LSCP scripts (Bug 39)
1520 - another thread safety fix for lscp "load engine" and "set
1521 channel audio output device"
1522 - fixed a crash which occurs when reassigning the same engine
1523 on a sampler channel with connected MIDI device
1524 - fixed a crash which occurs when changing the number of ports of a MIDI
1525 device connected to a sampler channel to number less then or equal
1526 to the index of the port to which the sampler channel is connected.
1527 - The previous bindings were not been disconnected when altering
1528 the ALSA_SEQ_BINDINGS parameter. Introduced a NONE keyword for
1529 unsubscribing from all bindings (e.g. ALSA_SEQ_BINDINGS=NONE).
1530 - The active stream/voice count statistic was incorrect.
1531 - notification events were not been sent for some sampler
1532 channel changes
1533 - added default min and max values to restrict the number of allowed
1534 audio output channels and MIDI input ports
1535 - the connection to the PCM interface is now closed when destroying
1536 an audio output device
1537 - files with slash in their path or filename could not be loaded
1539 * test cases:
1540 - updated and fixed (haven't been touched in a while)
1542 Version 0.4.0 (24 November 2006)
1544 * packaging changes:
1545 - changed deprecated copyright attribute to license; added ldconfig
1546 to post-(un)install steps; added devel package for liblinuxsampler;
1547 to linuxsampler.spec (RPM)
1548 - install necessary development header files for allowing 3rd party
1549 applications to link against liblinuxsampler
1550 - liblinuxsampler's API documentation can be generated with 'make docs'
1551 (Doxygen required)
1552 - added benchmark to automatically detect the best triangle LFO
1553 implementation (currently either an integer math solution or a
1554 di-harmonic approximation), automatic detection can be overriden
1555 with --enable-signed-triang-algo=x and --enable-unsigned-triang-algo=x
1556 configure script argument though (mandatory for cross-compilation)
1557 - do not automatically pick optimized gcc flags if the user already
1558 provided some on his own (as CXXFLAGS)
1559 - added compile time option to disable processing of All-Notes-Off MIDI
1560 messages
1561 - added compile time options to allow disabling the various audio and
1562 MIDI drivers
1563 - fixed automatic GCC CPU switch detection on PPC
1564 (patch by Ebrahim Mayat)
1566 * Gigasampler format engine:
1567 - extensive synthesis optimization
1568 (reimplementation of EGs and LFO(s), removed synthesis parameter
1569 prerendering and the synthesis parameter matrix in general, splitting
1570 each audio fragment into subfragments now where each subfragment uses
1571 constant pitch and filter coefficients. The volume coefficient is
1572 linearly interpolated inside a subfragment, unless
1573 --disable-interpolate-volume is set.)
1574 - fine tuning of the EG modulation parameters
1575 - improved filter cutoff calculation by adding support for the
1576 following gig parameters: Cutoff freq (used when no cutoff
1577 controller is defined), Control invert, Minimum cutoff, Velocity
1578 curve and Velocity range. The keyboard tracking now scales
1579 cutoff frequency instead of resonance.
1580 - added support for gig parameter Resonance.
1581 - fixed bug in sysex handling (patch by Juan Linietsky)
1582 - added global volume attenuation of -9 dB (0.35f) to prevent clipping
1583 which can be overridden with --enable-global-attenuation
1584 - EG fixes: made the length of "attack hold" stage more
1585 accurate. Release stage can now start before attack stage
1586 ends. Cancel release didn't work when sustain was zero. Attack
1587 time now has a minimal value to prevent clicks.
1588 - fixed pitch changes larger than one octave
1589 - fixed EG3 (pitch envelope) synthesis which was neutral all the time
1590 - implemented portamento mode and solo mode (a.k.a 'mono mode'):
1591 all modes can be altered via standard GM messages, that is CC5 for
1592 altering portamento time, CC65 for enabling / disabling portamento
1593 mode, CC126 for enabling solo mode and CC127 for disabling solo mode
1594 - fine tuning of the curves for volume (CC7), pan (CC10 and gig
1595 parameter) and crossfade
1596 - added support for the "attenuation controller threshold" gig
1597 parameter
1598 - added smoothing of volume changes caused by control change
1599 messages
1600 - sample loop parameters are now taken from the DimensionRegion
1601 instead of the wave chunk
1602 - fixed keyswitching for v3 gigs with a number of keyswitch splits
1603 not equal to a power of two
1604 - reimplementation of the filter algorithm. The new filter is more
1605 accurate and supports all gig filter types, including bandreject
1606 and lowpass turbo.
