ViewVC logotype

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 3836 - (show annotations) (download)
Sat Nov 28 14:47:14 2020 UTC (3 years, 5 months ago) by schoenebeck
File size: 101302 byte(s)
* NKSP parser: fixed memory leak; string tokens were allocated as C
  strings and never freed.

* Bumped version (2.1.1.svn66).

1 Version SVN trunk (?)
3 * general changes:
4 - Fixed compiler error in Pool.h.
5 - Require C++14 compiler support.
6 - Autoconf: Added m4/ax_cxx_compile_stdcxx.m4 macro which is used
7 for checking in configure for C++14 support (as mandatory
8 requirement) and automatically adds compiler argument if required
9 (e.g. -std=C++14).
10 - RTMath: Implemented floating point comparison methods
11 fEqual32(float,float) and fEqual64(double,double) which take the
12 expected floating point tolerances into account.
13 - Added public API C++ class "LFO", which is a cluster class encapsulating
14 all the sampler's LFO implementations to be used by 3rd party applications
15 (e.g. by Gigedit).
16 - Added int math square LFO implementation.
17 - Added int math saw LFO implementation.
18 - Added numeric complex nr sine LFO implementation.
19 - Marked class LFOTriangleDiHarmonic as deprecated
20 (will be removed in future).
21 - Added LFOAll.h which includes all LFO implementation's header files.
22 - Refactored LFO class names and their file names (PulseLFO -> LFOPulse,
23 LFOSawIntMath -> LFOSawIntMathNew, SawLFO -> LFOSawIntMathOld,
24 SineLFO -> LFOSineBuiltinFn, LFOSine -> LFOSineNumericComplexNr,
25 SquareLFO -> LFOSquarePulse and separated the latter to its own header
26 file).
27 - Renamed type LFOSigned -> LFOTriangleSigned.
28 - Renamed type LFOUnsigned -> LFOTriangleUnsigned.
29 - Using now cubic interpolation as resampling algorithm by default; and
30 linear interpolation is deprecated now.
31 - Event:Init(): memset() entire Event::Param union to zero instead of trying
32 to do that for its individual members; the latter became error prone due
33 to the union's growing complexity meanwhile.
34 - EngineChannel: Overhaul of RPN/NRPN related methods: Added new
35 *Rpn*Parameter*() methods as replacement for previous *Rpn*Controller*()
36 methods and marked the old ones as deprecated; and likewise added new
37 *Nrpn*Parameter*() methods as replacement for previous *Nrpn*Controller*()
38 methods and marked those old ones deprecated as well.
39 - Added support for MIDI CC #38 (data entry LSB): this allows higher value
40 resolution for certain (e.g. hard coded) RPN and NRPN parameters.
41 - Implemented support for compressed RPN/NRPN message sequences (that is to
42 allow processing MIDI data entry messages without always having to be
43 preceded strictly by RPN/NRPN parameter selection messages on each value
44 change).
45 - Added support for MIDI CC #96 (data increment) and MIDI CC #97 (data
46 decrement).
47 - Added support for 'null function RPN/NRPN' (which is RPN/NRPN parameter
48 MSB=127 and LSB=127) and which is disabling subsequent data
49 entry/increment/decrement CC messages.
50 - Fixed crash when editing key groups (a.k.a. exclusive groups) with an
51 instrument editor.
52 - Fixed playing noise at the end of a disk stream under certain conditions.
53 - Fixed potential crash with high pitch at end of RAM playback.
54 - Added configure checks whether C++ compiler supports so called
55 'designated initializers'.
56 - Added global backtraceAsString() function for debugging purposes
57 (POSIX systems only, no Windows implementation yet).
58 - Thread class: Added new methods pushCancelable(), popCancelable(), name(),
59 nameOfCaller() and setNameOfCaller().
60 - configure: Always check for availability of pthread_testcancel() and use
61 it if so (previously it was only used if explicitly enabled by configure
62 option).
63 - Thread class: raise compiler warning if pthread_testcancel() is not
64 available, because it can lead to deadlocks.
65 - Mutex class: Implemented optional bug detection and deadlock debugging
66 features.
67 - Ref<> class is now thread safe (as lock-free and wait-free
68 implementation).
69 - Fixed compilation error with Bison >= 3.6.0.
70 - Fixed rare crash when switching between instruments that had instrument
71 scripts loaded.
73 * Real-time instrument scripts:
74 - Added method ScriptVM::setExitResultEnabled() which allows to
75 explicitly enable the built-in exit() function to optionally accept
76 one function argument; the value of the passed exit() function
77 argument will then become available by calling
78 VMExecContext::exitResult() after script execution.
79 - 64 bit support for NKSP integer scripts variables (declare $foo).
80 - Variable names, function names and preprocessor condition names must start
81 with a regular character (a-z or A-Z); starting them with a digit or
82 underscore is not allowed.
83 - NKSP parser fix: equal comparison operator "=" and not equal comparison
84 operator "#" must only accept integer operands.
85 - NKSP language: Implemented support for standard units like Hertz, seconds,
86 Bel including support for metric unit prefixes; so one can now e.g.
87 conveniently use numbers in scripts like "5us" meaning "5 microseconds",
88 or e.g. "12kHz" meaning "12 kilo Hertz", or e.g. "-14mdB" meaning
89 "minus 14 Millidecibel", or e.g. "28c" meaning "28 cents" (for tuning).
90 - NKSP language: Introduced "final" operator "!" which is specifically
91 intended for synthesis parameter values to denote that the synthesis
92 parameter value is intended to be the "final" value for that synthesis
93 parameter that should explicitly be used by the engine and thus causing
94 the sampler engine to ignore all other modulation sources for the same
95 synthesis parameter (like e.g. LFO, EG); by simply prefixing a value,
96 variable or formula with this new "!" operator the expression is marked as
97 being "final".
98 - NKSP script editor API: Added support for detecting standard unit tokens
99 and their potential metric prefix token.
100 - NKSP language: Added support for NKSP real number literals and
101 arithmetic operations on them (e.g. "(3.9 + 2.9) / 12.3 - 42.0").
102 - NKSP language: Added support for NKSP real number (floating point) script
103 variables (declare ~foo).
104 - NKSP language: Added support for NKSP real number (floating point) array
105 script variables (declare ?foo[]).
106 - Built-in script function "message()" accepts now real number argument as
107 well.
108 - Added built-in script function "real_to_int()" and its short hand form
109 "int()" for casting from real number to integer in NKSP scripts.
110 - Added built-in script function "int_to_real()" and its short hand form
111 "real()" for casting from integer to real number in NKSP scripts.
112 - Allow built-in exit() function to potentially accept real number type
113 argument as well.
114 - Built-in script functions may have a different return type depending on
115 the arguments passed to the function.
116 - Built-in script function "abs()" optionally accepts and returns real
117 number.
118 - Built-in script functions "min()" and "max()" optionally accept real
119 number arguments and return real number as result in that case.
120 - NKSP VM API: Allow units and 'final'ness to be returned as result from
121 built-in functions (added methods VMFunction::returnUnitType() and
122 VMFunction::returnsFinal() for that purpose which must be implemented by
123 built-in function implementations).
124 - NKSP language: Allow metric unit prefixes of numeric scalar and array
125 variables to be changed freely at runtime (unlike unit types like Hz etc.
126 which are still sticky, parse-time features of variables which cannot be
127 changed at runtime for the intentional sake of determinism).
128 - NKSP language: 'final' values are prohibited for array variables for now
129 (attempt causes a parsers error).
130 - NKSP language: expressions with unit types (e.g. Hz) are prohibited for
131 conditions of runtime control structures like if(), while(), select()
132 (attempt causes a parser error).
133 - NKSP VM API: Allow built-in functions to perform their own, individual
134 parse time checks of arguments going to be passed to the function at
135 runtime (added method VMFunction::checkArgs() for that purpose).
136 - NKSP language: raise parser warning if only one operand of binary
137 operators (like logical 'or' comparison) contain a 'final' value (because
138 it would always yield in a 'final' result in such cases).
139 - NKSP language: Allow comparison (=, #, <, >, <=, >=) of values with
140 different metric unit prefixes, which will behave as expected (e.g.
141 result of expression '1000us < 2ms' is true).
142 - NKSP language: Allow adding values with different metric unit prefixes
143 (e.g. result of expression '100Hz + 5kHz' is '5100Hz').
144 - NKSP language: Allow subtracting values with different metric unit
145 prefixes (e.g. result of expression '1ms - 20us' is '980us').
146 - NKSP language: Allow multiplying with any metric unit prefixes
147 (e.g. result of expression '2k * 3ms' is '6s'), however multiplications
148 with unit types on both sides (e.g. '2s * 2s') is still prohibited since
149 we don't have any considerable practical use for a term like '4s^2'
150 (hence any attempt multiplying two unit types still causes parser error).
151 - NKSP language: Allow dividing by any metric unit prefixes and allow
152 division of same unit type on both sides (e.g. expression '8kHz / 1000Hz'
153 yields in unit free result '8'). So this is now a way to cast units away
154 e.g. for passing the result to other expressions, certain function calls
155 or variables which are not accepting any units (or that specific unit).
156 - NKSP language: integer arrays and real number arrays can now be converted
157 to strings (e.g. for dumping their content with message() calls for
158 script debugging purposes).
159 - NKSP language: expressions and variables with units are now correctly
160 casted to strings (e.g. with message() calls).
161 - NKSP language: comparing real numbers for equalness (e.g. '~foo = 3.1') or
162 unequalness (e.g. '~foo # 3.1') is now less strict and takes the expected
163 floating point tolerances into account.
