ViewVC logotype

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 3967 - (show annotations) (download)
Sun Jun 20 11:09:30 2021 UTC (2 days, 7 hours ago) by schoenebeck
File size: 103109 byte(s)
* gig engine: Use libgig's gig::File::GetInstrument() method with method
  gig::File::CountInstruments() instead of now deprecated
  gig::File::GetFirstInstrument() / gig::File::GetNextInstrument() pair.

* Bumped version (2.2.0.svn4).

1 Version SVN trunk (?)
3 * GigaStudio/Gigasampler format engine:
4 - Use libgig's new IO-per-thread feature to avoid file I/O concurrency
5 issues with .gig file.
6 - Use libgig's new method gig::File::GetSample() with method
7 gig::File::CountSamples() instead of now deprecated
8 gig::File::GetFirstSample() / gig::File::GetNextSample() pair.
9 - Use libgig's new method gig::Instrument::GetRegionAt() with method
10 gig::Instrument:CountRegions() instead of now deprecated
11 gig::Instrument::GetFirstRegion() / gig::Instrument::GetNextRegion() pair.
12 - Use libgig's gig::File::GetInstrument() method with method
13 gig::File::CountInstruments() instead of now deprecated
14 gig::File::GetFirstInstrument() / gig::File::GetNextInstrument() pair.
16 Version 2.2.0 (9 May 2021)
18 * general changes:
19 - Fixed compiler error in Pool.h.
20 - Require C++14 compiler support.
21 - Autoconf: Added m4/ax_cxx_compile_stdcxx.m4 macro which is used
22 for checking in configure for C++14 support (as mandatory
23 requirement) and automatically adds compiler argument if required
24 (e.g. -std=C++14).
25 - RTMath: Implemented floating point comparison methods
26 fEqual32(float,float) and fEqual64(double,double) which take the
27 expected floating point tolerances into account.
28 - Added public API C++ class "LFO", which is a cluster class encapsulating
29 all the sampler's LFO implementations to be used by 3rd party applications
30 (e.g. by Gigedit).
31 - Added int math square LFO implementation.
32 - Added int math saw LFO implementation.
33 - Added numeric complex nr sine LFO implementation.
34 - Marked class LFOTriangleDiHarmonic as deprecated
35 (will be removed in future).
36 - Added LFOAll.h which includes all LFO implementation's header files.
37 - Refactored LFO class names and their file names (PulseLFO -> LFOPulse,
38 LFOSawIntMath -> LFOSawIntMathNew, SawLFO -> LFOSawIntMathOld,
39 SineLFO -> LFOSineBuiltinFn, LFOSine -> LFOSineNumericComplexNr,
40 SquareLFO -> LFOSquarePulse and separated the latter to its own header
41 file).
42 - Renamed type LFOSigned -> LFOTriangleSigned.
43 - Renamed type LFOUnsigned -> LFOTriangleUnsigned.
44 - Using now cubic interpolation as resampling algorithm by default; and
45 linear interpolation is deprecated now.
46 - Event:Init(): memset() entire Event::Param union to zero instead of trying
47 to do that for its individual members; the latter became error prone due
48 to the union's growing complexity meanwhile.
49 - EngineChannel: Overhaul of RPN/NRPN related methods: Added new
50 *Rpn*Parameter*() methods as replacement for previous *Rpn*Controller*()
51 methods and marked the old ones as deprecated; and likewise added new
52 *Nrpn*Parameter*() methods as replacement for previous *Nrpn*Controller*()
53 methods and marked those old ones deprecated as well.
54 - Added support for MIDI CC #38 (data entry LSB): this allows higher value
55 resolution for certain (e.g. hard coded) RPN and NRPN parameters.
56 - Implemented support for compressed RPN/NRPN message sequences (that is to
57 allow processing MIDI data entry messages without always having to be
58 preceded strictly by RPN/NRPN parameter selection messages on each value
59 change).
60 - Added support for MIDI CC #96 (data increment) and MIDI CC #97 (data
61 decrement).
62 - Added support for 'null function RPN/NRPN' (which is RPN/NRPN parameter
63 MSB=127 and LSB=127) and which is disabling subsequent data
64 entry/increment/decrement CC messages.
65 - Fixed crash when editing key groups (a.k.a. exclusive groups) with an
66 instrument editor.
67 - Fixed playing noise at the end of a disk stream under certain conditions.
68 - Fixed potential crash with high pitch at end of RAM playback.
69 - Added configure checks whether C++ compiler supports so called
70 'designated initializers'.
71 - Added global backtraceAsString() function for debugging purposes
72 (POSIX systems only, no Windows implementation yet).
73 - Thread class: Added new methods pushCancelable(), popCancelable(), name(),
74 nameOfCaller() and setNameOfCaller().
75 - configure: Always check for availability of pthread_testcancel() and use
76 it if so (previously it was only used if explicitly enabled by configure
77 option).
78 - Thread class: raise compiler warning if pthread_testcancel() is not
79 available, because it can lead to deadlocks.
80 - Mutex class: Implemented optional bug detection and deadlock debugging
81 features.
82 - Ref<> class is now thread safe (as lock-free and wait-free
83 implementation).
84 - Fixed compilation error with Bison >= 3.6.0.
85 - Fixed rare crash when switching between instruments that had instrument
86 scripts loaded.
87 - Allow EffectChain::InsertEffect() as alternative to
88 EffectChain::AppendEffect() (i.e. with index equal to current chain size).
89 - ConstCapacityArray class: Fixed bug in remove() method.
90 - Fixed crash on certain MIDI RPN and NRPN messages that occurred if no
91 real-time instrument script was loaded.
93 * Real-time instrument scripts:
94 - Added method ScriptVM::setExitResultEnabled() which allows to
95 explicitly enable the built-in exit() function to optionally accept
96 one function argument; the value of the passed exit() function
97 argument will then become available by calling
98 VMExecContext::exitResult() after script execution.
99 - 64 bit support for NKSP integer scripts variables (declare $foo).
100 - Variable names, function names and preprocessor condition names must start
101 with a regular character (a-z or A-Z); starting them with a digit or
102 underscore is not allowed.
103 - NKSP parser fix: equal comparison operator "=" and not equal comparison
104 operator "#" must only accept integer operands.
105 - NKSP language: Implemented support for standard units like Hertz, seconds,
106 Bel including support for metric unit prefixes; so one can now e.g.
107 conveniently use numbers in scripts like "5us" meaning "5 microseconds",
108 or e.g. "12kHz" meaning "12 kilo Hertz", or e.g. "-14mdB" meaning
109 "minus 14 Millidecibel", or e.g. "28c" meaning "28 cents" (for tuning).
110 - NKSP language: Introduced "final" operator "!" which is specifically
111 intended for synthesis parameter values to denote that the synthesis
112 parameter value is intended to be the "final" value for that synthesis
113 parameter that should explicitly be used by the engine and thus causing
114 the sampler engine to ignore all other modulation sources for the same
115 synthesis parameter (like e.g. LFO, EG); by simply prefixing a value,
116 variable or formula with this new "!" operator the expression is marked as
117 being "final".
118 - NKSP script editor API: Added support for detecting standard unit tokens
119 and their potential metric prefix token.
120 - NKSP language: Added support for NKSP real number literals and
121 arithmetic operations on them (e.g. "(3.9 + 2.9) / 12.3 - 42.0").
122 - NKSP language: Added support for NKSP real number (floating point) script
123 variables (declare ~foo).
124 - NKSP language: Added support for NKSP real number (floating point) array
125 script variables (declare ?foo[]).
126 - Built-in script function "message()" accepts now real number argument as
127 well.
128 - Added built-in script function "real_to_int()" and its short hand form
129 "int()" for casting from real number to integer in NKSP scripts.
130 - Added built-in script function "int_to_real()" and its short hand form
131 "real()" for casting from integer to real number in NKSP scripts.
132 - Allow built-in exit() function to potentially accept real number type
133 argument as well.
134 - Built-in script functions may have a different return type depending on
135 the arguments passed to the function.
136 - Built-in script function "abs()" optionally accepts and returns real
137 number.
138 - Built-in script functions "min()" and "max()" optionally accept real
139 number arguments and return real number as result in that case.
140 - NKSP VM API: Allow units and 'final'ness to be returned as result from
141 built-in functions (added methods VMFunction::returnUnitType() and
142 VMFunction::returnsFinal() for that purpose which must be implemented by
143 built-in function implementations).
144 - NKSP language: Allow metric unit prefixes of numeric scalar and array
145 variables to be changed freely at runtime (unlike unit types like Hz etc.
146 which are still sticky, parse-time features of variables which cannot be
147 changed at runtime for the intentional sake of determinism).
148 - NKSP language: 'final' values are prohibited for array variables for now
149 (attempt causes a parsers error).
150 - NKSP language: expressions with unit types (e.g. Hz) are prohibited for
151 conditions of runtime control structures like if(), while(), select()
152 (attempt causes a parser error).
153 - NKSP VM API: Allow built-in functions to perform their own, individual
154 parse time checks of arguments going to be passed to the function at
155 runtime (added method VMFunction::checkArgs() for that purpose).
156 - NKSP language: raise parser warning if only one operand of binary
157 operators (like logical 'or' comparison) contain a 'final' value (because
158 it would always yield in a 'final' result in such cases).
159 - NKSP language: Allow comparison (=, #, <, >, <=, >=) of values with
160 different metric unit prefixes, which will behave as expected (e.g.
161 result of expression '1000us < 2ms' is true).
162 - NKSP language: Allow adding values with different metric unit prefixes
163 (e.g. result of expression '100Hz + 5kHz' is '5100Hz').
164 - NKSP language: Allow subtracting values with different metric unit
165 prefixes (e.g. result of expression '1ms - 20us' is '980us').
166 - NKSP language: Allow multiplying with any metric unit prefixes
167 (e.g. result of expression '2k * 3ms' is '6s'), however multiplications
168 with unit types on both sides (e.g. '2s * 2s') is still prohibited since
169 we don't have any considerable practical use for a term like '4s^2'
170 (hence any attempt multiplying two unit types still causes parser error).
171 - NKSP language: Allow dividing by any metric unit prefixes and allow
172 division of same unit type on both sides (e.g. expression '8kHz / 1000Hz'
173 yields in unit free result '8'). So this is now a way to cast units away
174 e.g. for passing the result to other expressions, certain function calls
175 or variables which are not accepting any units (or that specific unit).
176 - NKSP language: integer arrays and real number arrays can now be converted
177 to strings (e.g. for dumping their content with message() calls for
178 script debugging purposes).
179 - NKSP language: expressions and variables with units are now correctly
180 casted to strings (e.g. with message() calls).
181 - NKSP language: comparing real numbers for equalness (e.g. '~foo = 3.1') or
182 unequalness (e.g. '~foo # 3.1') is now less strict and takes the expected
183 floating point tolerances into account.
184 - NKSP VM API: Added methods VMScalarNumberExpr::evalCastInt() and
185 VMScalarNumberExpr::evalCastReal().
186 - NKSP VM API: Added base class 'VMNumberArrayExpr' for classes
187 'VMIntArrayExpr' and 'VMRealArrayExpr'.
188 - NKSP VM API: replaced all unitPrefix() (parse time) methods by
189 unitFactor() (runtime) methods.
190 - Built-in function "exit()" supports now returning units and 'final'ness
191 for test cases.
192 - The following built-in functions support now units as well: "abs()",
193 "random()", "inc()", "dec()", "in_range()", "min()", "max()",
194 "real_to_int()", "int()", "int_to_real()" and "real()".
195 - Built-in functions "array_equal()", "search()" and "sort()" support now
196 real number arrays (correctly) as well.
197 - Added individual parse time checks of arguments to be passed to built-in
198 functions "random()", "inc()", "dec()", "in_range()", "min()", "max()",
199 "array_equal()" and "search()" specific for their individual purposes.
200 - NKSP VM refactoring: Renamed all methods, functions and classes matching
201 pattern *ScalarNumber* to simply *Number* (that is i.e. classes
202 VMScalarNumberExpr -> VMNumberExpr, ScalarNumberExpr -> NumberExpr,
203 ScalarNumberVariable -> NumberVariable, ScalarNumberBinaryOp ->
204 NumberBinaryOp, VMScalarNumberResultFunction -> VMNumberResultFunction,
205 method VMExpr::asScalarNumberExpr() -> VMExpr::asNumber(), function
206 isScalarNumber() -> isNumber()).