1607 - real support for 24 bit samples - samples are not truncated to
1608 16 bits anymore
1609 - support for aftertouch (channel pressure, not polyphonic
1610 aftertouch)
1611 - LFO1 behaviour fixed (dampening from max volume instead of
1612 amplifying from 0)
1614 * LSCP server:
1615 - fixed application exit on broken pipe error (fixes bug #20)
1616 - fixed the notification messages delay due to lack of
1617 network activity (fixes bug #26)
1618 - fixed parser bug which occured on space(s) within device parameters
1621 * audio driver:
1622 - added aRts audio output driver (by no means RT safe)
1624 * MIDI driver:
1625 - fixed legacy sysex code which caused dispatching of MIDI SysEx
1626 messages several times instead of once
1627 - API extension for MIDI drivers which already supply exact time stamps
1628 for events (i.e. for offline rendering based MIDI drivers)
1629 - added checks for bad MIDI data
1631 * linuxsampler application:
1632 - show available sampler engine types on startup
1634 * general changes:
1635 - support for muting sampler channels and solo mode of the same, two new
1636 LSCP commands ("SET CHANNEL MUTE" and "SET CHANNEL SOLO") and two new
1637 fields ("MUTE" and "SOLO") for command "GET CHANNEL INFO" were
1638 introduced for this, the behavior is the same like on a mixer console
1639 (patch by Grigor Iliev, a bit adjusted). Also added configure option
1640 --enable-process-muted-channels which can be used to enable the
1641 processing of muted channels.
1642 - support for sostenuto pedal
1643 - support for monitoring the total number of active voices
1646 - fixed some memory management errors
1647 - fixed some concurrency problems that could lead to crashes when
1648 LSCP commands were executed
1649 - fixed crash when instrument loading failed previously on the same
1650 sampler channel (fixes bug #36)
1652 Version 0.3.3 (15 July 2005)
1654 * packaging changes:
1655 - fixed compilation with gcc 4.0
1657 * Gigasampler format engine:
1658 - LFO preprocessing was performed even though the respective LFO was
1659 disabled by the instrument patch (minor efficiency fix)
1660 - if period time of chosen audio device is too small (< MIN_RELEASE_TIME)
1661 for volume ramp downs in the same fragment (needed for current voice
1662 stealing implementation) then simply inform the user with a warning
1663 about possible click sounds and reduce the volume ramp down
1664 appropriately instead of cancelling the audio device connection
1666 Version 0.3.2 (24 June 2005)
1668 * packaging changes:
1669 - updated autotools build files to compile on OS X
1670 (conditional compilation of CoreMIDI and MidiShare drivers)
1671 - hand-crafted assembly optimization code can be disabled with
1672 './configure --disable-asm' (definitely not recommended)
1673 - fixed 'make dist' rule to include all necessary files
1674 - require automake (>= 1.5) for 'make -f Makefile.cvs'
1675 (due to 'dist-bzip2' automake option)
1677 * Gigasampler format engine:
1678 - support for the gig parameters for "release velocity response" curves
1679 - fine tuning of the EGADSR envelope
1680 - volume of release triggered samples now depends on note-on velocity,
1681 note length and gig parameter "release trigger decay" instead of
1682 note-off velocity.
1683 - revised voice stealing
1684 (fixes crash and endless loop caused by voice stealing)
1685 - don't reset scale tuning on instrument or audio output device change
1686 - handle key group conflicts right at the beginning of each fragment
1687 instead of when the respective voice is actually launched
1688 (fixes undefined behavior if stolen voices belonged to a key group -
1689 this case was followed by a "killed voice survived" error message)
1690 - fixed minor issue with null/silence samples
1691 (those stole voices even though they don't need a voice at all which
1692 resulted in "voice stealing didn't work out" messages)
1693 - don't reset volume, pan, pitch and MIDI controller values on
1694 instrument or audio output device change
1696 * LSCP server:
1697 - fixed some crashes (patch by Grigor Iliev, fixes #19)
1698 - fixed LSCP event "CHANNEL_INFO" notification
1699 (e.g. did not notify on volume changes or MIDI program change events)
1701 * linuxsampler application:
1702 - added command line parameters --lscp-addr and --lscp-port to override
1703 default IP address and port of LSCP server
1705 Version 0.3.1 (24 May 2005)
1707 * initial release

  ViewVC Help
Powered by ViewVC