164 - NKSP VM API: Added methods VMScalarNumberExpr::evalCastInt() and
165 VMScalarNumberExpr::evalCastReal().
166 - NKSP VM API: Added base class 'VMNumberArrayExpr' for classes
167 'VMIntArrayExpr' and 'VMRealArrayExpr'.
168 - NKSP VM API: replaced all unitPrefix() (parse time) methods by
169 unitFactor() (runtime) methods.
170 - Built-in function "exit()" supports now returning units and 'final'ness
171 for test cases.
172 - The following built-in functions support now units as well: "abs()",
173 "random()", "inc()", "dec()", "in_range()", "min()", "max()",
174 "real_to_int()", "int()", "int_to_real()" and "real()".
175 - Built-in functions "array_equal()", "search()" and "sort()" support now
176 real number arrays (correctly) as well.
177 - Added individual parse time checks of arguments to be passed to built-in
178 functions "random()", "inc()", "dec()", "in_range()", "min()", "max()",
179 "array_equal()" and "search()" specific for their individual purposes.
180 - NKSP VM refactoring: Renamed all methods, functions and classes matching
181 pattern *ScalarNumber* to simply *Number* (that is i.e. classes
182 VMScalarNumberExpr -> VMNumberExpr, ScalarNumberExpr -> NumberExpr,
183 ScalarNumberVariable -> NumberVariable, ScalarNumberBinaryOp ->
184 NumberBinaryOp, VMScalarNumberResultFunction -> VMNumberResultFunction,
185 method VMExpr::asScalarNumberExpr() -> VMExpr::asNumber(), function
186 isScalarNumber() -> isNumber()).
187 - NKSP VM API: Added 4 overridden methods to class VMNumberExpr:
188 evalCastInt(MetricPrefix_t), evalCastInt(MetricPrefix_t,MetricPrefix_t),
189 evalCastReal(MetricPrefix_t), evalCastReal(MetricPrefix_t,MetricPrefix_t)
190 as convenient methods for automatically converting values to expected
191 metric value basis.
192 - Built-in function "wait()" accepts now both integers and real numbers as
193 argument.
194 - NKSP VM API cleanup: Get rid of legacy method
195 VMFunction::argType(vmint iArg) which was already superseded by its new
196 replacement VMFunction::acceptsArgType(vmint iArg, ExprType_t type).
197 - NKSP parser: if wrong argument type was passed to a built-in function and
198 that built-in function accepts more than one data type for the argument,
199 then show all supported data types as parser error message.
200 - Built-in function "play_note()" accepts now real numbers and seconds as
201 unit type as well for its 3rd and 4th function arguments.
202 - The following built-in functions accept now real numbers as well for their
203 2nd function argument: "change_vol()", "change_tune()", "change_cutoff()",
204 "change_attack()", "change_decay()", "change_release()",
205 "change_sustain()", "change_cutoff_attack()", "change_cutoff_decay()",
206 "change_cutoff_sustain()", "change_cutoff_release()",
207 "change_amp_lfo_freq()", "change_cutoff_lfo_freq()",
208 "change_pitch_lfo_freq()", "change_vol_time()", "change_tune_time()",
209 "change_pan_time()", "fade_in()", "fade_out()", "change_play_pos()".
210 - Fixed built-in function "change_play_pos()" not having accepted metric
211 prefixes at all.
212 - Fixed the following built-in functions having misinterpreted values given
213 with unit type (for their 2nd argument) as if they were relative values
214 (that is as if they were passed without a unit type): "change_attack()",
215 "change_decay()", "change_release()", "change_cutoff_attack()",
216 "change_cutoff_decay()", "change_cutoff_release()".
217 - Fixed the following built-in functions having applied completely wrong
218 'final' values: "change_sustain()", "change_cutoff_sustain()" (since the
219 respective EGs being their modulation sink assume uint data type with
220 value range 0..1000 instead of 0.0..1.0.
221 - Added individual parse-time checks of function arguments for the following
222 built-in functions: "play_note()", "note_off()", "set_event_mark()",
223 "delete_event_mark()", "by_marks()", "change_cutoff()", "change_attack()",
224 "change_decay()", "change_release()", "change_cutoff_attack()",
225 "change_cutoff_decay()", "change_cutoff_release()",
226 "change_amp_lfo_freq()", "change_cutoff_lfo_freq()",
227 "change_pitch_lfo_freq()", "change_vol_time()", "change_tune_time()" and
228 "change_pan_time()".
229 - Don't abort function call if unit type was used and at the same time
230 'final' operator was omitted for the primary value argument of the
231 following built-in functions: "change_cutoff()", "change_attack()",
232 "change_decay()", "change_release()", "change_cutoff_attack()",
233 "change_cutoff_decay()", "change_cutoff_release()",
234 "change_amp_lfo_freq()", "change_cutoff_lfo_freq()",
235 "change_pitch_lfo_freq()", "change_vol_time()", "change_tune_time()",
236 "change_pan_time()", instead imply 'final'ness at runtime and raise an
237 appropriate parser warning at parse time.
238 - Added built-in real number functions "round()", "ceil()", "floor()",
239 "sqrt()", "log()", "log2()", "log10()", "exp()", "pow()", "sin()",
240 "cos()", "tan()", "asin()", "acos()", "atan()".
241 - Added built-in script real number constant "~NI_MATH_PI".
242 - Added built-in script real number constant "~NI_MATH_E".
243 - NKSP language: Allow unary '+' operator.
244 - Added built-in script functions "msb()" and "lsb()".
245 - NKSP language: Added support for RPN event handler
246 ("on rpn ... end on" in instrument scripts).
247 - NKSP language: Added support for NRPN event handler
248 ("on nrpn ... end on" in instrument scripts).
249 - Added built-in read-only variables "$RPN_ADDRESS" and "$RPN_VALUE" which
250 may be read from the new RPN/NRPN script handlers to get the (N)RPN
251 parameter that had been changed and its new value.
252 - Added built-in const variables "$NI_CB_TYPE_RPN" and "$NI_CB_TYPE_NRPN"
253 which are identifying the new (N)RPN handlers as such at script runtime.
254 - Implemented built-in script function "set_rpn()".
255 - Implemented built-in script function "set_nrpn()".
256 - Fix: built-in functions returning an array type were broken and caused a
257 crash when the built-in function's (supposed to be array) result value was
258 used by the script VM.
259 - Removed upper value constraint for 2nd argument of built-in functions
260 "change_amp_lfo_freq()", "change_cutoff_lfo_freq()",
261 "change_pitch_lfo_freq()","change_reso()", "change_cutoff()",
262 "change_amp_lfo_depth()", "change_cutoff_lfo_depth()" and
263 "change_pitch_lfo_depth()", to allow e.g. passing value 2000000 to double
264 the LFO frequency / resonance / cutoff frequency.
265 - Reduced code involved for parsing NKSP variable declarations by using 2
266 generalized grammar rules with optional components (i.e. optional value
267 assignment and optional variable qualifiers) instead of previous linear
268 combinations of grammar rules, which also fixes some edge case.
269 - NKSP language parser: track code block locations also by a) raw byte
270 position in NKSP source code string and b) length of that code block in
271 bytes (along to the already existing code block tracking by line/column);
272 and added appropriate new members to public C++ API classes like
273 'firstByte' and 'lengthBytes' to class VMSourceToken and struct CodeBlock.
274 - NKSP language: Added support for 'patch' variable qualifier (capture their
275 locations and allow replacement of their assignment expression via VM C++
276 API).
277 - Fixed incorrect EG times if standard measuring unit was passed to built-in
278 functions "change_attack()", "change_decay()" or "change_release()".
279 - Fixed intermediate function result values never having reflected any
280 standard measuring unit type.
281 - Fixed re-entrant issue with function calls which caused wrong result
282 values if the same function was called multiple times in a term
283 (specifically if metric prefixes were used).
284 - NKSP language: Just throw a warning, not an error if an array variable of
285 size zero was declared.
286 - NKSP language: Allow omitting explicit array size on array variable
287 declaration if combined with immediate value assignment
288 (e.g. declare %foo[] := ( 1, 2, 3 ) ).
289 - NKSP language: emit warning if an array variable was declared with bigger
290 array size than amount of initial values been assigned, and initialize the
291 missing array elements with zero in this case.
292 - Fixed built-in NKSP exit() function to behave as return statement.
293 - NKSP parser: Fixed crash if unary '-' operator was used on a non-number
294 data type.
295 - NKSP VM: Fixed the "release" handler being executed too often under
296 certain situations if polyphonic data was passed from "note" handler to
297 "release" handler.
298 - Fixed memory leak in NKSP parser: string tokens were allocated as C
299 strings and never freed.
301 * test cases:
302 - Fixed compiler errors in test cases.
303 - Updated README for how to compile & run test cases.
304 - Updated test case
305 MutexTest::testDoubleLockStillBlocksConcurrentThread() to latest
306 expected behaviour of the Mutex class implementation (recursive
307 mutex type).
308 - Added test cases for NKSP core language aspects and core built-in
309 functions.
310 - Fixed thread tests segfaulting on Linux.
311 - NKSP: Added real number test cases for built-in functions exit(),
312 int_to_real(), real(), real_to_int() and int(), as well as for the
313 plus, minus and negate language operators.
314 - Added massive amount of NKSP test cases for standard measuring units and
315 'final' operator usage cases.
316 - Added NKSP test cases for (floating point tolerance aware) real number
317 equalness / unequalness comparison.
318 - Added NKSP int array and real array tests for value assignment and
319 initialization of arrays.