207 - NKSP VM API: Added 4 overridden methods to class VMNumberExpr:
208 evalCastInt(MetricPrefix_t), evalCastInt(MetricPrefix_t,MetricPrefix_t),
209 evalCastReal(MetricPrefix_t), evalCastReal(MetricPrefix_t,MetricPrefix_t)
210 as convenient methods for automatically converting values to expected
211 metric value basis.
212 - Built-in function "wait()" accepts now both integers and real numbers as
213 argument.
214 - NKSP VM API cleanup: Get rid of legacy method
215 VMFunction::argType(vmint iArg) which was already superseded by its new
216 replacement VMFunction::acceptsArgType(vmint iArg, ExprType_t type).
217 - NKSP parser: if wrong argument type was passed to a built-in function and
218 that built-in function accepts more than one data type for the argument,
219 then show all supported data types as parser error message.
220 - Built-in function "play_note()" accepts now real numbers and seconds as
221 unit type as well for its 3rd and 4th function arguments.
222 - The following built-in functions accept now real numbers as well for their
223 2nd function argument: "change_vol()", "change_tune()", "change_cutoff()",
224 "change_attack()", "change_decay()", "change_release()",
225 "change_sustain()", "change_cutoff_attack()", "change_cutoff_decay()",
226 "change_cutoff_sustain()", "change_cutoff_release()",
227 "change_amp_lfo_freq()", "change_cutoff_lfo_freq()",
228 "change_pitch_lfo_freq()", "change_vol_time()", "change_tune_time()",
229 "change_pan_time()", "fade_in()", "fade_out()", "change_play_pos()".
230 - Fixed built-in function "change_play_pos()" not having accepted metric
231 prefixes at all.
232 - Fixed the following built-in functions having misinterpreted values given
233 with unit type (for their 2nd argument) as if they were relative values
234 (that is as if they were passed without a unit type): "change_attack()",
235 "change_decay()", "change_release()", "change_cutoff_attack()",
236 "change_cutoff_decay()", "change_cutoff_release()".
237 - Fixed the following built-in functions having applied completely wrong
238 'final' values: "change_sustain()", "change_cutoff_sustain()" (since the
239 respective EGs being their modulation sink assume uint data type with
240 value range 0..1000 instead of 0.0..1.0.
241 - Added individual parse-time checks of function arguments for the following
242 built-in functions: "play_note()", "note_off()", "set_event_mark()",
243 "delete_event_mark()", "by_marks()", "change_cutoff()", "change_attack()",
244 "change_decay()", "change_release()", "change_cutoff_attack()",
245 "change_cutoff_decay()", "change_cutoff_release()",
246 "change_amp_lfo_freq()", "change_cutoff_lfo_freq()",
247 "change_pitch_lfo_freq()", "change_vol_time()", "change_tune_time()" and
248 "change_pan_time()".
249 - Don't abort function call if unit type was used and at the same time
250 'final' operator was omitted for the primary value argument of the
251 following built-in functions: "change_cutoff()", "change_attack()",
252 "change_decay()", "change_release()", "change_cutoff_attack()",
253 "change_cutoff_decay()", "change_cutoff_release()",
254 "change_amp_lfo_freq()", "change_cutoff_lfo_freq()",
255 "change_pitch_lfo_freq()", "change_vol_time()", "change_tune_time()",
256 "change_pan_time()", instead imply 'final'ness at runtime and raise an
257 appropriate parser warning at parse time.
258 - Added built-in real number functions "round()", "ceil()", "floor()",
259 "sqrt()", "log()", "log2()", "log10()", "exp()", "pow()", "sin()",
260 "cos()", "tan()", "asin()", "acos()", "atan()".
261 - Added built-in script real number constant "~NI_MATH_PI".
262 - Added built-in script real number constant "~NI_MATH_E".
263 - NKSP language: Allow unary '+' operator.
264 - Added built-in script functions "msb()" and "lsb()".
265 - NKSP language: Added support for RPN event handler
266 ("on rpn ... end on" in instrument scripts).
267 - NKSP language: Added support for NRPN event handler
268 ("on nrpn ... end on" in instrument scripts).
269 - Added built-in read-only variables "$RPN_ADDRESS" and "$RPN_VALUE" which
270 may be read from the new RPN/NRPN script handlers to get the (N)RPN
271 parameter that had been changed and its new value.
272 - Added built-in const variables "$NI_CB_TYPE_RPN" and "$NI_CB_TYPE_NRPN"
273 which are identifying the new (N)RPN handlers as such at script runtime.
274 - Implemented built-in script function "set_rpn()".
275 - Implemented built-in script function "set_nrpn()".
276 - Fix: built-in functions returning an array type were broken and caused a
277 crash when the built-in function's (supposed to be array) result value was
278 used by the script VM.
279 - Removed upper value constraint for 2nd argument of built-in functions
280 "change_amp_lfo_freq()", "change_cutoff_lfo_freq()",
281 "change_pitch_lfo_freq()","change_reso()", "change_cutoff()",
282 "change_amp_lfo_depth()", "change_cutoff_lfo_depth()" and
283 "change_pitch_lfo_depth()", to allow e.g. passing value 2000000 to double
284 the LFO frequency / resonance / cutoff frequency.
285 - Reduced code involved for parsing NKSP variable declarations by using 2
286 generalized grammar rules with optional components (i.e. optional value
287 assignment and optional variable qualifiers) instead of previous linear
288 combinations of grammar rules, which also fixes some edge case.
289 - NKSP language parser: track code block locations also by a) raw byte
290 position in NKSP source code string and b) length of that code block in
291 bytes (along to the already existing code block tracking by line/column);
292 and added appropriate new members to public C++ API classes like
293 'firstByte' and 'lengthBytes' to class VMSourceToken and struct CodeBlock.
294 - NKSP language: Added support for 'patch' variable qualifier (capture their
295 locations and allow replacement of their assignment expression via VM C++
296 API).
297 - Fixed incorrect EG times if standard measuring unit was passed to built-in
298 functions "change_attack()", "change_decay()" or "change_release()".
299 - Fixed intermediate function result values never having reflected any
300 standard measuring unit type.
301 - Fixed re-entrant issue with function calls which caused wrong result
302 values if the same function was called multiple times in a term
303 (specifically if metric prefixes were used).
304 - NKSP language: Just throw a warning, not an error if an array variable of
305 size zero was declared.
306 - NKSP language: Allow omitting explicit array size on array variable
307 declaration if combined with immediate value assignment
308 (e.g. declare %foo[] := ( 1, 2, 3 ) ).
309 - NKSP language: emit warning if an array variable was declared with bigger
310 array size than amount of initial values been assigned, and initialize the
311 missing array elements with zero in this case.
312 - Fixed built-in NKSP exit() function to behave as return statement.
313 - NKSP parser: Fixed crash if unary '-' operator was used on a non-number
314 data type.
315 - NKSP VM: Fixed the "release" handler being executed too often under
316 certain situations if polyphonic data was passed from "note" handler to
317 "release" handler.
318 - Fixed memory leak in NKSP parser: string tokens were allocated as C
319 strings and never freed.
320 - Fixed incorrect polyphonic data transfer from wrong note handler to
321 release handler; which also fixes no release handler being executed
322 sometimes, and due to the latter it also fixes potential crashes as some
323 polyphonic script events were never released and the engine hence was
324 running out of free script events.
325 - Fixed crash when handling array results of built-in functions
326 (e.g. "num_elements(by_marks($MARK_1)").
327 - NKSP VM: Fixed dead note events not been removed from user's event groups
328 (i.e. causing incorrect behaviour of built-in NKSP functions by_marks()
329 and set_event_mark() after max. size of event group list was exhausted).
331 * test cases:
332 - Fixed compiler errors in test cases.
333 - Updated README for how to compile & run test cases.
334 - Updated test case
335 MutexTest::testDoubleLockStillBlocksConcurrentThread() to latest
336 expected behaviour of the Mutex class implementation (recursive
337 mutex type).
338 - Added test cases for NKSP core language aspects and core built-in
339 functions.
340 - Fixed thread tests segfaulting on Linux.
341 - NKSP: Added real number test cases for built-in functions exit(),
342 int_to_real(), real(), real_to_int() and int(), as well as for the
343 plus, minus and negate language operators.
344 - Added massive amount of NKSP test cases for standard measuring units and
345 'final' operator usage cases.
346 - Added NKSP test cases for (floating point tolerance aware) real number
347 equalness / unequalness comparison.
348 - Added NKSP int array and real array tests for value assignment and
349 initialization of arrays.
350 - Added NKSP test cases for built-in functions "round()", "ceil()",
351 "floor()", "sqrt()", "log()", "log2()", "log10()", "exp()", "pow()",
352 "sin()", "cos()", "tan()", "asin()", "acos()", "atan()".
353 - Added NKSP test cases for unary '+' operator.
354 - Added NKSP test cases for core built-in variables.
355 - Added thorough NKSP test cases for variable declarations.
357 * GigaStudio/Gigasampler format engine:
358 - LFOTriangleIntMath and LFOTriangleIntAbsMath: Fixed FlipPhase=true
359 behaviour for start_level_mid.
360 - Changed LFO start levels: LFO1 and LFO2 both to mid, LFO3 to max. start
361 level (see discussion "GigaStudio LFO compatibility" on mailing list
362 from 2019-09-26 for details).
363 - Changed default wave form for all 3 LFOs to sine (instead of triangle;
364 see discussion "GigaStudio LFO compatibility" on mailing list from
365 2019-09-26 for details).
366 - Format extension: Added support for different LFO wave forms (currently
367 either sine [default], triangle, saw or square).
368 - Format extension: Added support for LFO phase displacement (0��..360��).
369 - Format extension: Added support for flipping LFO polarity on LFO 3
370 (in the original gig format this was only available for LFO 1 and LFO 2).
371 - Format extension: Support for additional filter types:
372 lowpass 1/2/4/6-pole, highpass 1/2/4/6-pole, bandpass 2-pole,
373 bandreject 2-pole.
374 - Fixd potential crash with polynomial interpolator and max. pitch.
375 - If minimum cutoff is defined, remap cutoff controller CC values to really
376 span the entire possible cutoff range (and avoiding a dead controller
377 zone).
378 - Implemented support for loading instrument scripts with 'patch' variables
379 (by patching these script variables with potentially overridden ones by
380 gig Instrument).
382 * SFZ format engine:
383 - Fixed support for regions with loccN/hiccN conditions on more than one
384 MIDI controller.
385 - Fixed internal opcode 'xfin_hivel' falsely being set by user opcode
386 'xfin_lovel' (patch by Christian Czezatke).
388 * Benchmarks:
389 - Fixed benchmarks/triang.cpp falsely having favoured "int math abs"
390 algorithm (since result of 2nd run was not accumulated).
391 - Added benchmark for saw wave (benchmarks/saw.cpp).
392 - Added benchmark for sine wave (benchmarks/sine.cpp).
393 - Added benchmark for square wave (benchmarks/square.cpp).
394 - Increased amount of benchmarks runs by factor 6 to achieve benchmark times
395 which are large enough on modern systems.
397 Version 2.1.1 (27 Jul 2019)
399 * Real-time instrument scripts:
400 - Fixed behavior of built-in NKSP functions change_sustain(),
401 change_cutoff_attack(), change_cutoff_decay(), change_cutoff_sustain()
402 and change_cutoff_release().
404 * general changes:
405 - Only play release trigger samples on sustain pedal up if this behaviour
406 was explicitly requested by the instrument (otherwise only on note-off).
407 - Fixed compiler warnings.
408 - Fixed compilation error when cross-compiling to Mac.
409 - FX Sends: Provide more useful error messages on routing problems
410 (see bug #169).
411 - LSCP doc: Be more clear describing the two distinct approaches
412 of using external vs. internal effects (see bug #169).
413 - "optional" class: Fixed comparison operators.
415 * Gigasampler/GigaStudio format engine:
416 - Format extension: If requested by instrument then don't play release
417 trigger sample on note-off events.
419 * SFZ format engine:
420 - Fixed memory leak when releasing samples
421 (fixes bug #307, patch by Jacek Roszkowski)
422 - Fixed potential crash when a sample is shared by more than one region
423 (fixes bug #308, patch by Jacek Roszkowski).
424 - Opcode 'sample': Added support for built-in sample '*silence'
425 (fixes bug #310, patch by Jacek Roszkowski).
427 Version 2.1.0 (25 Nov 2017)
429 * SFZ format engine:
430 - added support for <global>, <master> and #define (patch by Alby M)
431 - Removed code duplication in SFZ file loading code.
432 - Added support for sfz extension opcode 'script' which may be used to
433 load real-time instrument script file (NKSP script language).
434 - Implemented opcode set_ccN (initial patch by Giovanni Senatore).
435 - Fixed unintended volume fade-in of voices under certain conditions.
436 - sfz parser: allow missing space between header and opcode
438 * Gigasampler/GigaStudio format engine:
439 - Fixed clicks and pumping noise with Lowpass Turbo filter on very low
440 cutoff settings.
441 - Got rid of resembling an ancient GSt misbehavior which did not pitch at
442 all if an up-pitch of more than 40 semi tones was requested (I don't
443 think there is any stock gig sound that requires this behavior to
444 resemble its original sound).
445 - Added support for controlling whether the individual EGADSR stages may
446 be aborted (as LinuxSampler extension to the original GigaStudio 4
447 format).
449 * general changes:
450 - fixed printf type errors (mostly in debug messages)
451 - use unique_ptr instead of auto_ptr when building with C++11
452 - Added RTAVLTree class which is a real-time safe ordered multi-map, thus
453 allowing to sort data efficiently in real-time safe manner.
454 - RTList class: added methods for moving/inserting elements to arbitrary
455 position within a list.
456 - RTList class: added method fromPtr() for creating an Iterator object from
457 a raw element pointer.
458 - Attempt to partly fix resetting engine channels vs. resetting engine, an
459 overall cleanup of the Reset*(), ConnectAudioDevice(),
460 DisconnectAudioDevice() API methods would still be desirable though,
461 because the current situation is still inconsistent and error prone.
462 - Active voices are now internally grouped to "Note" objects, instead of
463 being directly assigned to a keyboard key. This allows more fine graded
464 processing of voices, which is i.e. required for certain instrument
465 script features.
466 - Fix: Release trigger voices were not spawned on sustain pedal up
467 (CC #64) events.
468 - Fix: Release trigger voices can now also distinguish correctly between
469 note-off and sustain pedal up events.
470 - All engines: Increased ramp speed of volume smoother and pan smoother
471 (while slow rate of crossfade smoother is preserved) to allow quick
472 volume and pan changes by instrument scripts for instance.
473 - gig/sf2/sfz: Fixed aftertouch channel pressure events being processed
474 properly and thus were ignored for certain uses.
475 - Added new C++ API method
476 VirtualMidiDevice::SendChannelPressureToSampler().
477 - windows, 32-bit: fixed potential crashes by making sure the stack in
478 sub threads is 16-byte aligned
479 - fixed numerous compiler warnings
480 - Fixed invalid (note-on) event ID being assigned to new Note objects.
481 - Revised fundamental C++ classes "Thread", "Mutex" and "Condition" which
482 fixes potential undefined behavior.
483 - Fixed Note object leak when triggering notes on keys which did not
484 have a valid sample mapped (fixes bug #252).
485 - Fixed compilation errors when compiling with CONFIG_DEVMODE enabled.
486 - linuxsampler binary fix: option --create-instruments-db ignored
487 subsequent optional argument due to glibc's implementation oddity
488 which expects a "=" sign, but no space between them.
490 * packaging changes:
491 - removed unnecessary dependency to libuuid
492 (originated by libgig's usage of it)
493 - Automake: set environment variable GCC_COLORS=auto to allow GCC to
494 auto detect whether it (sh/c)ould output its messages in color.
495 - Debian: Fixed packaging error about invalid "Source-Version"
496 substitution variable.
497 - Debian: Raised Debian compatibility level to Debian 9 "Stretch".
498 - Debian: Added build dependency to libsqlite3-dev for building
499 linuxsampler with instruments DB support.
501 * Real-time instrument scripts:
502 - Implemented scheduler for delayed MIDI events and for suspended scripts.
503 - Built-in script function "wait()": implemented support for function's
504 "duration-us" argument, thus scripts using this function are now
505 correctly resumed after the requested amount of microseconds.
506 - Built-in script function "play_note()": implemented support for
507 function's "duration-us" argument, thus notes triggered with this
508 argument are now correctly released after the requested amount of
509 microseconds.
510 - Fix: script events were not cleared when engine channel was reset,
511 potentially causing undefined behavior.
512 - Fixed crash which happened when trying to reference an undeclared
513 variable.
514 - Built-in script function "play_note()": Added support for passing
515 special value -1 for "duration-us" argument, which will cause the
516 triggered note to be released once the original note was released.
517 - Instrument script classes now exported with the liblinuxsampler C++ API.
518 - Added new API method ScriptVM::syntaxHighlighting() which provides
519 a convenient syntax highlighting backend for external instrument
520 script editor applications.
521 - Added new C++ API class "ScriptVMFactory".
522 - Extended parser issues to provide not only first line and first
523 column, but also last line and last column of issue (thus marking
524 the precise span of the issue within the source code).
525 - Reload script automatically after being modified by an instrument
526 editor.
527 - NKSP language grammar correction: allow empty event handler bodies
528 like "on note end on".
529 - Implemented built-in script function "change_vol()".
530 - Implemented built-in script function "change_tune()".
531 - Implemented built-in script function "change_pan()".
532 - Implemented built-in script function "change_cutoff()".
533 - Implemented built-in script function "change_reso()".
534 - Implemented built-in script function "event_status()".
535 - Added built-in script constants "$EVENT_STATUS_INACTIVE" and
536 "$EVENT_STATUS_NOTE_QUEUE" both for being used as flags for
537 "event_status()" function.
538 - NKSP language: Added support for bitwise operators ".or.", ".and."
539 and ".not.".
540 - NKSP language scanner: Fixed IDs matching to require at least one
541 character (i.e. when matching function names or variable names).
542 - NKSP language scanner: disabled unusued rules.
543 - Fixed behavior of play_note() and note_off() functions which must
544 be distinguished engine internally from "real" MIDI note on/off
545 events in order to avoid misbehaviors like hanging notes.
546 - Implemented built-in script variable "$KSP_TIMER".
547 - Implemented built-in script variable "$NKSP_REAL_TIMER".
548 - Implemented built-in script variable "$NKSP_PERF_TIMER".
549 - Implemented built-in script variable "$ENGINE_UPTIME".
550 - Implemented built-in script function "inc()".
551 - Implemented built-in script function "dec()".
552 - NKSP language fix: division expressions were evaluated too often.
553 - NKSP language fix: string concatenation operator was right
554 associative instead of left (to right).
555 - NKSP language correction: allow empty statements in entire language
556 (i.e. "if end if").
557 - Implemented built-in script function "stop_wait()".
558 - Implemented built-in script variable "$NI_CALLBACK_ID".
559 - Implemented built-in script variable "$NI_CALLBACK_TYPE".
560 - Implemented built-in script variable "$NKSP_IGNORE_WAIT".
561 - Added support for read-only built-in variables (respectively
562 handled by the script parser).
563 - Added built-in script constant "$NI_CB_TYPE_INIT".
564 - Added built-in script constant "$NI_CB_TYPE_NOTE".
565 - Added built-in script constant "$NI_CB_TYPE_RELEASE".
566 - Added built-in script constant "$NI_CB_TYPE_CONTROLLER".
567 - NKSP Language: Added support for user defined script functions.
568 - Implemented built-in script function "change_attack()".
569 - Implemented built-in script function "change_decay()".
570 - Implemented built-in script function "change_release()".
571 - Fixed all change_*() built-in script functions to apply their
572 synthesis parameter changes immediately in case the respective note
573 was triggered at the same time, instead of scheduling the parameter
574 change, especially because it would cause some parameter types's
575 changes either to be ramped (i.e. change_vol()) or other types even
576 to have not effect at all (i.e. change_attack()).
577 - Implemented built-in script function "sh_left()".
578 - Implemented built-in script function "sh_right()".
579 - Implemented built-in script function "min()".
580 - Implemented built-in script function "max()".
581 - NKSP Fix: "init" event handler was not always executed when a script
582 was loaded.
583 - NKSP built-in wait() script function: abort script execution if
584 a negative or zero wait time was passed as argument (since this is
585 a common indication of a bug either of the script or even of the
586 engine, which could lead to RT instability or even worse).
587 - ScriptVM: Implemented automatic suspension of RT safety
588 threatening scripts.
589 - Provide more user friendly error messages on syntax errors.
590 - Fixed NKSP parser warning "Not a statement" when assigning an
591 initializer list to an array variable.
592 - Implemented built-in script array variable "%ALL_EVENTS".
593 - Implemented built-in script function "in_range()".
594 - Implemented built-in script function "change_amp_lfo_depth()".
595 - Implemented built-in script function "change_amp_lfo_freq()".
596 - Implemented built-in script function "change_pitch_lfo_depth()".
597 - Implemented built-in script function "change_pitch_lfo_freq()".
598 - Implemented built-in script function "change_vol_time()".
599 - Implemented built-in script function "change_tune_time()".
600 - Implemented built-in script function "fade_in()".
601 - Implemented built-in script function "fade_out()".
602 - Fixed acceptance of wrong data type of parameters passed to built-in
603 script functions "change_vol()", "change_tune()", "change_pan()",
604 "change_cutoff()", "change_reso()", "change_attack()", "change_decay()",
605 "change_release()", "change_amp_lfo_depth()", "change_amp_lfo_freq()",
606 "change_pitch_lfo_depth()" and "change_pitch_lfo_freq()".
607 - Added built-in script function "get_event_par()" and implemented some
608 of its possible parameter selections.
609 - Added built-in script function "set_event_par()" and implemented some
610 of its possible parameter selections.
611 - Fixed a bunch of scheduler time related bugs.
612 - Fixed polyphonic variables not being reset to zero after usage.
613 - Built-in "ignore_event()" function: argument is now optional, like with
614 built-in function "ignore_controller()".
615 - Implemented built-in script function "change_velo()".
616 - Implemented built-in script function "change_note()".
617 - Adjusted behavior of "change_vol()" and "change_tune()" to a more
618 intuitive behavior if used in combination with "change_vol_time()" or
619 "change_tune_time()" respectively: now tuning/volume changes are only
620 assigned (without delay) immediately to a new note if the respective
621 timing function has not been called before, otherwise the volume/tuning
622 changes are automatically faded (before, only the event's time stamp was
623 relevant).
624 - Implemented built-in script function "array_equal()".
625 - Implemented built-in script function "search()".
626 - Implemented built-in script function "sort()".
627 - NKSP Fix: Never suspend "init" event handlers.
628 - Implemented built-in script function "same_region()" (currently only
629 available for gig format engine).
630 - Added built-in script constant "$NKSP_LINEAR".
631 - Added built-in script constant "$NKSP_EASE_IN_EASE_OUT".
632 - Implemented built-in script function "change_vol_curve()".
633 - Implemented built-in script function "change_tune_curve()".
634 - built-in "play_note()" function now supports a sample playback start
635 offset with argument 3, where special value -1 means to use the regular
636 sample offset as defined by the instrument file.
637 - Built-in array variable %KEY_DOWN[] is now a read-only variable.
638 - Built-in variable $EVENT_NOTE is now a read-only variable.
639 - Built-in variable $EVENT_VELOCITY is now a read-only variable.
640 - built-in "play_note()" function now accepts -2 for its fourth argument
641 (note duration) which means the life time of the note shall be sticked
642 to the requested note number of argument 1.
643 - Fix: built-in "play_note()" function now returns 0 as result value if
644 -1 was passed for its fourth argument (note duration) and the respective
645 parent note is already gone.
646 - Implemented built-in script function "change_play_pos()".
647 - NKSP language: Added support for user declared const array variables.
648 - NKSP language: Raise parser warning if array variable is accessed with
649 an index that exceeds the array's size.
650 - NKSP language fix: Unknown characters were not handled correctly.
651 - NKSP language: Added support for "synchronized .. end synchronized"
652 code blocks.
653 - Implemented built-in script function "abort()" which allows to abort
654 another script handler by passing its callback ID.
655 - Fixed potential memory access bug and potential undefined behavior of
656 "init" event handlers.
657 - Print a time stamp along to each call of built-in function "message()".
658 - ScriptVM API: Added VMParserContext::preprocessorComments() which allows
659 to retrieve all code blocks filtered out by the preprocessor.
660 - Added built-in script function "fork()".