320 - Added NKSP test cases for built-in functions "round()", "ceil()",
321 "floor()", "sqrt()", "log()", "log2()", "log10()", "exp()", "pow()",
322 "sin()", "cos()", "tan()", "asin()", "acos()", "atan()".
323 - Added NKSP test cases for unary '+' operator.
324 - Added NKSP test cases for core built-in variables.
325 - Added thorough NKSP test cases for variable declarations.
327 * GigaStudio/Gigasampler format engine:
328 - LFOTriangleIntMath and LFOTriangleIntAbsMath: Fixed FlipPhase=true
329 behaviour for start_level_mid.
330 - Changed LFO start levels: LFO1 and LFO2 both to mid, LFO3 to max. start
331 level (see discussion "GigaStudio LFO compatibility" on mailing list
332 from 2019-09-26 for details).
333 - Changed default wave form for all 3 LFOs to sine (instead of triangle;
334 see discussion "GigaStudio LFO compatibility" on mailing list from
335 2019-09-26 for details).
336 - Format extension: Added support for different LFO wave forms (currently
337 either sine [default], triangle, saw or square).
338 - Format extension: Added support for LFO phase displacement (0��..360��).
339 - Format extension: Added support for flipping LFO polarity on LFO 3
340 (in the original gig format this was only available for LFO 1 and LFO 2).
341 - Format extension: Support for additional filter types:
342 lowpass 1/2/4/6-pole, highpass 1/2/4/6-pole, bandpass 2-pole,
343 bandreject 2-pole.
344 - Fixd potential crash with polynomial interpolator and max. pitch.
345 - If minimum cutoff is defined, remap cutoff controller CC values to really
346 span the entire possible cutoff range (and avoiding a dead controller
347 zone).
348 - Implemented support for loading instrument scripts with 'patch' variables
349 (by patching these script variables with potentially overridden ones by
350 gig Instrument).
352 * SFZ format engine:
353 - Fixed support for regions with loccN/hiccN conditions on more than one
354 MIDI controller.
355 - Fixed internal opcode 'xfin_hivel' falsely being set by user opcode
356 'xfin_lovel' (patch by Christian Czezatke).
358 * Benchmarks:
359 - Fixed benchmarks/triang.cpp falsely having favoured "int math abs"
360 algorithm (since result of 2nd run was not accumulated).
361 - Added benchmark for saw wave (benchmarks/saw.cpp).
362 - Added benchmark for sine wave (benchmarks/sine.cpp).
363 - Added benchmark for square wave (benchmarks/square.cpp).
364 - Increased amount of benchmarks runs by factor 6 to achieve benchmark times
365 which are large enough on modern systems.
367 Version 2.1.1 (27 Jul 2019)
369 * Real-time instrument scripts:
370 - Fixed behavior of built-in NKSP functions change_sustain(),
371 change_cutoff_attack(), change_cutoff_decay(), change_cutoff_sustain()
372 and change_cutoff_release().
374 * general changes:
375 - Only play release trigger samples on sustain pedal up if this behaviour
376 was explicitly requested by the instrument (otherwise only on note-off).
377 - Fixed compiler warnings.
378 - Fixed compilation error when cross-compiling to Mac.
379 - FX Sends: Provide more useful error messages on routing problems
380 (see bug #169).
381 - LSCP doc: Be more clear describing the two distinct approaches
382 of using external vs. internal effects (see bug #169).
383 - "optional" class: Fixed comparison operators.
385 * Gigasampler/GigaStudio format engine:
386 - Format extension: If requested by instrument then don't play release
387 trigger sample on note-off events.
389 * SFZ format engine:
390 - Fixed memory leak when releasing samples
391 (fixes bug #307, patch by Jacek Roszkowski)
392 - Fixed potential crash when a sample is shared by more than one region
393 (fixes bug #308, patch by Jacek Roszkowski).
394 - Opcode 'sample': Added support for built-in sample '*silence'
395 (fixes bug #310, patch by Jacek Roszkowski).
397 Version 2.1.0 (25 Nov 2017)
399 * SFZ format engine:
400 - added support for <global>, <master> and #define (patch by Alby M)
401 - Removed code duplication in SFZ file loading code.
402 - Added support for sfz extension opcode 'script' which may be used to
403 load real-time instrument script file (NKSP script language).
404 - Implemented opcode set_ccN (initial patch by Giovanni Senatore).
405 - Fixed unintended volume fade-in of voices under certain conditions.
406 - sfz parser: allow missing space between header and opcode
408 * Gigasampler/GigaStudio format engine:
409 - Fixed clicks and pumping noise with Lowpass Turbo filter on very low
410 cutoff settings.
411 - Got rid of resembling an ancient GSt misbehavior which did not pitch at
412 all if an up-pitch of more than 40 semi tones was requested (I don't
413 think there is any stock gig sound that requires this behavior to
414 resemble its original sound).
415 - Added support for controlling whether the individual EGADSR stages may
416 be aborted (as LinuxSampler extension to the original GigaStudio 4
417 format).
419 * general changes:
420 - fixed printf type errors (mostly in debug messages)
421 - use unique_ptr instead of auto_ptr when building with C++11
422 - Added RTAVLTree class which is a real-time safe ordered multi-map, thus
423 allowing to sort data efficiently in real-time safe manner.
424 - RTList class: added methods for moving/inserting elements to arbitrary
425 position within a list.
426 - RTList class: added method fromPtr() for creating an Iterator object from
427 a raw element pointer.
428 - Attempt to partly fix resetting engine channels vs. resetting engine, an
429 overall cleanup of the Reset*(), ConnectAudioDevice(),
430 DisconnectAudioDevice() API methods would still be desirable though,
431 because the current situation is still inconsistent and error prone.
432 - Active voices are now internally grouped to "Note" objects, instead of
433 being directly assigned to a keyboard key. This allows more fine graded
434 processing of voices, which is i.e. required for certain instrument
435 script features.
436 - Fix: Release trigger voices were not spawned on sustain pedal up
437 (CC #64) events.
438 - Fix: Release trigger voices can now also distinguish correctly between
439 note-off and sustain pedal up events.
440 - All engines: Increased ramp speed of volume smoother and pan smoother
441 (while slow rate of crossfade smoother is preserved) to allow quick
442 volume and pan changes by instrument scripts for instance.
443 - gig/sf2/sfz: Fixed aftertouch channel pressure events being processed
444 properly and thus were ignored for certain uses.
445 - Added new C++ API method
446 VirtualMidiDevice::SendChannelPressureToSampler().
447 - windows, 32-bit: fixed potential crashes by making sure the stack in
448 sub threads is 16-byte aligned
449 - fixed numerous compiler warnings
450 - Fixed invalid (note-on) event ID being assigned to new Note objects.
451 - Revised fundamental C++ classes "Thread", "Mutex" and "Condition" which
452 fixes potential undefined behavior.
453 - Fixed Note object leak when triggering notes on keys which did not
454 have a valid sample mapped (fixes bug #252).
455 - Fixed compilation errors when compiling with CONFIG_DEVMODE enabled.
456 - linuxsampler binary fix: option --create-instruments-db ignored
457 subsequent optional argument due to glibc's implementation oddity
458 which expects a "=" sign, but no space between them.
460 * packaging changes:
461 - removed unnecessary dependency to libuuid
462 (originated by libgig's usage of it)
463 - Automake: set environment variable GCC_COLORS=auto to allow GCC to
464 auto detect whether it (sh/c)ould output its messages in color.
465 - Debian: Fixed packaging error about invalid "Source-Version"
466 substitution variable.
467 - Debian: Raised Debian compatibility level to Debian 9 "Stretch".
468 - Debian: Added build dependency to libsqlite3-dev for building
469 linuxsampler with instruments DB support.
471 * Real-time instrument scripts:
472 - Implemented scheduler for delayed MIDI events and for suspended scripts.
473 - Built-in script function "wait()": implemented support for function's
474 "duration-us" argument, thus scripts using this function are now
475 correctly resumed after the requested amount of microseconds.
476 - Built-in script function "play_note()": implemented support for
477 function's "duration-us" argument, thus notes triggered with this
478 argument are now correctly released after the requested amount of
479 microseconds.
480 - Fix: script events were not cleared when engine channel was reset,
481 potentially causing undefined behavior.
482 - Fixed crash which happened when trying to reference an undeclared
483 variable.
484 - Built-in script function "play_note()": Added support for passing
485 special value -1 for "duration-us" argument, which will cause the
486 triggered note to be released once the original note was released.
487 - Instrument script classes now exported with the liblinuxsampler C++ API.
488 - Added new API method ScriptVM::syntaxHighlighting() which provides
489 a convenient syntax highlighting backend for external instrument
490 script editor applications.
491 - Added new C++ API class "ScriptVMFactory".
492 - Extended parser issues to provide not only first line and first
493 column, but also last line and last column of issue (thus marking
494 the precise span of the issue within the source code).
495 - Reload script automatically after being modified by an instrument
496 editor.
497 - NKSP language grammar correction: allow empty event handler bodies
498 like "on note end on".
499 - Implemented built-in script function "change_vol()".
500 - Implemented built-in script function "change_tune()".
501 - Implemented built-in script function "change_pan()".
502 - Implemented built-in script function "change_cutoff()".
503 - Implemented built-in script function "change_reso()".
504 - Implemented built-in script function "event_status()".
505 - Added built-in script constants "$EVENT_STATUS_INACTIVE" and
506 "$EVENT_STATUS_NOTE_QUEUE" both for being used as flags for
507 "event_status()" function.
508 - NKSP language: Added support for bitwise operators ".or.", ".and."
509 and ".not.".
510 - NKSP language scanner: Fixed IDs matching to require at least one
511 character (i.e. when matching function names or variable names).