661 - Added built-in array variable %NKSP_CALLBACK_CHILD_ID[].
662 - Added built-in variable $NKSP_CALLBACK_PARENT_ID.
663 - Fixed potential crash when accessing dynamic built-in array variables.
664 - Added built-in script function "callback_status()".
665 - Added built-in constant $CALLBACK_STATUS_TERMINATED.
666 - Added built-in constant $CALLBACK_STATUS_QUEUE.
667 - Added built-in constant $CALLBACK_STATUS_RUNNING.
668 - Removed max. value limitation of built-in functions "change_attack()",
669 "change_decay()" and "change_release()" to i.e. allow passing 2000000
670 for doubling the respective time.
671 - NKSP script editor syntax highlighting API: Fixed app termination due
672 to a lexer start condition stack underrun.
673 - NKSP preprocessor: Fixed wrong behavior on nested USE_CODE_IF() and
674 USE_CODE_IF_NOT() preprocessor statements.
675 - NKSP: Added built-in preprocessor condition NKSP_NO_MESSAGE, which
676 can be set to disable all subsequent built-in "message()" function calls
677 on preprocessor level.
678 - Implemented built-in script function "change_sustain()".
679 - NKSP script editor syntax highlighting API: catch all fatal lexer errors,
680 to avoid the editor app to crash on ill-formed text input.
681 - Added built-in script function "change_pan_time()".
682 - Added built-in script function "change_pan_curve()".
683 - Added built-in script function "change_cutoff_attack()".
684 - Added built-in script function "change_cutoff_decay()".
685 - Added built-in script function "change_cutoff_sustain()".
686 - Added built-in script function "change_cutoff_release()".
687 - Added built-in script function "change_cutoff_lfo_depth()".
688 - Added built-in script function "change_cutoff_lfo_freq()".
690 * Instruments DB:
691 - Fixed memory access bug of general DB access code which lead to
692 undefined behavior.
693 - Cleanup of instruments DB file creation and opening code.
694 - The instrument DB path of linuxsampler's --create-instruments-db argument
695 is now optional, if it is missing, then a default location is used.
696 - Added support for scanning SFZ (.sfz) files.
697 - Added support for scanning Sound Font (.sf2) files.
698 - Fixed undefined DB transaction behavior.
700 Version 2.0.0 (15 July 2015)
702 * packaging changes:
703 - fixed building with newer MinGW-w64
704 - Mac OS X: support the new dir for Core Audio SDK
705 - Mac OS X: fixed building outside source directory
706 - made sure all source files for hostplugins are included when
707 doing "make dist"
708 - removed empty directories left from the cvs to svn migration
709 - fixed compilation with gcc 4.6.1
710 - another "make dist" fix, for the LV2 plugin
711 - made --enable-pthread-testcancel default on Mac OS X
712 - minor Makefile fix for building lscpparser when build and source
713 directories are separate
714 - Mac OS X: made it possible to specify plugin installation dir to
715 configure
716 - Mac OS X: Makefile fix for the install-strip target
717 - fixed compilation with gcc 4.7
718 - fixed configure script error with old autoconf versions
719 - lsatomic.h: use gcc provided atomic functions if building with
720 gcc 4.7 and C++11
721 - modernized configure script
722 - fixed linkage error when building with
723 LDFLAGS="-Wl,--no-undefined" (#190)
724 - fixed compilation with Clang 3.2
725 - removed usage of deprecated Automake variable INCLUDES
726 - fixed building with C++11
727 - build fix: ChangeFlagRelaxed.h was missing in makefile
728 - build fix: libsndfile compiler flags were missing in some
729 makefiles
730 - fix for building with bison 3.0 (#202)
731 - Mac OS X: added temporary hack allowing to spawn gigedit as callback
732 on the process's main thread
733 - fixed build error on newer MinGW
734 - support building with older jack versions
735 - support building with spaces in vst sdk path
736 - enabled automake 'subdir-objects' option and moved external
737 source references (vst, au, asio) from makefiles to cpp files,
738 in order to get rid of warnings from automake 1.14
740 * general changes:
741 - Refactoring: moved the independent code from
742 the Gigasampler format engine to base classes
743 - command line option '--profile' is currently disabled, since the
744 respective profiling code is currently broken
745 - Introduced support for optional environment variable
746 "LINUXSAMPLER_PLUGIN_DIR" which allows to override the directory
747 where the sampler shall look for instrument editor plugins
748 (patch by Luis Garrido, slightly modified).
749 - implemented Roland GS NRPN 1ArrH which allows to set volume per note
750 - implemented Roland GS NRPN 1CrrH which allows to set pan per note
751 - implemented Roland GS NRPN 1DrrH which allows to set reverb send per
752 note (in this implementation of the sampler its simply hard routed to
753 the 1st effect send of the sampler channel, no matter what the actual
754 effect type is)
755 - implemented Roland GS NRPN 1ErrH which allows to set chorus send per
756 note (in this implementation of the sampler its simply hard routed to
757 the 2nd effect send of the sampler channel, no matter what the actual
758 effect type is)
759 - implemented support for internal LADSPA effects
760 - raised limit of program change queue from 100 to 512
761 (as suggested by Alex Stone)
762 - implemented sine LFO, pulse LFO and saw LFO
763 - experimental support for per voice equalization
764 - added command line option --exec-after-init
765 - Introduced new C++ API method:
766 EngineChannel::InstrumentFileName(int index)
767 allowing to retrieve the whole list of files used for the loaded
768 instrument on an engine channel (a.k.a. part). Some GigaStudio
769 instruments for example are splitted over several files like
770 "Foo.gig", "Foo.gx01", "Foo.gx02", ...
771 - Added new C++ API method Sampler::GetGlobalMaxVoices().
772 - Added new C++ API method Sampler::GetGlobalMaxStreams().
773 - Added new C++ API method Sampler::SetGlobalMaxVoices().
774 - Added new C++ API method Sampler::SetGlobalMaxStreams().
775 - Various "const" and "restrict" optimizations.
776 - all engines: add pan CC value to instrument pan parameter before
777 applying panning, instead of using two separate pan functions in
778 series (#182)
779 - added a lock guard class for exception safe mutex handling and
780 used it everywhere appropriate
781 - Immediately apply scale tuning changes to active voices.
782 - Exposed scale tuning to C++ API (along to the already existing standard
783 SysEx way).
784 - lsatomic.h fixes: seq_cst load and store were suboptimal for x86
785 and broken for ppc64. (Seq_cst loads and stores are actually not
786 used in LS, so the bug wasn't noticable.)
787 - lsatomic.h: added ARMv7 support
788 - Added support for multiple MIDI input ports per sampler channel (and
789 added various new C++ methods for this new feature / design change, old
790 C++ API methods for managing SamplerChannel's MIDI inputs are now marked
791 as deprecated but are still there and should provide full behavior
792 backward compatibility).
793 - AbstractEngine::GSChecksum(): don't allocate memory on the stack (was
794 unsafe and caused compilation error with clang 2.x).
795 - Bugfix: only process the latest MIDI program change event.
796 - Introducing the LSCP shell, which provides convenient control of the
797 sampler from the command line by providing LSCP aware features.
798 - VirtualMidiDevice: Added support for program change.
799 - VirtualMidiDevice: Added support for bank select (MSB & LSB).
800 - VirtualMidiDevice: Added support for pitch bend.
801 - Aftertouch: extended API to explicitly handle channel pressure and
802 polyphonic key pressure events (so far polyphonic pressure was not
803 supported at all, and channel pressure was rerouted as CC128 but not
804 used so far).
805 - Added initial support for real-time instrument scripts. The script VM
806 code is shared by all sampler engine implemementations, however only the
807 gig file format currently provides support for storing instrument scripts
808 (as LinuxSampler extension to the original GigaStudio 4 format).
810 * Gigasampler format engine:
811 - implemented the "round robin keyboard" dimension
812 - fixed round robin and random dimensions for cases when number of
813 dimension zones is not a power of two
814 - made round robin use a counter for each region instead of each
815 key
816 - bugfix: pitch LFO controller "internal+aftertouch" was broken
817 - bugfix: filter keyboard tracking was broken
818 - filter performance fix (an unnecessary copy was made of the
819 filter parameters in each sub fragment)
820 - handle special case when pan parameter in gig file has max or
821 min value
822 - Exclusive Groups: don't ever stop voices of the same note,
823 doesn't sound naturally with a drumkit
824 - fixed EG1 modulation when attack or release is zero
825 - Fixed support for 'aftertouch' attenuation controller.
826 - Fixed crash that happened with velocity split sounds under certain
827 conditions (see also previous commit on libgig).
828 - fixed behaviour of filter LFO
830 * SFZ format engine:
831 - Initial implementation (not usable yet)
832 - added support for v2 multiple stage envelope generators
833 - added a fine-tuned v1 envelope generator instead of using the
834 one from the gig engine
835 - fixed memory leak and memory handling errors
836 - added support for trigger=first, trigger=legato and sw_previous
837 - allow non-numerical key values ("C#4" for example)
838 - "key" opcode now sets pitch_keycenter too
839 - fixed error when unloading instrument with same sample used by
840 multiple regions
841 - added some opcode aliases, like loopmode for loop_mode, to be
842 more compatible
843 - added support for trigger=release and rt_decay
844 - added support for off_mode=normal
845 - added support for random, seq_position, seq_length and volume
846 - added v1 LFO opcodes to parser (no support in engine yet)
847 - added support for amp_veltrack and amp_velcurve_N
848 - fine-tuned the default velocity curve
849 - added support for transpose
850 - fixed crash when using sw_down/up
851 - improved logic for sw_lokey/hikey/up/down/last
852 - added more v1 aliases to parser
853 - reduced memory usage for sfz data
854 - RT-safeness: avoid malloc in audio thread
855 - fixed a bug that could cause voice stealing to fail
856 - optimized sample lookup
857 - improved support for exclusive groups (group, off_by and
858 off_mode)
859 - added support for controller triggered regions (on_locc/on_hicc)
860 - added support for loop_mode=one_shot
861 - fixed playback of 16 bit wav files on big endian CPUs
862 - added support for Ogg Vorbis sample files
863 - use loop markers from sample file if loop_start and loop_end are
864 not set in sfz file
865 - implemeted filters. Filter types: lowpass, bandpass, bandreject
866 and highpass. 1, 2, 4 and 6 pole filters. Opcodes: fil_type,
867 cutoff, resonance, fil_veltrack, fil_keytrack, fil_keycenter,
868 cutoff_cc, cutoff_chanaft.