512 - NKSP language scanner: disabled unusued rules.
513 - Fixed behavior of play_note() and note_off() functions which must
514 be distinguished engine internally from "real" MIDI note on/off
515 events in order to avoid misbehaviors like hanging notes.
516 - Implemented built-in script variable "$KSP_TIMER".
517 - Implemented built-in script variable "$NKSP_REAL_TIMER".
518 - Implemented built-in script variable "$NKSP_PERF_TIMER".
519 - Implemented built-in script variable "$ENGINE_UPTIME".
520 - Implemented built-in script function "inc()".
521 - Implemented built-in script function "dec()".
522 - NKSP language fix: division expressions were evaluated too often.
523 - NKSP language fix: string concatenation operator was right
524 associative instead of left (to right).
525 - NKSP language correction: allow empty statements in entire language
526 (i.e. "if end if").
527 - Implemented built-in script function "stop_wait()".
528 - Implemented built-in script variable "$NI_CALLBACK_ID".
529 - Implemented built-in script variable "$NI_CALLBACK_TYPE".
530 - Implemented built-in script variable "$NKSP_IGNORE_WAIT".
531 - Added support for read-only built-in variables (respectively
532 handled by the script parser).
533 - Added built-in script constant "$NI_CB_TYPE_INIT".
534 - Added built-in script constant "$NI_CB_TYPE_NOTE".
535 - Added built-in script constant "$NI_CB_TYPE_RELEASE".
536 - Added built-in script constant "$NI_CB_TYPE_CONTROLLER".
537 - NKSP Language: Added support for user defined script functions.
538 - Implemented built-in script function "change_attack()".
539 - Implemented built-in script function "change_decay()".
540 - Implemented built-in script function "change_release()".
541 - Fixed all change_*() built-in script functions to apply their
542 synthesis parameter changes immediately in case the respective note
543 was triggered at the same time, instead of scheduling the parameter
544 change, especially because it would cause some parameter types's
545 changes either to be ramped (i.e. change_vol()) or other types even
546 to have not effect at all (i.e. change_attack()).
547 - Implemented built-in script function "sh_left()".
548 - Implemented built-in script function "sh_right()".
549 - Implemented built-in script function "min()".
550 - Implemented built-in script function "max()".
551 - NKSP Fix: "init" event handler was not always executed when a script
552 was loaded.
553 - NKSP built-in wait() script function: abort script execution if
554 a negative or zero wait time was passed as argument (since this is
555 a common indication of a bug either of the script or even of the
556 engine, which could lead to RT instability or even worse).
557 - ScriptVM: Implemented automatic suspension of RT safety
558 threatening scripts.
559 - Provide more user friendly error messages on syntax errors.
560 - Fixed NKSP parser warning "Not a statement" when assigning an
561 initializer list to an array variable.
562 - Implemented built-in script array variable "%ALL_EVENTS".
563 - Implemented built-in script function "in_range()".
564 - Implemented built-in script function "change_amp_lfo_depth()".
565 - Implemented built-in script function "change_amp_lfo_freq()".
566 - Implemented built-in script function "change_pitch_lfo_depth()".
567 - Implemented built-in script function "change_pitch_lfo_freq()".
568 - Implemented built-in script function "change_vol_time()".
569 - Implemented built-in script function "change_tune_time()".
570 - Implemented built-in script function "fade_in()".
571 - Implemented built-in script function "fade_out()".
572 - Fixed acceptance of wrong data type of parameters passed to built-in
573 script functions "change_vol()", "change_tune()", "change_pan()",
574 "change_cutoff()", "change_reso()", "change_attack()", "change_decay()",
575 "change_release()", "change_amp_lfo_depth()", "change_amp_lfo_freq()",
576 "change_pitch_lfo_depth()" and "change_pitch_lfo_freq()".
577 - Added built-in script function "get_event_par()" and implemented some
578 of its possible parameter selections.
579 - Added built-in script function "set_event_par()" and implemented some
580 of its possible parameter selections.
581 - Fixed a bunch of scheduler time related bugs.
582 - Fixed polyphonic variables not being reset to zero after usage.
583 - Built-in "ignore_event()" function: argument is now optional, like with
584 built-in function "ignore_controller()".
585 - Implemented built-in script function "change_velo()".
586 - Implemented built-in script function "change_note()".
587 - Adjusted behavior of "change_vol()" and "change_tune()" to a more
588 intuitive behavior if used in combination with "change_vol_time()" or
589 "change_tune_time()" respectively: now tuning/volume changes are only
590 assigned (without delay) immediately to a new note if the respective
591 timing function has not been called before, otherwise the volume/tuning
592 changes are automatically faded (before, only the event's time stamp was
593 relevant).
594 - Implemented built-in script function "array_equal()".
595 - Implemented built-in script function "search()".
596 - Implemented built-in script function "sort()".
597 - NKSP Fix: Never suspend "init" event handlers.
598 - Implemented built-in script function "same_region()" (currently only
599 available for gig format engine).
600 - Added built-in script constant "$NKSP_LINEAR".
601 - Added built-in script constant "$NKSP_EASE_IN_EASE_OUT".
602 - Implemented built-in script function "change_vol_curve()".
603 - Implemented built-in script function "change_tune_curve()".
604 - built-in "play_note()" function now supports a sample playback start
605 offset with argument 3, where special value -1 means to use the regular
606 sample offset as defined by the instrument file.
607 - Built-in array variable %KEY_DOWN[] is now a read-only variable.
608 - Built-in variable $EVENT_NOTE is now a read-only variable.
609 - Built-in variable $EVENT_VELOCITY is now a read-only variable.
610 - built-in "play_note()" function now accepts -2 for its fourth argument
611 (note duration) which means the life time of the note shall be sticked
612 to the requested note number of argument 1.
613 - Fix: built-in "play_note()" function now returns 0 as result value if
614 -1 was passed for its fourth argument (note duration) and the respective
615 parent note is already gone.
616 - Implemented built-in script function "change_play_pos()".
617 - NKSP language: Added support for user declared const array variables.
618 - NKSP language: Raise parser warning if array variable is accessed with
619 an index that exceeds the array's size.
620 - NKSP language fix: Unknown characters were not handled correctly.
621 - NKSP language: Added support for "synchronized .. end synchronized"
622 code blocks.
623 - Implemented built-in script function "abort()" which allows to abort
624 another script handler by passing its callback ID.
625 - Fixed potential memory access bug and potential undefined behavior of
626 "init" event handlers.
627 - Print a time stamp along to each call of built-in function "message()".
628 - ScriptVM API: Added VMParserContext::preprocessorComments() which allows
629 to retrieve all code blocks filtered out by the preprocessor.
630 - Added built-in script function "fork()".
631 - Added built-in array variable %NKSP_CALLBACK_CHILD_ID[].
632 - Added built-in variable $NKSP_CALLBACK_PARENT_ID.
633 - Fixed potential crash when accessing dynamic built-in array variables.
634 - Added built-in script function "callback_status()".
635 - Added built-in constant $CALLBACK_STATUS_TERMINATED.
636 - Added built-in constant $CALLBACK_STATUS_QUEUE.
637 - Added built-in constant $CALLBACK_STATUS_RUNNING.
638 - Removed max. value limitation of built-in functions "change_attack()",
639 "change_decay()" and "change_release()" to i.e. allow passing 2000000
640 for doubling the respective time.
641 - NKSP script editor syntax highlighting API: Fixed app termination due
642 to a lexer start condition stack underrun.
643 - NKSP preprocessor: Fixed wrong behavior on nested USE_CODE_IF() and
644 USE_CODE_IF_NOT() preprocessor statements.
645 - NKSP: Added built-in preprocessor condition NKSP_NO_MESSAGE, which
646 can be set to disable all subsequent built-in "message()" function calls
647 on preprocessor level.
648 - Implemented built-in script function "change_sustain()".
649 - NKSP script editor syntax highlighting API: catch all fatal lexer errors,
650 to avoid the editor app to crash on ill-formed text input.
651 - Added built-in script function "change_pan_time()".
652 - Added built-in script function "change_pan_curve()".
653 - Added built-in script function "change_cutoff_attack()".
654 - Added built-in script function "change_cutoff_decay()".
655 - Added built-in script function "change_cutoff_sustain()".
656 - Added built-in script function "change_cutoff_release()".
657 - Added built-in script function "change_cutoff_lfo_depth()".
658 - Added built-in script function "change_cutoff_lfo_freq()".
660 * Instruments DB:
661 - Fixed memory access bug of general DB access code which lead to
662 undefined behavior.
663 - Cleanup of instruments DB file creation and opening code.
664 - The instrument DB path of linuxsampler's --create-instruments-db argument
665 is now optional, if it is missing, then a default location is used.
666 - Added support for scanning SFZ (.sfz) files.
667 - Added support for scanning Sound Font (.sf2) files.
668 - Fixed undefined DB transaction behavior.