869 - bugfix: zero ampeg_sustain didn't work
870 - added support for velocity effect on amplifier envelope time
871 (ampeg_vel2attack, ampeg_vel2decay, ampeg_vel2sustain and
872 ampeg_vel2release)
873 - added support for EG hold (ampeg_hold)
874 - added support for sample offset (offset)
875 - use the newly introduced signal units model
876 - implemented opcodes lfoN_delay,
877 lfoN_freq, lfoN_pan, lfoN_cutoff, lfoN_resonance
878 - implemented opcodes ampeg_delay, ampeg_vel2delay
879 pitcheg_delay, pitcheg_start, pitcheg_attack, pitcheg_hold,
880 pitcheg_decay, pitcheg_sustain, pitcheg_release, pitcheg_vel2delay,
881 pitcheg_vel2attack, pitcheg_vel2hold, pitcheg_vel2decay,
882 pitcheg_vel2sustain, pitcheg_vel2release, pitcheg_depth
883 - implemented opcodes pitchlfo_delay, pitchlfo_freq,
884 pitchlfo_depth, fillfo_delay, fillfo_freq, fillfo_depth,
885 amplfo_delay, amplfo_freq, amplfo_depth
886 - implemented opcodes fileg_delay, fileg_start, fileg_attack,
887 fileg_hold, fileg_decay, fileg_sustain, fileg_release, fileg_vel2delay,
888 fileg_vel2attack, fileg_vel2hold, fileg_vel2decay, fileg_vel2sustain,
889 fileg_vel2release
890 - implemented opcode lfoN_wave
891 - implemented opcode pitchlfo_depthccN
892 - implemented opcodes lfoN_phase, lfoN_phase_onccX,
893 lfoN_pitch, lfoN_pitch_onccX
894 - implemented opcodes pitchlfo_fade,
895 fillfo_fade, amplfo_fade, lfoN_fade, lfoN_fade_onccX
896 - implemented opcodes fillfo_freqccN,
897 pitchlfo_freqccN, amplfo_freqccN, lfoN_freq_onccX
898 - implemented opcodes ampeg_delayccN, ampeg_startccN,
899 ampeg_attackccN, ampeg_holdccN, ampeg_decayccN, ampeg_sustainccN,
900 ampeg_releaseccN, egN_timeX_onccY, egN_levelX_onccY
901 - lfoN_* and egN_* opcodes defined in group sections
902 are now taken into account
903 - implemented curves
904 - implemented opcodes volume_onccN, volume_curveccN
905 - implemented opcode volume_smoothccN
906 - implemented opcodes fillfo_depthccN, amplfo_depthccN,
907 lfoN_volume, lfoN_volume_onccX, lfoN_volume_smoothccX,
908 lfoN_freq_smoothccX, lfoN_pitch_smoothccX, lfoN_pan_onccX,
909 lfoN_pan_smoothccX, lfoN_cutoff_onccX, lfoN_cutoff_smoothccX,
910 lfoN_resonance_onccX, lfoN_resonance_smoothccX, lfoN_delay_onccX
911 - implemented opcode end
912 - implemented opcodes egN_amplitude_onccX,
913 egN_volume, egN_volume_onccX, egN_cutoff, egN_cutoff_onccX,
914 egN_pitch, egN_pitch_onccX, egN_resonance, egN_resonance_onccX
915 - implemented opcodes xfin_lokey, xfin_hikey,
916 xfout_lokey, xfout_hikey, xf_keycurve, xfin_lovel, xfin_hivel,
917 xfout_lovel, xfout_hivel, xf_velcurve, xfin_loccN, xfin_hiccN,
918 xfout_loccN, xfout_hiccN, xf_cccurve
919 - implemented opcodes pan_onccN, pan_smoothccN,
920 pan_curveccN, egN_pan, egN_pan_curve, egN_pan_onccX,
921 egN_pan_curveccX, pitch_veltrack
922 - when failed to parse a sfz file
923 print the line number on which the error occurs
924 - use common pool of CC objects to minimize RAM usage
925 - implemented opcodes amplfo_delay_onccN,
926 amplfo_fade_onccN, fillfo_delay_onccN, fillfo_fade_onccN,
927 pitchlfo_delay_onccN, pitchlfo_fade_onccN
928 - implemented opcodes fileg_delay_onccN,
929 fileg_start_onccN, fileg_attack_onccN, fileg_hold_onccN,
930 fileg_decay_onccN, fileg_sustain_onccN, fileg_release_onccN,
931 fileg_depth_onccN, pitcheg_delay_onccN, pitcheg_start_onccN,
932 pitcheg_attack_onccN, pitcheg_hold_onccN, pitcheg_decay_onccN,
933 pitcheg_sustain_onccN, pitcheg_release_onccN, pitcheg_depth_onccN
934 - implemented automatic aliasing *ccN <-> *_onccN
935 - *lfo_freqccN wasn't working when the respective *lfo_freq
936 was not set or was set to zero
937 - lfoN_freq_onccX wasn't working when lfoN_freq
938 was not set or was set to zero
939 - implemented opcodes resonance_onccN, resonance_smoothccN,
940 resonance_curveccN, cutoff_smoothccN, cutoff_curveccN
941 - implemented opcodes fillfo_depthchanaft,
942 fillfo_freqchanaft, amplfo_depthchanaft, amplfo_freqchanaft,
943 pitchlfo_depthchanaft, pitchlfo_freqchanaft
944 - implemented opcodes pitch_onccN,
945 pitch_curveccN, pitch_smoothccN, pitch_stepccN
946 - implemented opcodes volume_stepccN, pan_stepccN,
947 cutoff_stepccN, resonance_stepccN, lfoN_freq_stepccX,
948 lfoN_volume_stepccX, lfoN_pitch_stepccX, lfoN_pan_stepccX,
949 lfoN_cutoff_stepccX, lfoN_resonance_stepccX
950 - implemented opcodes eq1_freq, eq2_freq, eq3_freq,
951 eq1_freqccN, eq2_freqccN, eq3_freqccN, eq1_bw, eq2_bw, eq3_bw,
952 eq1_bwccN, eq2_bwccN, eq3_bwccN, eq1_gain, eq2_gain, eq3_gain,
953 eq1_gainccN, eq2_gainccN, eq3_gainccN
954 - implemented opcodes delay, delay_onccN, delay_random,
955 delay_samples, delay_samples_onccN
956 - implemented opcodes egN_eq1gain, egN_eq2gain, egN_eq3gain,
957 egN_eq1gain_onccX, egN_eq2gain_onccX, egN_eq3gain_onccX, egN_eq1freq,
958 egN_eq2freq, egN_eq3freq, egN_eq1freq_onccX, egN_eq2freq_onccX,
959 egN_eq3freq_onccX, egN_eq1bw, egN_eq2bw, egN_eq3bw, egN_eq1bw_onccX,
960 egN_eq2bw_onccX, egN_eq3bw_onccX, lfoN_eq1gain, lfoN_eq2gain,
961 lfoN_eq3gain, lfoN_eq1gain_onccX, lfoN_eq2gain_onccX, lfoN_eq3gain_onccX,
962 lfoN_eq1gain_smoothccX, lfoN_eq2gain_smoothccX, lfoN_eq3gain_smoothccX,
963 lfoN_eq1gain_stepccX, lfoN_eq2gain_stepccX, lfoN_eq3gain_stepccX,
964 lfoN_eq1freq, lfoN_eq2freq, lfoN_eq3freq, lfoN_eq1freq_onccX,
965 lfoN_eq2freq_onccX, lfoN_eq3freq_onccX, lfoN_eq1freq_smoothccX,
966 lfoN_eq2freq_smoothccX, lfoN_eq3freq_smoothccX, lfoN_eq1freq_stepccX,
967 lfoN_eq2freq_stepccX, lfoN_eq3freq_stepccX, lfoN_eq1bw, lfoN_eq2bw,
968 lfoN_eq3bw, lfoN_eq1bw_onccX, lfoN_eq2bw_onccX, lfoN_eq3bw_onccX,
969 lfoN_eq1bw_smoothccX, lfoN_eq2bw_smoothccX, lfoN_eq3bw_smoothccX,
970 lfoN_eq1bw_stepccX, lfoN_eq2bw_stepccX, lfoN_eq3bw_stepccX
971 - implemented opcodes eq1_vel2freq, eq2_vel2freq,
972 eq3_vel2freq, eq1_vel2gain, eq2_vel2gain, eq3_vel2gain
973 - sfz parser: allow double spaces in sample filenames
974 - sfz parser: allow absolute paths for sample filenames
975 - use linear decay and release for filter and pitch EG
976 - bugfix: only the first amp_veltrack definition in a file was
977 used
978 - bugfix: looping was disabled if loop_start was set to 0
979 - allow regions with end=-1 to turn off other regions using the
980 group and off_by opcodes (#168)
981 - made end=0 play the whole sample
982 - fixed support for lochan and hichan opcodes (#155)
983 - fixed crash when using lochan/hichan opcodes (#187)
984 - sfz parser: allow -200 to 200 for pan_oncc opcode (#182)
985 - added FLAC support (#191)
986 - sfz parser bugfix: lines starting with whitespace were ignored
987 - added amplitude opcode
988 - added support for "#include" instruction
989 (modified patch which was originally posted by Sergey on LS mailing list)
990 - bugfix: generation of velocity curves etc should not be done
991 after each "#include", only after the main file is parsed
992 - bugfix: line numbers in error messages were wrong after "#include"
993 - added support for float and 32 bit sample files
995 * SoundFont format engine:
996 - Initial implementation (not usable yet)
997 - RT-safeness: avoid malloc in audio thread
998 - fixed a bug that could cause voice stealing to fail
999 - fine-tuned amplitude EG (by switching from gig to sfz EG)
1000 - initial implementation of Vibrato LFO and Modulation LFO
1001 - initial implementation of cutoff filter
1002 - use linear decay and release for filter and pitch EG
1004 * Host plugins (VST, AU, LV2, DSSI):
1005 - AU bugfix: failed to destroy its audio/MIDI devices
1006 - Listen to all interfaces on Mac OS X (INADDR_ANY)
1007 - VST bugfix: If the host called resume() before and after
1008 changing sample rate or block size, the number of channels was
1009 incorrectly set to two. This caused silence in Cubase 5.
1010 - save engine type (gig, sfz or sf2) in plugin state
1011 - VST: when opening Fantasia, look for both 32 and 64 bit Java on
1012 64 bit Windows
1013 - AU: changed number of output channels from one stereo to 16
1014 stereo
1015 - VST: made it possible to build the VST plugin for Mac
1016 - AU: link AU plugin dynamically if --disable-shared isn't
1017 specified
1018 - LV2 "state" extension support (patch by David Robillard)
1019 - VST bugfix: instrument loading hang and crashed the host when
1020 the plugin was loaded a second time (#174)
1021 - plugin bugfix: instrument loading hang when the plugin was
1022 loaded a second time (this time it's for Linux and Mac, previous
1023 similar fix was for Windows)
1024 - thread safety fixes for the instrument loading thread
1025 - LV2: use the new lv2 package if present
1026 - VST: try to open Fantasia automatically on Linux and Mac too (on
1027 Linux, the Fantasia jar should be placed in <prefix>/share/java)
1028 - VST: fixed crashes on Linux Ardour and EnergyXT
1029 - DSSI bugfix: it wasn't possible to change engine type. The MIDI
1030 port and audio channel routing for DSSI plugins are now visible.
1031 - LV2: use urid and atom extensions instead of deprecated uri-map
1032 and event
1033 - LV2: lv2 package 1.0 is now required to build the LV2 plugin
1034 - LV2: changed number of output channels to 16 stereo, just like
1035 the VST and AU plugins
1036 - LV2: fixed save/restore of SFZ state (patch by David Robillard)
1037 - LV2: made LV2 plugin buildable on Windows and Mac
1038 - VST: implemented retrieval and switching of programs using the
1039 sampler's internal MIDI instrument mapping system
1041 * Instrument editor interface:
1042 - Changed instrument editor plugin interface, providing additional
1043 informations like the EngineChannel for which the instrument editor was
1044 spawned for. This allows the instrument editors to interact more actively
1045 with the sampler.
1047 * MIDI driver:
1048 - ALSA MIDI driver supports now "NAME" device parameter, for overriding
1049 the ALSA sequencer client name
1050 - removed limit of maximum amount of MIDI ports per MIDI device, since
1051 there is no reason for this limit
1052 - MME: fixed memory handling bug found with cppcheck
1053 - MME: removed compiler warning
1054 - CoreMIDI: implemented driver specific port parameter "CORE_MIDI_BINDINGS",
1055 which allows to retrieve the list of CoreMIDI clients / ports and to connect
1056 to them a la JACK, via the usual sampler APIs
1057 - CoreMIDI: added driver specific port parameter "AUTO_BIND", if enabled
1058 the driver will automatically connect to other CoreMIDI clients' ports
1059 (e.g. external MIDI devices being attached to the Mac)
1060 - added support for MIDI note on velocity filter
1061 - CoreMIDI: fixed memory deallocation error
1062 - Fixed variable underflow in VirtualMidiDevice, which caused graphical
1063 virtual keyboards in frontends / instrument editors being stuck.
1064 - Bugfix in VirtualMidiDevice: process note on with velocity 0 as note off.
1065 - Implemented missing handling of MIDI "running status".
1066 - CoreMIDI fix: a MIDIPacket can contain more than one event per packet.
1067 - MME bugfix: driver wasn't closed properly
1068 - CoreMIDI: automatically connect to all input sources by default (driver
1069 parameter "AUTO_BIND").
1070 - CoreMIDI: fixed auto bind feature to CoreMIDI ports that go online
1071 - CoreMIDI: fixed minor memory leak
1073 * audio driver:
1074 - ASIO driver fixes for newer gcc versions (fix from PortAudio)
1075 - JACK audio: react on sample rate changes.
1076 - JACK audio: react on buffer size changes.
1077 - JACK audio: jack_port_get_buffer() was cached and called outside
1078 RT context.
1079 - ASIO driver: removed compiler warnings
1080 - CoreAudio: fixed minor error handling bug
1081 - ASIO driver: be more verbose when no ASIO card could be found (fixes #203)
1082 - JACK audio: return the JACK server's current sample rate as default value
1083 for audio device parameter "SAMPLERATE" (fixes #166).