670 Version 2.0.0 (15 July 2015)
672 * packaging changes:
673 - fixed building with newer MinGW-w64
674 - Mac OS X: support the new dir for Core Audio SDK
675 - Mac OS X: fixed building outside source directory
676 - made sure all source files for hostplugins are included when
677 doing "make dist"
678 - removed empty directories left from the cvs to svn migration
679 - fixed compilation with gcc 4.6.1
680 - another "make dist" fix, for the LV2 plugin
681 - made --enable-pthread-testcancel default on Mac OS X
682 - minor Makefile fix for building lscpparser when build and source
683 directories are separate
684 - Mac OS X: made it possible to specify plugin installation dir to
685 configure
686 - Mac OS X: Makefile fix for the install-strip target
687 - fixed compilation with gcc 4.7
688 - fixed configure script error with old autoconf versions
689 - lsatomic.h: use gcc provided atomic functions if building with
690 gcc 4.7 and C++11
691 - modernized configure script
692 - fixed linkage error when building with
693 LDFLAGS="-Wl,--no-undefined" (#190)
694 - fixed compilation with Clang 3.2
695 - removed usage of deprecated Automake variable INCLUDES
696 - fixed building with C++11
697 - build fix: ChangeFlagRelaxed.h was missing in makefile
698 - build fix: libsndfile compiler flags were missing in some
699 makefiles
700 - fix for building with bison 3.0 (#202)
701 - Mac OS X: added temporary hack allowing to spawn gigedit as callback
702 on the process's main thread
703 - fixed build error on newer MinGW
704 - support building with older jack versions
705 - support building with spaces in vst sdk path
706 - enabled automake 'subdir-objects' option and moved external
707 source references (vst, au, asio) from makefiles to cpp files,
708 in order to get rid of warnings from automake 1.14
710 * general changes:
711 - Refactoring: moved the independent code from
712 the Gigasampler format engine to base classes
713 - command line option '--profile' is currently disabled, since the
714 respective profiling code is currently broken
715 - Introduced support for optional environment variable
716 "LINUXSAMPLER_PLUGIN_DIR" which allows to override the directory
717 where the sampler shall look for instrument editor plugins
718 (patch by Luis Garrido, slightly modified).
719 - implemented Roland GS NRPN 1ArrH which allows to set volume per note
720 - implemented Roland GS NRPN 1CrrH which allows to set pan per note
721 - implemented Roland GS NRPN 1DrrH which allows to set reverb send per
722 note (in this implementation of the sampler its simply hard routed to
723 the 1st effect send of the sampler channel, no matter what the actual
724 effect type is)
725 - implemented Roland GS NRPN 1ErrH which allows to set chorus send per
726 note (in this implementation of the sampler its simply hard routed to
727 the 2nd effect send of the sampler channel, no matter what the actual
728 effect type is)
729 - implemented support for internal LADSPA effects
730 - raised limit of program change queue from 100 to 512
731 (as suggested by Alex Stone)
732 - implemented sine LFO, pulse LFO and saw LFO
733 - experimental support for per voice equalization
734 - added command line option --exec-after-init
735 - Introduced new C++ API method:
736 EngineChannel::InstrumentFileName(int index)
737 allowing to retrieve the whole list of files used for the loaded
738 instrument on an engine channel (a.k.a. part). Some GigaStudio
739 instruments for example are splitted over several files like
740 "Foo.gig", "Foo.gx01", "Foo.gx02", ...
741 - Added new C++ API method Sampler::GetGlobalMaxVoices().
742 - Added new C++ API method Sampler::GetGlobalMaxStreams().
743 - Added new C++ API method Sampler::SetGlobalMaxVoices().
744 - Added new C++ API method Sampler::SetGlobalMaxStreams().
745 - Various "const" and "restrict" optimizations.
746 - all engines: add pan CC value to instrument pan parameter before
747 applying panning, instead of using two separate pan functions in
748 series (#182)
749 - added a lock guard class for exception safe mutex handling and
750 used it everywhere appropriate
751 - Immediately apply scale tuning changes to active voices.
752 - Exposed scale tuning to C++ API (along to the already existing standard
753 SysEx way).
754 - lsatomic.h fixes: seq_cst load and store were suboptimal for x86
755 and broken for ppc64. (Seq_cst loads and stores are actually not
756 used in LS, so the bug wasn't noticable.)
757 - lsatomic.h: added ARMv7 support
758 - Added support for multiple MIDI input ports per sampler channel (and
759 added various new C++ methods for this new feature / design change, old
760 C++ API methods for managing SamplerChannel's MIDI inputs are now marked
761 as deprecated but are still there and should provide full behavior
762 backward compatibility).
763 - AbstractEngine::GSChecksum(): don't allocate memory on the stack (was
764 unsafe and caused compilation error with clang 2.x).
765 - Bugfix: only process the latest MIDI program change event.
766 - Introducing the LSCP shell, which provides convenient control of the
767 sampler from the command line by providing LSCP aware features.
768 - VirtualMidiDevice: Added support for program change.
769 - VirtualMidiDevice: Added support for bank select (MSB & LSB).
770 - VirtualMidiDevice: Added support for pitch bend.
771 - Aftertouch: extended API to explicitly handle channel pressure and
772 polyphonic key pressure events (so far polyphonic pressure was not
773 supported at all, and channel pressure was rerouted as CC128 but not
774 used so far).
775 - Added initial support for real-time instrument scripts. The script VM
776 code is shared by all sampler engine implemementations, however only the
777 gig file format currently provides support for storing instrument scripts
778 (as LinuxSampler extension to the original GigaStudio 4 format).
780 * Gigasampler format engine:
781 - implemented the "round robin keyboard" dimension
782 - fixed round robin and random dimensions for cases when number of
783 dimension zones is not a power of two
784 - made round robin use a counter for each region instead of each
785 key
786 - bugfix: pitch LFO controller "internal+aftertouch" was broken
787 - bugfix: filter keyboard tracking was broken
788 - filter performance fix (an unnecessary copy was made of the
789 filter parameters in each sub fragment)
790 - handle special case when pan parameter in gig file has max or
791 min value
792 - Exclusive Groups: don't ever stop voices of the same note,
793 doesn't sound naturally with a drumkit
794 - fixed EG1 modulation when attack or release is zero
795 - Fixed support for 'aftertouch' attenuation controller.
796 - Fixed crash that happened with velocity split sounds under certain
797 conditions (see also previous commit on libgig).
798 - fixed behaviour of filter LFO
800 * SFZ format engine:
801 - Initial implementation (not usable yet)
802 - added support for v2 multiple stage envelope generators
803 - added a fine-tuned v1 envelope generator instead of using the
804 one from the gig engine
805 - fixed memory leak and memory handling errors
806 - added support for trigger=first, trigger=legato and sw_previous
807 - allow non-numerical key values ("C#4" for example)
808 - "key" opcode now sets pitch_keycenter too
809 - fixed error when unloading instrument with same sample used by
810 multiple regions
811 - added some opcode aliases, like loopmode for loop_mode, to be
812 more compatible
813 - added support for trigger=release and rt_decay
814 - added support for off_mode=normal
815 - added support for random, seq_position, seq_length and volume
816 - added v1 LFO opcodes to parser (no support in engine yet)
817 - added support for amp_veltrack and amp_velcurve_N
818 - fine-tuned the default velocity curve
819 - added support for transpose
820 - fixed crash when using sw_down/up
821 - improved logic for sw_lokey/hikey/up/down/last
822 - added more v1 aliases to parser
823 - reduced memory usage for sfz data
824 - RT-safeness: avoid malloc in audio thread
825 - fixed a bug that could cause voice stealing to fail
826 - optimized sample lookup
827 - improved support for exclusive groups (group, off_by and
828 off_mode)
829 - added support for controller triggered regions (on_locc/on_hicc)
830 - added support for loop_mode=one_shot
831 - fixed playback of 16 bit wav files on big endian CPUs
832 - added support for Ogg Vorbis sample files
833 - use loop markers from sample file if loop_start and loop_end are
834 not set in sfz file
835 - implemeted filters. Filter types: lowpass, bandpass, bandreject
836 and highpass. 1, 2, 4 and 6 pole filters. Opcodes: fil_type,
837 cutoff, resonance, fil_veltrack, fil_keytrack, fil_keycenter,
838 cutoff_cc, cutoff_chanaft.
839 - bugfix: zero ampeg_sustain didn't work
840 - added support for velocity effect on amplifier envelope time
841 (ampeg_vel2attack, ampeg_vel2decay, ampeg_vel2sustain and
842 ampeg_vel2release)
843 - added support for EG hold (ampeg_hold)
844 - added support for sample offset (offset)
845 - use the newly introduced signal units model
846 - implemented opcodes lfoN_delay,
847 lfoN_freq, lfoN_pan, lfoN_cutoff, lfoN_resonance
848 - implemented opcodes ampeg_delay, ampeg_vel2delay
849 pitcheg_delay, pitcheg_start, pitcheg_attack, pitcheg_hold,
850 pitcheg_decay, pitcheg_sustain, pitcheg_release, pitcheg_vel2delay,
851 pitcheg_vel2attack, pitcheg_vel2hold, pitcheg_vel2decay,
852 pitcheg_vel2sustain, pitcheg_vel2release, pitcheg_depth
853 - implemented opcodes pitchlfo_delay, pitchlfo_freq,
854 pitchlfo_depth, fillfo_delay, fillfo_freq, fillfo_depth,
855 amplfo_delay, amplfo_freq, amplfo_depth
856 - implemented opcodes fileg_delay, fileg_start, fileg_attack,
857 fileg_hold, fileg_decay, fileg_sustain, fileg_release, fileg_vel2delay,
858 fileg_vel2attack, fileg_vel2hold, fileg_vel2decay, fileg_vel2sustain,
859 fileg_vel2release
860 - implemented opcode lfoN_wave
861 - implemented opcode pitchlfo_depthccN
862 - implemented opcodes lfoN_phase, lfoN_phase_onccX,
863 lfoN_pitch, lfoN_pitch_onccX
864 - implemented opcodes pitchlfo_fade,
865 fillfo_fade, amplfo_fade, lfoN_fade, lfoN_fade_onccX
866 - implemented opcodes fillfo_freqccN,
867 pitchlfo_freqccN, amplfo_freqccN, lfoN_freq_onccX
868 - implemented opcodes ampeg_delayccN, ampeg_startccN,
869 ampeg_attackccN, ampeg_holdccN, ampeg_decayccN, ampeg_sustainccN,
870 ampeg_releaseccN, egN_timeX_onccY, egN_levelX_onccY
871 - lfoN_* and egN_* opcodes defined in group sections
872 are now taken into account
873 - implemented curves
874 - implemented opcodes volume_onccN, volume_curveccN
875 - implemented opcode volume_smoothccN
876 - implemented opcodes fillfo_depthccN, amplfo_depthccN,
877 lfoN_volume, lfoN_volume_onccX, lfoN_volume_smoothccX,
878 lfoN_freq_smoothccX, lfoN_pitch_smoothccX, lfoN_pan_onccX,
879 lfoN_pan_smoothccX, lfoN_cutoff_onccX, lfoN_cutoff_smoothccX,
880 lfoN_resonance_onccX, lfoN_resonance_smoothccX, lfoN_delay_onccX
881 - implemented opcode end
882 - implemented opcodes egN_amplitude_onccX,
883 egN_volume, egN_volume_onccX, egN_cutoff, egN_cutoff_onccX,
884 egN_pitch, egN_pitch_onccX, egN_resonance, egN_resonance_onccX
885 - implemented opcodes xfin_lokey, xfin_hikey,
886 xfout_lokey, xfout_hikey, xf_keycurve, xfin_lovel, xfin_hivel,
887 xfout_lovel, xfout_hivel, xf_velcurve, xfin_loccN, xfin_hiccN,
888 xfout_loccN, xfout_hiccN, xf_cccurve
889 - implemented opcodes pan_onccN, pan_smoothccN,
890 pan_curveccN, egN_pan, egN_pan_curve, egN_pan_onccX,
891 egN_pan_curveccX, pitch_veltrack
892 - when failed to parse a sfz file
893 print the line number on which the error occurs
894 - use common pool of CC objects to minimize RAM usage
895 - implemented opcodes amplfo_delay_onccN,
896 amplfo_fade_onccN, fillfo_delay_onccN, fillfo_fade_onccN,
897 pitchlfo_delay_onccN, pitchlfo_fade_onccN
898 - implemented opcodes fileg_delay_onccN,
899 fileg_start_onccN, fileg_attack_onccN, fileg_hold_onccN,
900 fileg_decay_onccN, fileg_sustain_onccN, fileg_release_onccN,
901 fileg_depth_onccN, pitcheg_delay_onccN, pitcheg_start_onccN,
902 pitcheg_attack_onccN, pitcheg_hold_onccN, pitcheg_decay_onccN,
903 pitcheg_sustain_onccN, pitcheg_release_onccN, pitcheg_depth_onccN
904 - implemented automatic aliasing *ccN <-> *_onccN
905 - *lfo_freqccN wasn't working when the respective *lfo_freq
906 was not set or was set to zero
907 - lfoN_freq_onccX wasn't working when lfoN_freq
908 was not set or was set to zero
909 - implemented opcodes resonance_onccN, resonance_smoothccN,
910 resonance_curveccN, cutoff_smoothccN, cutoff_curveccN
911 - implemented opcodes fillfo_depthchanaft,
912 fillfo_freqchanaft, amplfo_depthchanaft, amplfo_freqchanaft,
913 pitchlfo_depthchanaft, pitchlfo_freqchanaft
914 - implemented opcodes pitch_onccN,
915 pitch_curveccN, pitch_smoothccN, pitch_stepccN
916 - implemented opcodes volume_stepccN, pan_stepccN,
917 cutoff_stepccN, resonance_stepccN, lfoN_freq_stepccX,
918 lfoN_volume_stepccX, lfoN_pitch_stepccX, lfoN_pan_stepccX,
919 lfoN_cutoff_stepccX, lfoN_resonance_stepccX
920 - implemented opcodes eq1_freq, eq2_freq, eq3_freq,
921 eq1_freqccN, eq2_freqccN, eq3_freqccN, eq1_bw, eq2_bw, eq3_bw,
922 eq1_bwccN, eq2_bwccN, eq3_bwccN, eq1_gain, eq2_gain, eq3_gain,
923 eq1_gainccN, eq2_gainccN, eq3_gainccN
924 - implemented opcodes delay, delay_onccN, delay_random,
925 delay_samples, delay_samples_onccN
926 - implemented opcodes egN_eq1gain, egN_eq2gain, egN_eq3gain,
927 egN_eq1gain_onccX, egN_eq2gain_onccX, egN_eq3gain_onccX, egN_eq1freq,
928 egN_eq2freq, egN_eq3freq, egN_eq1freq_onccX, egN_eq2freq_onccX,
929 egN_eq3freq_onccX, egN_eq1bw, egN_eq2bw, egN_eq3bw, egN_eq1bw_onccX,
930 egN_eq2bw_onccX, egN_eq3bw_onccX, lfoN_eq1gain, lfoN_eq2gain,
931 lfoN_eq3gain, lfoN_eq1gain_onccX, lfoN_eq2gain_onccX, lfoN_eq3gain_onccX,
932 lfoN_eq1gain_smoothccX, lfoN_eq2gain_smoothccX, lfoN_eq3gain_smoothccX,
933 lfoN_eq1gain_stepccX, lfoN_eq2gain_stepccX, lfoN_eq3gain_stepccX,
934 lfoN_eq1freq, lfoN_eq2freq, lfoN_eq3freq, lfoN_eq1freq_onccX,
935 lfoN_eq2freq_onccX, lfoN_eq3freq_onccX, lfoN_eq1freq_smoothccX,
936 lfoN_eq2freq_smoothccX, lfoN_eq3freq_smoothccX, lfoN_eq1freq_stepccX,
937 lfoN_eq2freq_stepccX, lfoN_eq3freq_stepccX, lfoN_eq1bw, lfoN_eq2bw,
938 lfoN_eq3bw, lfoN_eq1bw_onccX, lfoN_eq2bw_onccX, lfoN_eq3bw_onccX,
939 lfoN_eq1bw_smoothccX, lfoN_eq2bw_smoothccX, lfoN_eq3bw_smoothccX,
940 lfoN_eq1bw_stepccX, lfoN_eq2bw_stepccX, lfoN_eq3bw_stepccX
941 - implemented opcodes eq1_vel2freq, eq2_vel2freq,
942 eq3_vel2freq, eq1_vel2gain, eq2_vel2gain, eq3_vel2gain
943 - sfz parser: allow double spaces in sample filenames
944 - sfz parser: allow absolute paths for sample filenames
945 - use linear decay and release for filter and pitch EG
946 - bugfix: only the first amp_veltrack definition in a file was
947 used
948 - bugfix: looping was disabled if loop_start was set to 0
949 - allow regions with end=-1 to turn off other regions using the
950 group and off_by opcodes (#168)
951 - made end=0 play the whole sample
952 - fixed support for lochan and hichan opcodes (#155)
953 - fixed crash when using lochan/hichan opcodes (#187)
954 - sfz parser: allow -200 to 200 for pan_oncc opcode (#182)
955 - added FLAC support (#191)
956 - sfz parser bugfix: lines starting with whitespace were ignored
957 - added amplitude opcode
958 - added support for "#include" instruction
959 (modified patch which was originally posted by Sergey on LS mailing list)
960 - bugfix: generation of velocity curves etc should not be done
961 after each "#include", only after the main file is parsed
962 - bugfix: line numbers in error messages were wrong after "#include"
963 - added support for float and 32 bit sample files
965 * SoundFont format engine:
966 - Initial implementation (not usable yet)
967 - RT-safeness: avoid malloc in audio thread
968 - fixed a bug that could cause voice stealing to fail
969 - fine-tuned amplitude EG (by switching from gig to sfz EG)
970 - initial implementation of Vibrato LFO and Modulation LFO
971 - initial implementation of cutoff filter
972 - use linear decay and release for filter and pitch EG
974 * Host plugins (VST, AU, LV2, DSSI):
975 - AU bugfix: failed to destroy its audio/MIDI devices
976 - Listen to all interfaces on Mac OS X (INADDR_ANY)
977 - VST bugfix: If the host called resume() before and after
978 changing sample rate or block size, the number of channels was
979 incorrectly set to two. This caused silence in Cubase 5.
980 - save engine type (gig, sfz or sf2) in plugin state
981 - VST: when opening Fantasia, look for both 32 and 64 bit Java on
982 64 bit Windows
983 - AU: changed number of output channels from one stereo to 16
984 stereo
985 - VST: made it possible to build the VST plugin for Mac
986 - AU: link AU plugin dynamically if --disable-shared isn't
987 specified
988 - LV2 "state" extension support (patch by David Robillard)
989 - VST bugfix: instrument loading hang and crashed the host when
990 the plugin was loaded a second time (#174)
991 - plugin bugfix: instrument loading hang when the plugin was
992 loaded a second time (this time it's for Linux and Mac, previous
993 similar fix was for Windows)
994 - thread safety fixes for the instrument loading thread
995 - LV2: use the new lv2 package if present
996 - VST: try to open Fantasia automatically on Linux and Mac too (on
997 Linux, the Fantasia jar should be placed in <prefix>/share/java)
998 - VST: fixed crashes on Linux Ardour and EnergyXT
999 - DSSI bugfix: it wasn't possible to change engine type. The MIDI
1000 port and audio channel routing for DSSI plugins are now visible.