1085 * LSCP server:
1086 - added support for sending MIDI CC messages via LSCP command
1087 "SEND CHANNEL MIDI_DATA CC <sampler-chan> <ctrl> <val>"
1088 - added LSCP command "GET AVAILABLE_EFFECTS"
1089 - added LSCP command "LIST AVAILABLE_EFFECTS"
1090 - added LSCP command "GET EFFECT INFO <effect-index>"
1091 - added LSCP command "CREATE EFFECT_INSTANCE <effect-index>"
1092 - added LSCP command
1093 "CREATE EFFECT_INSTANCE <effect-system> <module> <effect-name>"
1094 - added LSCP command "DESTROY EFFECT_INSTANCE <effect-instance>"
1095 - added LSCP command "GET EFFECT_INSTANCES"
1096 - added LSCP command "LIST EFFECT_INSTANCES"
1097 - added LSCP command "GET EFFECT_INSTANCE INFO <effect-instance>"
1098 - added LSCP command
1099 "GET EFFECT_INSTANCE_INPUT_CONTROL INFO <effect-instance> <input-control>"
1101 <effect-instance> <input-control> <value>"
1102 - added LSCP command "GET SEND_EFFECT_CHAINS <audio-device>"
1103 - added LSCP command "LIST SEND_EFFECT_CHAINS <audio-device>"
1104 - added LSCP command "ADD SEND_EFFECT_CHAIN <audio-device>"
1105 - added LSCP command
1106 "REMOVE SEND_EFFECT_CHAIN <audio-device> <effect-chain>"
1107 - added LSCP command
1108 "GET SEND_EFFECT_CHAIN INFO <audio-device> <effect-chain>"
1109 - added LSCP command "APPEND SEND_EFFECT_CHAIN EFFECT <audio-device>
1110 <effect-chain> <effect-instance>"
1111 - added LSCP command "INSERT SEND_EFFECT_CHAIN EFFECT <audio-device>
1112 <effect-chain> <effect-chain-pos> <effect-instance>"
1113 - added LSCP command "REMOVE SEND_EFFECT_CHAIN EFFECT <audio-device>
1114 <effect-chain> <chain-pos>"
1115 - added LSCP command "SET FX_SEND EFFECT <sampler_channel>
1116 <fx_send_id> <effect_chain> <chain_pos>"
1117 - added LSCP command "REMOVE FX_SEND EFFECT <sampler_channel> <fx_send_id>"
1121 - provide comprehensive error messages on LSCP syntax errors
1122 (suggesting expected next non-terminal symbols)
1123 - Fixed client connection not being closed after network errors.
1125 * LSCP shell:
1126 - Added support for auto correction of obvious and trivial syntax mistakes.
1127 - Added support for auto completion by tab key.
1128 - Show currently available auto completion while typing.
1129 - Added support for browsing command history with up / down keys.
1130 - Show all possible next symbols immediately right to the current command
1131 line while typing (no double tab required for this feature, as it would
1132 be the case in other shells).
1133 - Added support for moving cursor left/right with arrow keys.
1134 - Added support for built-in LSCP reference documentation, which will
1135 automatically show the relevant LSCP reference section on screen as soon
1136 as one specific LSCP command was detected while typing on the command
1137 line.
1139 * Real-time instrument scripts:
1140 - Implemented built-in script array variable %CC.
1141 - Implemented built-in script int variable $CC_NUM.
1142 - Implemented built-in script int variable $EVENT_NOTE.
1143 - Implemented built-in script int variable $EVENT_VELOCITY.
1144 - Implemented built-in script constant variable $VCC_MONO_AT.
1145 - Implemented built-in script constant variable $VCC_PITCH_BEND.
1146 - Implemented execution of script event handler "init".
1147 - Implemented execution of script event handler "controller".
1148 - Implemented execution of script event handler "note".
1149 - Implemented execution of script event handler "release".
1150 - Implemented built-in script function "play_note()" (only two of the
1151 max. four function arguments are currently implemented yet though).
1152 - Implemented built-in script int variable $EVENT_ID.
1153 - Implemented built-in script function "ignore_event()"
1154 - Implemented built-in script function "ignore_controller()" (may have one
1155 or no argument).
1156 - Implemented built-in script function "set_controller()".
1157 - Added extended script VM for the Gigasampler/GigaStudio format sampler
1158 engine, which extends the general instrument script VM with Giga format
1159 specific variables and functions.
1160 - Giga format scripts: added built-in script int constant variables
1171 - Giga format scripts: Implemented built-in script function
1172 "gig_set_dim_zone(event_id, dimension, zone)".
1173 - Implemented built-in script int array variable %KEY_DOWN.
1174 - Implemented built-in script function "abs()".
1175 - Implemented built-in script function "random()".
1176 - Implemented built-in script function "num_elements()".
1177 - Implemented built-in script function "note_off()".
1178 - Implemented built-in script function "set_event_mark()".
1179 - Implemented built-in script function "delete_event_mark()".
1180 - Implemented built-in script function "by_marks()".
1181 - Added built-in script int const variables $MARK_1 to $MARK_28.
1182 - Built-in script functions "ignore_event()", "note_off()" and
1183 "gig_set_dim_zone()" now also accept an array of event IDs as argument
1184 (i.e. return value of new script function "by_marks()").
1185 - Pass/preserve polyphonic variable data from respective "note" event
1186 handler to "release" event handler.
1187 - Fixed crash when using built-in script function "by_marks()".
1189 * Bug fixes:
1190 - Fixed crash which may occur when MIDI key + transpose is out of range
1191 - minor valgrind fixes
1192 - fixed crash which occurred when changing an already deployed sampler
1193 channel to a different engine type
1194 - fixed crash when deleting a sampler channel or changing engine
1195 type while an instrument load was in progress
1196 - bugfix: playing a note while changing the instrument could cause
1197 a crash, or give "Handing back unknown region" error messages
1198 - bugfix: calling SET VOICES (which Fantasia does on start and
1199 refresh) could cause instruments to be unloaded
1200 - Mac OS X: fixed name collision of enum in EffectControl and
1201 wrong name of destructor in AudioOutputDeviceCoreAudio.cpp
1202 - Mac OS X: fixed hanging threads
1203 - Windows: disabled the previous thread fix on non-Mac systems, as
1204 it caused hanging threads on Windows
1205 - Fixed possible crashes due to corrupted MIDI/audio device list
1206 after MIDI/audio device creation failure
1207 - When creating MIDI instrument map entries with "PERSISTENT" type, the
1208 instruments were uselessly precached with zero samples, however it still
1209 took the full preloading time and on 1st program change the respective
1210 instrument was completely reloaded again.
1211 - fixed handling of rapid bank select and program change messages
1212 sent to the same sampler channel (patch from the Open Octave
1213 project, slightly adjusted)
1214 - fixed crash when trying to create an effect instance with controls
1215 which min and/or max values depend on the sample rate
1216 - fixed bug #162
1217 - bugfix: LADSPA_PATH was not evaluated correctly when containing
1218 multiple paths (#165)
1219 - thread safety fixes for the instrument loading thread
1220 - bugfix: instrument loading crashed for sfz and sf2 in Ardour
1221 (#176)
1222 - more thread safety fixes for the instrument loading thread
1223 - sfz/sf2 engine: fixed crash when using small audio fragment size
1224 - Mac OS X: fixed crash when unloading plugin on 10.7 and later
1225 - Mac OS X: fixed process hang when unloading 32-bit plugin (bug
1226 introduced in previous fix)
1227 - fixed crash when a channel received a program change while
1228 playing a note in a key group
1229 - fixed erroneous error message piping in VoiceBase.h
1230 (labelled "Disk stream not available in time")
1231 - Update effects on sample rate & period size changes (to avoid
1232 crashes and noise on such transitions).
1233 - ignore missing LADSPA paths without ignoring valid LADSPA paths
1234 (fixes #208)
1235 - Fixed bug in LSCP grammar definition which caused a statement like
1236 "GET SERVER INFOasdf\n" to be accepted as valid statement (was so far
1237 practically irrelevant, however it caused problems with the new LSCP
1238 shell's auto completion feature).
1239 - Fixed MIDI program change messages being ignored if quickly executed
1240 after each other (fixes #231).
1242 Version 1.0.0 (31 July 2009)
1244 * packaging changes:
1245 - autoconf bugfix: the PKG_CONFIG variable wasn't initialized properly,
1246 causing e.g. the libgig test to fail when
1247 "./configure --disable-jack-driver" was used
1248 (patch by Alexis Ballier)
1249 - fixed compilation with gcc 4.3
1250 - fixes for building on OS X (thanks to Ebrahim Mayat for testing)
1251 - fixed configure so it detects x86_64 (#107)
1252 - fixes for building with newer MinGW versions
1253 - fix for building with bison 2.4 (#111)
1254 - fixed building with libgig installed in a non-standard directory
1255 - minor fix in configure for mmsystem.h detection on MinGW
1256 - Windows: look for editor plugins and Fantasia using base
1257 directory of liblinuxsampler dll (look in the same directory and one
1258 directory above)
1259 - configure script fix: removed unconditional use of SSE
1260 - fixed building with sqlite installed in a non-standard directory
1261 - when cross-compiling, don't try to create instruments.db
1262 - fix for new mingw-w64 version, which has usleep
1264 * general changes:
1265 - bugfix: on some POSIX systems instrument editor plugins refused to
1266 load as we used a non-portable Linux specific struct field
1267 (fixes bug #70, patch by Ronald Baljeu)
1268 - fixed endless loop which occured when loading an instrument editor
1269 plugin DLL on Windows systems failed
1270 - fixed memory leaks that occurred when liblinuxsampler was unloaded
1271 - optimized the SynchronizedConfig class so it doesn't wait
1272 unnecessarily long after an update
1273 - added support for notifying instrument editors on note-on / note-off
1274 events (e.g. to highlight the pressed keys on the virtual keyboard
1275 of gigedit)
1276 - added support for triggering notes by instrument editors (see above)
1277 - be verbose on DLL load errors (on Linux)
1278 - fixed di-harmonic triangle LFO implementation (this LFO implementation
1279 is only used by very few systems, most pick the int math
1280 implementation, which is usually faster)
1281 - fixes for audio drivers with varying buffer sizes
1282 - experimental support for running LinuxSampler as a DSSI, LV2 and
1283 VST plugin
1284 - notification events for stream/voice count statistics are now sent
1285 only when there are actual changes
1286 - added memory ordering constraints to improve stability on
1287 multi-core and multi-cpu systems
1288 - maximum voices and disk streams can now be altered at runtime
1289 - fixed CPU feature detection on x86_64 (maybe fixes #108)
1290 - automatic stacktrace mechanism is now turned off by default and can
1291 be switched on by command line option "--stacktrace" (the automatic
1292 stacktrace mechanism seems to be broken on most systems at the moment)
1293 - C++ API method InstrumentManager::LaunchInstrumentEditor() now returns
1294 a pointer to the launched InstrumentEditor object
1295 - added optional 3rd party user data parameter for following
1296 liblinuxsampler C++ API methods: InstrumentEditor::Main(),
1297 InstrumentEditor::Launch(),
1298 InstrumentManager::LaunchInstrumentEditor()
1299 - theoretical fix: made SynchronizedConfig follow C++0x memory
1300 model more strictly
1301 - fixes for using large audio device buffers
1302 - Windows: add the installation directory to the DLL search path
1303 when loading an editor plugin (solves problems with VST and
1304 gigedit on systems with other GTK versions installed)
1305 - updated linuxsampler man page
1307 * audio driver:
1308 - removed the nonsense audio channel constraint (which was hard coded to
1309 max. 100 audio channels) for most audio drivers
1310 - JACK audio driver did not offer a device parameter "SAMPLERATE" as
1311 opposed to the LSCP specs
1312 - bugfix: the SAMPLERATE parameter of some drivers (e.g. JACK)
1313 reflected the wrong value
1314 - fixed a memory management bug in ASIO driver
1315 - Makefile fix: JACK_CFLAGS wasn't used
1316 - JACK: use jack_client_open instead of the deprecated
1317 jack_client_new
1318 - added (experimental) CoreAudio driver
1319 - applied old fixes to the ASIO driver that were included in the
1320 previous binary release but accidentally never committed to CVS
1321 (fixes #117)
1322 - fixes for ASIO on mingw-w64 (iasio wrapper is not needed on
1323 win64)
1324 - VST: added support for sample rate and buffer size changes
1325 - VST: close editor (Fantasia) when the VST is removed
1326 - VST: avoid opening Fantasia more than once for each VST instance
1327 - VST: export main function as "main" on Linux too (fix for
1328 energyXT)
1329 - VST: changed number of output channels from one stereo to 16
1330 stereo
1331 - added channel routing, fxsends and midi maps to the settings
1332 stored in the plugin state
1333 - performance optimization of AudioChannel::MixTo() and
1334 AudioChannel::CopyTo() methods using GCC vector exensions
1335 (if available)
1336 - ASIO fixes: avoid initializing the device twice, avoid throwing
1337 exception when getting parameters from a disconnected device
1339 * MIDI driver:
1340 - added JACK MIDI driver
1341 - dispatch bank select as ordinary CC as well, the user might seriously
1342 want to (mis)use it for some purpose ("fixed" in all current MIDI
1343 input drivers: ALSA, CoreMIDI, JACK, MidiShare, MME)
1344 - bugfix: pitch bend wasn't working with JackMidi, VST, LV2, MME,
1345 CoreMidi or AU
1346 - fixed mingw-w64 compilation error in MME driver
1347 - made program change handling in MIDI thread real-time safe by
1348 moving the logic to a non-RT thread
1349 - fixed minor memory leak in ALSA driver
1351 * instruments database:
1352 - avoid time consuming samples scanning when adding instruments
1353 to the instruments database
1354 - added support for handling lost files in the instruments database
1355 - Implemented option for adding instruments in separate directories
1356 in the instruments database
1357 (patch by Chris Cherrett & Andrew Williams, a bit adjusted)
1358 - work-around for missing fnmatch function on Windows to make
1359 instrument database compilable
1360 - added instrument database support on Windows
1361 works with both standalone mode and VST plugin.