1001 - LV2: use urid and atom extensions instead of deprecated uri-map
1002 and event
1003 - LV2: lv2 package 1.0 is now required to build the LV2 plugin
1004 - LV2: changed number of output channels to 16 stereo, just like
1005 the VST and AU plugins
1006 - LV2: fixed save/restore of SFZ state (patch by David Robillard)
1007 - LV2: made LV2 plugin buildable on Windows and Mac
1008 - VST: implemented retrieval and switching of programs using the
1009 sampler's internal MIDI instrument mapping system
1011 * Instrument editor interface:
1012 - Changed instrument editor plugin interface, providing additional
1013 informations like the EngineChannel for which the instrument editor was
1014 spawned for. This allows the instrument editors to interact more actively
1015 with the sampler.
1017 * MIDI driver:
1018 - ALSA MIDI driver supports now "NAME" device parameter, for overriding
1019 the ALSA sequencer client name
1020 - removed limit of maximum amount of MIDI ports per MIDI device, since
1021 there is no reason for this limit
1022 - MME: fixed memory handling bug found with cppcheck
1023 - MME: removed compiler warning
1024 - CoreMIDI: implemented driver specific port parameter "CORE_MIDI_BINDINGS",
1025 which allows to retrieve the list of CoreMIDI clients / ports and to connect
1026 to them a la JACK, via the usual sampler APIs
1027 - CoreMIDI: added driver specific port parameter "AUTO_BIND", if enabled
1028 the driver will automatically connect to other CoreMIDI clients' ports
1029 (e.g. external MIDI devices being attached to the Mac)
1030 - added support for MIDI note on velocity filter
1031 - CoreMIDI: fixed memory deallocation error
1032 - Fixed variable underflow in VirtualMidiDevice, which caused graphical
1033 virtual keyboards in frontends / instrument editors being stuck.
1034 - Bugfix in VirtualMidiDevice: process note on with velocity 0 as note off.
1035 - Implemented missing handling of MIDI "running status".
1036 - CoreMIDI fix: a MIDIPacket can contain more than one event per packet.
1037 - MME bugfix: driver wasn't closed properly
1038 - CoreMIDI: automatically connect to all input sources by default (driver
1039 parameter "AUTO_BIND").
1040 - CoreMIDI: fixed auto bind feature to CoreMIDI ports that go online
1041 - CoreMIDI: fixed minor memory leak
1043 * audio driver:
1044 - ASIO driver fixes for newer gcc versions (fix from PortAudio)
1045 - JACK audio: react on sample rate changes.
1046 - JACK audio: react on buffer size changes.
1047 - JACK audio: jack_port_get_buffer() was cached and called outside
1048 RT context.
1049 - ASIO driver: removed compiler warnings
1050 - CoreAudio: fixed minor error handling bug
1051 - ASIO driver: be more verbose when no ASIO card could be found (fixes #203)
1052 - JACK audio: return the JACK server's current sample rate as default value
1053 for audio device parameter "SAMPLERATE" (fixes #166).
1055 * LSCP server:
1056 - added support for sending MIDI CC messages via LSCP command
1057 "SEND CHANNEL MIDI_DATA CC <sampler-chan> <ctrl> <val>"
1058 - added LSCP command "GET AVAILABLE_EFFECTS"
1059 - added LSCP command "LIST AVAILABLE_EFFECTS"
1060 - added LSCP command "GET EFFECT INFO <effect-index>"
1061 - added LSCP command "CREATE EFFECT_INSTANCE <effect-index>"
1062 - added LSCP command
1063 "CREATE EFFECT_INSTANCE <effect-system> <module> <effect-name>"
1064 - added LSCP command "DESTROY EFFECT_INSTANCE <effect-instance>"
1065 - added LSCP command "GET EFFECT_INSTANCES"
1066 - added LSCP command "LIST EFFECT_INSTANCES"
1067 - added LSCP command "GET EFFECT_INSTANCE INFO <effect-instance>"
1068 - added LSCP command
1069 "GET EFFECT_INSTANCE_INPUT_CONTROL INFO <effect-instance> <input-control>"
1071 <effect-instance> <input-control> <value>"
1072 - added LSCP command "GET SEND_EFFECT_CHAINS <audio-device>"
1073 - added LSCP command "LIST SEND_EFFECT_CHAINS <audio-device>"
1074 - added LSCP command "ADD SEND_EFFECT_CHAIN <audio-device>"
1075 - added LSCP command
1076 "REMOVE SEND_EFFECT_CHAIN <audio-device> <effect-chain>"
1077 - added LSCP command
1078 "GET SEND_EFFECT_CHAIN INFO <audio-device> <effect-chain>"
1079 - added LSCP command "APPEND SEND_EFFECT_CHAIN EFFECT <audio-device>
1080 <effect-chain> <effect-instance>"
1081 - added LSCP command "INSERT SEND_EFFECT_CHAIN EFFECT <audio-device>
1082 <effect-chain> <effect-chain-pos> <effect-instance>"
1083 - added LSCP command "REMOVE SEND_EFFECT_CHAIN EFFECT <audio-device>
1084 <effect-chain> <chain-pos>"
1085 - added LSCP command "SET FX_SEND EFFECT <sampler_channel>
1086 <fx_send_id> <effect_chain> <chain_pos>"
1087 - added LSCP command "REMOVE FX_SEND EFFECT <sampler_channel> <fx_send_id>"
1091 - provide comprehensive error messages on LSCP syntax errors
1092 (suggesting expected next non-terminal symbols)
1093 - Fixed client connection not being closed after network errors.
1095 * LSCP shell:
1096 - Added support for auto correction of obvious and trivial syntax mistakes.
1097 - Added support for auto completion by tab key.
1098 - Show currently available auto completion while typing.
1099 - Added support for browsing command history with up / down keys.
1100 - Show all possible next symbols immediately right to the current command
1101 line while typing (no double tab required for this feature, as it would
1102 be the case in other shells).
1103 - Added support for moving cursor left/right with arrow keys.
1104 - Added support for built-in LSCP reference documentation, which will
1105 automatically show the relevant LSCP reference section on screen as soon
1106 as one specific LSCP command was detected while typing on the command
1107 line.
1109 * Real-time instrument scripts:
1110 - Implemented built-in script array variable %CC.
1111 - Implemented built-in script int variable $CC_NUM.
1112 - Implemented built-in script int variable $EVENT_NOTE.
1113 - Implemented built-in script int variable $EVENT_VELOCITY.
1114 - Implemented built-in script constant variable $VCC_MONO_AT.
1115 - Implemented built-in script constant variable $VCC_PITCH_BEND.
1116 - Implemented execution of script event handler "init".
1117 - Implemented execution of script event handler "controller".
1118 - Implemented execution of script event handler "note".
1119 - Implemented execution of script event handler "release".
1120 - Implemented built-in script function "play_note()" (only two of the
1121 max. four function arguments are currently implemented yet though).
1122 - Implemented built-in script int variable $EVENT_ID.
1123 - Implemented built-in script function "ignore_event()"
1124 - Implemented built-in script function "ignore_controller()" (may have one
1125 or no argument).
1126 - Implemented built-in script function "set_controller()".
1127 - Added extended script VM for the Gigasampler/GigaStudio format sampler
1128 engine, which extends the general instrument script VM with Giga format
1129 specific variables and functions.
1130 - Giga format scripts: added built-in script int constant variables
1141 - Giga format scripts: Implemented built-in script function
1142 "gig_set_dim_zone(event_id, dimension, zone)".
1143 - Implemented built-in script int array variable %KEY_DOWN.
1144 - Implemented built-in script function "abs()".
1145 - Implemented built-in script function "random()".
1146 - Implemented built-in script function "num_elements()".
1147 - Implemented built-in script function "note_off()".
1148 - Implemented built-in script function "set_event_mark()".
1149 - Implemented built-in script function "delete_event_mark()".
1150 - Implemented built-in script function "by_marks()".
1151 - Added built-in script int const variables $MARK_1 to $MARK_28.
1152 - Built-in script functions "ignore_event()", "note_off()" and
1153 "gig_set_dim_zone()" now also accept an array of event IDs as argument
1154 (i.e. return value of new script function "by_marks()").
1155 - Pass/preserve polyphonic variable data from respective "note" event
1156 handler to "release" event handler.
1157 - Fixed crash when using built-in script function "by_marks()".
1159 * Bug fixes:
1160 - Fixed crash which may occur when MIDI key + transpose is out of range
1161 - minor valgrind fixes
1162 - fixed crash which occurred when changing an already deployed sampler
1163 channel to a different engine type
1164 - fixed crash when deleting a sampler channel or changing engine
1165 type while an instrument load was in progress
1166 - bugfix: playing a note while changing the instrument could cause
1167 a crash, or give "Handing back unknown region" error messages
1168 - bugfix: calling SET VOICES (which Fantasia does on start and
1169 refresh) could cause instruments to be unloaded
1170 - Mac OS X: fixed name collision of enum in EffectControl and
1171 wrong name of destructor in AudioOutputDeviceCoreAudio.cpp
1172 - Mac OS X: fixed hanging threads
1173 - Windows: disabled the previous thread fix on non-Mac systems, as
1174 it caused hanging threads on Windows
1175 - Fixed possible crashes due to corrupted MIDI/audio device list
1176 after MIDI/audio device creation failure
1177 - When creating MIDI instrument map entries with "PERSISTENT" type, the
1178 instruments were uselessly precached with zero samples, however it still
1179 took the full preloading time and on 1st program change the respective
1180 instrument was completely reloaded again.