1362 the instruments DB file is located in
1363 %USERPROFILE%\.linuxsampler\instruments.db which allows different
1364 databases for each windows user
1365 if no DB is present it automatically creates the .linuxsampler subdir
1366 and then creates an empty DB
1367 - fixed recursive import, which was broken on Windows
1369 * Gigasampler format engine:
1370 - fixed a memory leak that could happen when a channel was deleted
1371 while notes were playing
1372 - made it possible to load an instrument even if the audio thread
1373 isn't running
1374 - added partial support for the "Controller Triggered" MIDI rule,
1375 enough for piano gigs with pedal noise samples
1376 - minor fix: only mark FX sends as being modified if really the
1377 respective FX send MIDI controller was used
1378 - added support for GM global device volume SysEx message
1379 (can be selected at compile time to either apply globally to the
1380 whole sampler [default] or only to the sampler channels that are
1381 connected to the respective MIDI input port the SysEx message arrived
1382 on)
1383 - bugfix: notes triggered at position 0 in the audio buffer were
1384 sometimes wrongly killed in the same buffer, causing no sound to
1385 be played
1386 - added support for chromatic / drumkit mode Roland GS Sysex message
1387 which is usally used to switch a part between chromatic sounds and
1388 drumkit sounds (as opposed to their common default setting of MIDI
1389 part 10 being a drumkit part and all other ones chromatic parts), in
1390 LS however one can switch between the first 16 MIDI instrument maps
1391 defined for the sampler
1392 - made it possible to create multiple sample channels even if the
1393 audio thread isn't running
1394 - fixed crash when removing channel with active voices (#116)
1395 - bugfix: on sample reference changes (instrument editor), only
1396 un-cache the respective sample if it's really not used by any
1397 sampler engine anymore
1398 - re-cache samples in case they were changed by an instrument editor,
1399 e.g. when a sample was added while playing (#82)
1400 - fixed hanging notes which occured when note-off event had the exact
1401 same time stamp as the voice's note-on event and occured both in the
1402 same audio fragment cycle (fixes bug #112)
1403 - added support for the "fine tune" and "pitch bend range"
1404 instrument-level gig parameters
1405 - fixed minor artifacts in pitch bend handling
1406 - added support for GS Reset SysEx message
1407 - allow gig files to use unlimited downward pitch shifting
1408 - added a limit check for upward pitch shifting
1409 - bugfix: sometimes, when playing a note twice fast, the second
1410 note was silent
1411 - fixed crash happening when a pitch bend event arrived at the
1412 same time a new instrument was loading
1414 * LSCP server:
1415 - added new LSCP event "CHANNEL_MIDI" which can be used by frontends to
1416 react on MIDI data arriving on certain sampler channels (so far only
1417 Note-On and Note-Off events are sent via this LSCP event)
1418 - added new LSCP event "DEVICE_MIDI" which can be used by frontends to
1419 react on MIDI data arriving on certain MIDI input devices (so far only
1420 Note-On and Note-Off events are sent via this LSCP event)
1421 - added new LSCP commands: FIND LOST DB_INSTRUMENT_FILES and
1423 - added new LSCP command: SEND CHANNEL MIDI_DATA which can be used by
1424 frontends to send MIDI messages to specific sampler channel
1425 - added two additional fields to GET FILE INSTRUMENT INFO command -
1427 - bugfix: the bank number provided by MIDI_INSTRUMENT_INFO notifications
1428 was incorrect
1429 - Optimized the retrieval of the MIDI instrument mappings
1430 - added new LSCP command "SET VOICES" to globally alter the maximum
1431 amount of voices, added new LSCP event "GLOBAL_INFO:VOICES" which will
1432 be triggered respectively
1433 - added new LSCP command "SET STREAMS" to globally alter the maximum
1434 amount of disk streams, added new LSCP event "GLOBAL_INFO:STREAMS"
1435 which will be triggered respectively
1436 - bugfix: retry if "select" returns EINTR (this fixes a crash when
1437 a gigedit file dialog is opened)
1438 - close all connections when LSCPServer is deleted
1439 - hard close of all sockets on exit
1440 - bugfix: SET CHANNEL MIDI_INPUT_TYPE didn't work with the MME
1441 driver
1443 * Bug fixes:
1444 - fixed a crash which occurs when removing a sampler channel waiting
1445 to start instrument loading after another channel
1446 - fixed a crash which occurs when removing a sampler channel with
1447 instrument loading in progress (bug #113)
1448 - fixed termination caused by uncaught exception when adding MIDI
1449 instrument with PERSISTENT load mode
1450 - fixed possible iterator invalidations when resetting the sampler
1451 - fixed memory leaks when issuing the following LSCP commands:
1456 - fixed possible compilation error when sqlite is not present
1457 - fixed orphaned pointers when setting maximum voices limit (bug #118)
1458 - fixed crash when changing the audio output device of a sampler
1459 channel with loaded instrument and start playing notes
1460 - fixed endless loop in Engine::SuspendAll() (bug #120)
1461 - fixed a low-level atomic load function that was broken on 64-bit
1462 PowerPC, which probably could cause crashes on that platform
1463 - fixed a memory management error which could cause a crash when a
1464 plugin was unloaded
1465 - bugfix: two private structs had the same name, which could cause
1466 problems if the linker chose the wrong constructor
1467 - fixed low-level ConditionServer usage bug that caused lockups on
1468 Windows
1471 Version 0.5.1 (6 December 2007)
1473 * packaging changes:
1474 - added autoconf checks for pthread library
1475 - added autoconf check for pthread bug found on certain NPTL-enabled
1476 glibc versions (see Gentoo bug report #194076)
1477 - added autoconf checks for MS Windows
1479 * general changes:
1480 - bugfix: the thread used by an editor plugin didn't die when the
1481 editor closed
1482 - bugfix: Ringbuffer.h: fill_write_space_with_null() did not zero
1483 out all the space. operator--() did not apply size_mask after
1484 decrementing the read_ptr. DEFAULT_WRAP_ELEMENTS set to 0 in
1485 order to avoid problems with the _NonVolatileReader functions.
1486 - bugfix: Stream.h: added a missing divide by BytesPerSample in
1487 GetWriteSpace(). Since this function is currently only used in
1488 the stream's qsort() compare function, it didn't trigger any
1489 bugs.
1490 - Resampler.h, Synthesizer.h: cubic interpolation now works in
1491 24bit mode too. Faster method to read 24bit words on little
1492 endian machines (x86): replaced 3 byte reads + shifts with a 1
1493 unaligned 32bit read and shift
1494 - experimental support for MS Windows (MIDI input via MME, AUDIO
1495 output via ASIO)
1496 - made handling of SIGINT signal (Ctrl-C) a bit more robust
1497 - support for monitoring the total number of active disk streams
1498 (new LSCP commands: GET TOTAL_STREAM_COUNT,
1501 * AUDIO driver:
1502 - added Windows ASIO low latency audio driver
1504 * MIDI driver:
1505 - added MME Windows MIDI driver
1507 * LSCP server:
1508 - added support for Windows style path / filenames, however with
1509 forward slash path separators instead of backslash
1510 (i.e. "C:/foo/bar.gig")
1511 - allow naughty liblscp to send non-string device parameters within
1512 apostrophes as well
1513 - added new LSCP commands: "GET FILE INSTRUMENTS <file>",
1514 "LIST FILE INSTRUMENTS <file>" and
1515 "GET FILE INSTRUMENT INFO <file> <index>" for retrieving informations
1516 about an arbitrary instrument file on the system where the sampler is
1517 running on
1519 Version 0.5.0 (15 October 2007)
1521 * packaging changes:
1522 - config.h is not going to be installed along with liblinuxsampler's
1523 API header files anymore
1524 - only the API relevant header (and source) files will be exported to
1525 the Doxygen API documentation (explicitly listed in Doxyfile.in)
1526 - added completely new XCode project files for Mac OSX which is now
1527 capable to execute our autoconf environment, thus no need anymore
1528 to maintain the compile time configuration file (osx/version.h) for
1529 OSX manually (patch by Toshi Nagata)
1530 - fixed buggy boolean --enable-foo / --disable-foo configure script
1531 parameters
1532 - global.h now only covers global definitions that are needed for the
1533 C++ API header files, all implementation internal global definitions
1534 are now in global_private.h
1535 - atomic.h is not exposed to the C++ API anymore
1536 - no need to include config.h anymore for using LS's API header files
1537 - fixed warnings in API doc generation
1539 * general changes:
1540 - replaced the old, confusing MIDI program change mechanism by a
1541 flexible MIDI instrument mapper which allows to map arbitrary
1542 (MIDI bank, MIDI program) pairs with arbitrary
1543 (engine type, instrument file, file index) triplets which will be
1544 loaded on the respective channel when such MIDI program change
1545 messages arrive, beside that, each entry allows to define a life-time
1546 strategy for the instrument, so the user can define whether the
1547 instrument should i.e. be permanently loaded (to be able to switch
1548 quickly among sounds) or i.e. loaded just on demand when the
1549 respective program change arrives (to safe RAM space), as well as a
1550 global volume factor for each entry, so the user can adjust the volume
1551 dynamic of his mapped instrument collection without having to modify
1552 the instrument files, also one can manage arbitrary amount of such
1553 MIDI instrument maps and assign each sampler channel individually
1554 a certain map, so that i.e. separation of normal instruments and
1555 drumkits is possible
1556 - new notification events for tracking changes to audio/MIDI devices,
1557 MIDI instrument maps, MIDI instruments, FX sends, global volume.
1558 - sampler was limited to load max. 200 instruments in the background
1559 due to a constant size RingBuffer FIFO which is now replaced by a
1560 dynamic (unlimited) size std::list FIFO
1561 - added FX sends, these allow to route audio signals to arbitrary audio
1562 output channels for being processed by external effect processors
1563 (i.e. jack-rack), the send levels are controllable via arbitrary MIDI
1564 controllers
1565 - global (sampler wide) volume can now be controlled at runtime
1566 - Implemented new, improved notification system
1567 - fixed compilation errors regarding OSX
1568 (patch by Toshi Nagata)
1569 - implemented instruments database
1570 - added support for escape sequences to the instruments database
1571 - added highly experimental support for on-the-fly instrument editing
1572 within the sampler's process (by using instrument editor plugins),
1573 you'll notice the new "Registered instrument editors:" message on
1574 startup, added a new LSCP command:
1575 "EDIT CHANNEL INSTRUMENT <sampler-channel>"
1576 to spawn a matching instrument editor for the instrument on the
1577 given sampler channel, the plugin path can be overridden at compile
1578 time with ./configure --enable-plugin-dir=/some/dir
1579 - added experimental code for synchronizing instrument editors hosted
1580 in the sampler's process to safely edit instruments while playing
1581 without a crash (hopefully) by either suspending single regions
1582 wherever possible or if unavoidable whole engine(s)
1583 - fixed several issues in fundamental "Thread" class: set scheduling
1584 policy and priority on thread level, set a minimum stack size for
1585 thread (TODO: a reasonable value yet to be tested), bugfix: non-RT
1586 threads simply inherited properties of starting thread instead of
1587 setting their own policy and priority
1588 - minor fix in our automatic stack trace mechanism on crashes, the
1589 main process did not wait for the stack trace process to finish
1590 its output
1591 - fixed some minor memory leaks
1592 - reenabled assembly features support, at the moment only for
1593 enabling a fast denormal FPU mode (x86 platforms supporting SSE2)
1594 - minor assembly fix in x86 features detection (don't use the PIC
1595 register, to avoid relocations in the text segment at runtime)
1596 - POSIX callback functions of Thread.h are hidden
1598 * MIDI driver:
1599 - dispatch bank select (MSB and LSB) messages
1601 * audio driver:
1602 - the ALSA audio output driver parameters now reflect the correct
1603 parameter value ranges for the respective selected sound card
1604 (patch by Till Wimmer, a bit fixed and extended)
1606 * Gigasampler format engine:
1607 - if a filter is used and EG2 finishes before EG1, let the voice
1608 die when EG2 has finished (this fixes a problem with clicks and
1609 voice starvation for some gigs)
1610 - playback is no longer disabled during instrument loading
1611 - all notes playing on a channel that changes its instrument keep
1612 playing with the old instrument until they get a note off
1613 command
1614 - EG fix: a release value of zero could cause noises or crash
1615 - handle MIDI coarse tuning messages (MIDI RPN #0 MSB #2 LSB)
1616 - EG fine tuning: when attack is zero the EG starts at a level
1617 above max sustain level, which means that there is a decay phase
1618 even if sustain is 100%
1619 - more EG fixes: the level could sometimes go below zero and cause
1620 noises or crashes
1621 - minor fix of EGDecay (patch by Toshi Nagata)
1622 - fixed compiler error when --enable-override-filter-type was
1623 supplied to the configure script (fixes #46)
1624 - disk thread: queue sizes are now proportional to CONFIG_MAX_STREAMS
1625 instead of a fix value
1626 - behavior fix: on MIDI CC# 65 (portamento on / off), 126 (mono mode),
1627 127 (solo mode) only kill voices if the respective mode really
1628 changed
1630 * LSCP server:
1631 - fixed compile time error for old Bison versions
1632 (i.e. v1.28 found on MacOS 10.4, patch by Toshi Nagata)
1633 - parser now supports extended ASCII character set
1634 (up to ASCII code 255, i.e. includes now umlauts and accents)
1635 - filename arguments in LSCP commands now allow to use escape
1636 sequences, that is directly literal as one of: \', \", \\, \n, \r,
1637 \f, \t, \v, or as octal ASCII code value like \132, or as hex ASCII
1638 code value like \xf2) (fixes bug #24)
1639 - the following LSCP commands now also support escape sequences for at
1640 least one of their text-based parameters (i.e. name, description):
1647 - returns verbose syntax errors (line and column where syntax error
1648 occured, the unexpected character and the actually expected, possible
1649 character(s), the latter only if less than 5 possibilities)
1650 - made sure that LSCP syntax is not affected by gigedit locale
1651 settings
1652 - bugfix regarding strings parameter lists: all comma separated lists
1653 of strings were treated as being one string containing commas
1654 (fixes #57)
1656 * Bug fixes:
1657 - fixed crash occurring on certain LSCP scripts (Bug 39)
1658 - another thread safety fix for lscp "load engine" and "set
1659 channel audio output device"
1660 - fixed a crash which occurs when reassigning the same engine
1661 on a sampler channel with connected MIDI device
1662 - fixed a crash which occurs when changing the number of ports of a MIDI
1663 device connected to a sampler channel to number less then or equal
1664 to the index of the port to which the sampler channel is connected.
1665 - The previous bindings were not been disconnected when altering
1666 the ALSA_SEQ_BINDINGS parameter. Introduced a NONE keyword for
1667 unsubscribing from all bindings (e.g. ALSA_SEQ_BINDINGS=NONE).
1668 - The active stream/voice count statistic was incorrect.
1669 - notification events were not been sent for some sampler
1670 channel changes
1671 - added default min and max values to restrict the number of allowed
1672 audio output channels and MIDI input ports
1673 - the connection to the PCM interface is now closed when destroying
1674 an audio output device
1675 - files with slash in their path or filename could not be loaded
1677 * test cases:
1678 - updated and fixed (haven't been touched in a while)
1680 Version 0.4.0 (24 November 2006)
1682 * packaging changes:
1683 - changed deprecated copyright attribute to license; added ldconfig
1684 to post-(un)install steps; added devel package for liblinuxsampler;
1685 to linuxsampler.spec (RPM)
1686 - install necessary development header files for allowing 3rd party
1687 applications to link against liblinuxsampler
1688 - liblinuxsampler's API documentation can be generated with 'make docs'
1689 (Doxygen required)
1690 - added benchmark to automatically detect the best triangle LFO
1691 implementation (currently either an integer math solution or a
1692 di-harmonic approximation), automatic detection can be overriden
1693 with --enable-signed-triang-algo=x and --enable-unsigned-triang-algo=x
1694 configure script argument though (mandatory for cross-compilation)
1695 - do not automatically pick optimized gcc flags if the user already
1696 provided some on his own (as CXXFLAGS)
1697 - added compile time option to disable processing of All-Notes-Off MIDI
1698 messages
1699 - added compile time options to allow disabling the various audio and
1700 MIDI drivers
1701 - fixed automatic GCC CPU switch detection on PPC
1702 (patch by Ebrahim Mayat)
1704 * Gigasampler format engine:
1705 - extensive synthesis optimization
1706 (reimplementation of EGs and LFO(s), removed synthesis parameter
1707 prerendering and the synthesis parameter matrix in general, splitting
1708 each audio fragment into subfragments now where each subfragment uses
1709 constant pitch and filter coefficients. The volume coefficient is
1710 linearly interpolated inside a subfragment, unless
1711 --disable-interpolate-volume is set.)
1712 - fine tuning of the EG modulation parameters
1713 - improved filter cutoff calculation by adding support for the
1714 following gig parameters: Cutoff freq (used when no cutoff
1715 controller is defined), Control invert, Minimum cutoff, Velocity
1716 curve and Velocity range. The keyboard tracking now scales
1717 cutoff frequency instead of resonance.
1718 - added support for gig parameter Resonance.
1719 - fixed bug in sysex handling (patch by Juan Linietsky)
1720 - added global volume attenuation of -9 dB (0.35f) to prevent clipping
1721 which can be overridden with --enable-global-attenuation
1722 - EG fixes: made the length of "attack hold" stage more
1723 accurate. Release stage can now start before attack stage
1724 ends. Cancel release didn't work when sustain was zero. Attack
1725 time now has a minimal value to prevent clicks.
1726 - fixed pitch changes larger than one octave
1727 - fixed EG3 (pitch envelope) synthesis which was neutral all the time
1728 - implemented portamento mode and solo mode (a.k.a 'mono mode'):
1729 all modes can be altered via standard GM messages, that is CC5 for
1730 altering portamento time, CC65 for enabling / disabling portamento
1731 mode, CC126 for enabling solo mode and CC127 for disabling solo mode
1732 - fine tuning of the curves for volume (CC7), pan (CC10 and gig
1733 parameter) and crossfade
1734 - added support for the "attenuation controller threshold" gig
1735 parameter
1736 - added smoothing of volume changes caused by control change
1737 messages
1738 - sample loop parameters are now taken from the DimensionRegion
1739 instead of the wave chunk
1740 - fixed keyswitching for v3 gigs with a number of keyswitch splits
1741 not equal to a power of two
1742 - reimplementation of the filter algorithm. The new filter is more
1743 accurate and supports all gig filter types, including bandreject
1744 and lowpass turbo.
1745 - real support for 24 bit samples - samples are not truncated to
1746 16 bits anymore
1747 - support for aftertouch (channel pressure, not polyphonic
1748 aftertouch)
1749 - LFO1 behaviour fixed (dampening from max volume instead of
1750 amplifying from 0)
1752 * LSCP server:
1753 - fixed application exit on broken pipe error (fixes bug #20)
1754 - fixed the notification messages delay due to lack of
1755 network activity (fixes bug #26)
1756 - fixed parser bug which occured on space(s) within device parameters
1759 * audio driver:
1760 - added aRts audio output driver (by no means RT safe)
1762 * MIDI driver:
1763 - fixed legacy sysex code which caused dispatching of MIDI SysEx
1764 messages several times instead of once
1765 - API extension for MIDI drivers which already supply exact time stamps
1766 for events (i.e. for offline rendering based MIDI drivers)
1767 - added checks for bad MIDI data
1769 * linuxsampler application:
1770 - show available sampler engine types on startup
1772 * general changes:
1773 - support for muting sampler channels and solo mode of the same, two new
1774 LSCP commands ("SET CHANNEL MUTE" and "SET CHANNEL SOLO") and two new
1775 fields ("MUTE" and "SOLO") for command "GET CHANNEL INFO" were
1776 introduced for this, the behavior is the same like on a mixer console
1777 (patch by Grigor Iliev, a bit adjusted). Also added configure option
1778 --enable-process-muted-channels which can be used to enable the
1779 processing of muted channels.
1780 - support for sostenuto pedal
1781 - support for monitoring the total number of active voices
1784 - fixed some memory management errors
1785 - fixed some concurrency problems that could lead to crashes when
1786 LSCP commands were executed
1787 - fixed crash when instrument loading failed previously on the same
1788 sampler channel (fixes bug #36)
1790 Version 0.3.3 (15 July 2005)
1792 * packaging changes:
1793 - fixed compilation with gcc 4.0
1795 * Gigasampler format engine:
1796 - LFO preprocessing was performed even though the respective LFO was
1797 disabled by the instrument patch (minor efficiency fix)
1798 - if period time of chosen audio device is too small (< MIN_RELEASE_TIME)
1799 for volume ramp downs in the same fragment (needed for current voice
1800 stealing implementation) then simply inform the user with a warning
1801 about possible click sounds and reduce the volume ramp down
1802 appropriately instead of cancelling the audio device connection
1804 Version 0.3.2 (24 June 2005)
1806 * packaging changes:
1807 - updated autotools build files to compile on OS X
1808 (conditional compilation of CoreMIDI and MidiShare drivers)
1809 - hand-crafted assembly optimization code can be disabled with
1810 './configure --disable-asm' (definitely not recommended)
1811 - fixed 'make dist' rule to include all necessary files
1812 - require automake (>= 1.5) for 'make -f Makefile.cvs'
1813 (due to 'dist-bzip2' automake option)
1815 * Gigasampler format engine:
1816 - support for the gig parameters for "release velocity response" curves
1817 - fine tuning of the EGADSR envelope
1818 - volume of release triggered samples now depends on note-on velocity,
1819 note length and gig parameter "release trigger decay" instead of
1820 note-off velocity.
1821 - revised voice stealing
1822 (fixes crash and endless loop caused by voice stealing)
1823 - don't reset scale tuning on instrument or audio output device change
1824 - handle key group conflicts right at the beginning of each fragment
1825 instead of when the respective voice is actually launched
1826 (fixes undefined behavior if stolen voices belonged to a key group -
1827 this case was followed by a "killed voice survived" error message)
1828 - fixed minor issue with null/silence samples
1829 (those stole voices even though they don't need a voice at all which
1830 resulted in "voice stealing didn't work out" messages)
1831 - don't reset volume, pan, pitch and MIDI controller values on
1832 instrument or audio output device change
1834 * LSCP server:
1835 - fixed some crashes (patch by Grigor Iliev, fixes #19)
1836 - fixed LSCP event "CHANNEL_INFO" notification
1837 (e.g. did not notify on volume changes or MIDI program change events)
1839 * linuxsampler application:
1840 - added command line parameters --lscp-addr and --lscp-port to override
1841 default IP address and port of LSCP server
1843 Version 0.3.1 (24 May 2005)
1845 * initial release

  ViewVC Help
Powered by ViewVC