1181 - fixed handling of rapid bank select and program change messages
1182 sent to the same sampler channel (patch from the Open Octave
1183 project, slightly adjusted)
1184 - fixed crash when trying to create an effect instance with controls
1185 which min and/or max values depend on the sample rate
1186 - fixed bug #162
1187 - bugfix: LADSPA_PATH was not evaluated correctly when containing
1188 multiple paths (#165)
1189 - thread safety fixes for the instrument loading thread
1190 - bugfix: instrument loading crashed for sfz and sf2 in Ardour
1191 (#176)
1192 - more thread safety fixes for the instrument loading thread
1193 - sfz/sf2 engine: fixed crash when using small audio fragment size
1194 - Mac OS X: fixed crash when unloading plugin on 10.7 and later
1195 - Mac OS X: fixed process hang when unloading 32-bit plugin (bug
1196 introduced in previous fix)
1197 - fixed crash when a channel received a program change while
1198 playing a note in a key group
1199 - fixed erroneous error message piping in VoiceBase.h
1200 (labelled "Disk stream not available in time")
1201 - Update effects on sample rate & period size changes (to avoid
1202 crashes and noise on such transitions).
1203 - ignore missing LADSPA paths without ignoring valid LADSPA paths
1204 (fixes #208)
1205 - Fixed bug in LSCP grammar definition which caused a statement like
1206 "GET SERVER INFOasdf\n" to be accepted as valid statement (was so far
1207 practically irrelevant, however it caused problems with the new LSCP
1208 shell's auto completion feature).
1209 - Fixed MIDI program change messages being ignored if quickly executed
1210 after each other (fixes #231).
1212 Version 1.0.0 (31 July 2009)
1214 * packaging changes:
1215 - autoconf bugfix: the PKG_CONFIG variable wasn't initialized properly,
1216 causing e.g. the libgig test to fail when
1217 "./configure --disable-jack-driver" was used
1218 (patch by Alexis Ballier)
1219 - fixed compilation with gcc 4.3
1220 - fixes for building on OS X (thanks to Ebrahim Mayat for testing)
1221 - fixed configure so it detects x86_64 (#107)
1222 - fixes for building with newer MinGW versions
1223 - fix for building with bison 2.4 (#111)
1224 - fixed building with libgig installed in a non-standard directory
1225 - minor fix in configure for mmsystem.h detection on MinGW
1226 - Windows: look for editor plugins and Fantasia using base
1227 directory of liblinuxsampler dll (look in the same directory and one
1228 directory above)
1229 - configure script fix: removed unconditional use of SSE
1230 - fixed building with sqlite installed in a non-standard directory
1231 - when cross-compiling, don't try to create instruments.db
1232 - fix for new mingw-w64 version, which has usleep
1234 * general changes:
1235 - bugfix: on some POSIX systems instrument editor plugins refused to
1236 load as we used a non-portable Linux specific struct field
1237 (fixes bug #70, patch by Ronald Baljeu)
1238 - fixed endless loop which occured when loading an instrument editor
1239 plugin DLL on Windows systems failed
1240 - fixed memory leaks that occurred when liblinuxsampler was unloaded
1241 - optimized the SynchronizedConfig class so it doesn't wait
1242 unnecessarily long after an update
1243 - added support for notifying instrument editors on note-on / note-off
1244 events (e.g. to highlight the pressed keys on the virtual keyboard
1245 of gigedit)
1246 - added support for triggering notes by instrument editors (see above)
1247 - be verbose on DLL load errors (on Linux)
1248 - fixed di-harmonic triangle LFO implementation (this LFO implementation
1249 is only used by very few systems, most pick the int math
1250 implementation, which is usually faster)
1251 - fixes for audio drivers with varying buffer sizes
1252 - experimental support for running LinuxSampler as a DSSI, LV2 and
1253 VST plugin
1254 - notification events for stream/voice count statistics are now sent
1255 only when there are actual changes
1256 - added memory ordering constraints to improve stability on
1257 multi-core and multi-cpu systems
1258 - maximum voices and disk streams can now be altered at runtime
1259 - fixed CPU feature detection on x86_64 (maybe fixes #108)
1260 - automatic stacktrace mechanism is now turned off by default and can
1261 be switched on by command line option "--stacktrace" (the automatic
1262 stacktrace mechanism seems to be broken on most systems at the moment)
1263 - C++ API method InstrumentManager::LaunchInstrumentEditor() now returns
1264 a pointer to the launched InstrumentEditor object
1265 - added optional 3rd party user data parameter for following
1266 liblinuxsampler C++ API methods: InstrumentEditor::Main(),
1267 InstrumentEditor::Launch(),
1268 InstrumentManager::LaunchInstrumentEditor()
1269 - theoretical fix: made SynchronizedConfig follow C++0x memory
1270 model more strictly
1271 - fixes for using large audio device buffers
1272 - Windows: add the installation directory to the DLL search path
1273 when loading an editor plugin (solves problems with VST and
1274 gigedit on systems with other GTK versions installed)
1275 - updated linuxsampler man page
1277 * audio driver:
1278 - removed the nonsense audio channel constraint (which was hard coded to
1279 max. 100 audio channels) for most audio drivers
1280 - JACK audio driver did not offer a device parameter "SAMPLERATE" as
1281 opposed to the LSCP specs
1282 - bugfix: the SAMPLERATE parameter of some drivers (e.g. JACK)
1283 reflected the wrong value
1284 - fixed a memory management bug in ASIO driver
1285 - Makefile fix: JACK_CFLAGS wasn't used
1286 - JACK: use jack_client_open instead of the deprecated
1287 jack_client_new
1288 - added (experimental) CoreAudio driver
1289 - applied old fixes to the ASIO driver that were included in the
1290 previous binary release but accidentally never committed to CVS
1291 (fixes #117)
1292 - fixes for ASIO on mingw-w64 (iasio wrapper is not needed on
1293 win64)
1294 - VST: added support for sample rate and buffer size changes
1295 - VST: close editor (Fantasia) when the VST is removed
1296 - VST: avoid opening Fantasia more than once for each VST instance
1297 - VST: export main function as "main" on Linux too (fix for
1298 energyXT)
1299 - VST: changed number of output channels from one stereo to 16
1300 stereo
1301 - added channel routing, fxsends and midi maps to the settings
1302 stored in the plugin state
1303 - performance optimization of AudioChannel::MixTo() and
1304 AudioChannel::CopyTo() methods using GCC vector exensions
1305 (if available)
1306 - ASIO fixes: avoid initializing the device twice, avoid throwing
1307 exception when getting parameters from a disconnected device
1309 * MIDI driver:
1310 - added JACK MIDI driver
1311 - dispatch bank select as ordinary CC as well, the user might seriously
1312 want to (mis)use it for some purpose ("fixed" in all current MIDI
1313 input drivers: ALSA, CoreMIDI, JACK, MidiShare, MME)
1314 - bugfix: pitch bend wasn't working with JackMidi, VST, LV2, MME,
1315 CoreMidi or AU
1316 - fixed mingw-w64 compilation error in MME driver
1317 - made program change handling in MIDI thread real-time safe by
1318 moving the logic to a non-RT thread
1319 - fixed minor memory leak in ALSA driver
1321 * instruments database:
1322 - avoid time consuming samples scanning when adding instruments
1323 to the instruments database
1324 - added support for handling lost files in the instruments database
1325 - Implemented option for adding instruments in separate directories
1326 in the instruments database
1327 (patch by Chris Cherrett & Andrew Williams, a bit adjusted)
1328 - work-around for missing fnmatch function on Windows to make
1329 instrument database compilable
1330 - added instrument database support on Windows
1331 works with both standalone mode and VST plugin.
1332 the instruments DB file is located in
1333 %USERPROFILE%\.linuxsampler\instruments.db which allows different
1334 databases for each windows user
1335 if no DB is present it automatically creates the .linuxsampler subdir
1336 and then creates an empty DB
1337 - fixed recursive import, which was broken on Windows
1339 * Gigasampler format engine:
1340 - fixed a memory leak that could happen when a channel was deleted
1341 while notes were playing
1342 - made it possible to load an instrument even if the audio thread
1343 isn't running
1344 - added partial support for the "Controller Triggered" MIDI rule,
1345 enough for piano gigs with pedal noise samples
1346 - minor fix: only mark FX sends as being modified if really the
1347 respective FX send MIDI controller was used
1348 - added support for GM global device volume SysEx message
1349 (can be selected at compile time to either apply globally to the
1350 whole sampler [default] or only to the sampler channels that are
1351 connected to the respective MIDI input port the SysEx message arrived
1352 on)
1353 - bugfix: notes triggered at position 0 in the audio buffer were
1354 sometimes wrongly killed in the same buffer, causing no sound to
1355 be played
1356 - added support for chromatic / drumkit mode Roland GS Sysex message
1357 which is usally used to switch a part between chromatic sounds and
1358 drumkit sounds (as opposed to their common default setting of MIDI
1359 part 10 being a drumkit part and all other ones chromatic parts), in
1360 LS however one can switch between the first 16 MIDI instrument maps
1361 defined for the sampler
1362 - made it possible to create multiple sample channels even if the
1363 audio thread isn't running
1364 - fixed crash when removing channel with active voices (#116)
1365 - bugfix: on sample reference changes (instrument editor), only
1366 un-cache the respective sample if it's really not used by any
1367 sampler engine anymore
1368 - re-cache samples in case they were changed by an instrument editor,
1369 e.g. when a sample was added while playing (#82)
1370 - fixed hanging notes which occured when note-off event had the exact
1371 same time stamp as the voice's note-on event and occured both in the
1372 same audio fragment cycle (fixes bug #112)
1373 - added support for the "fine tune" and "pitch bend range"
1374 instrument-level gig parameters
1375 - fixed minor artifacts in pitch bend handling
1376 - added support for GS Reset SysEx message
1377 - allow gig files to use unlimited downward pitch shifting
1378 - added a limit check for upward pitch shifting
1379 - bugfix: sometimes, when playing a note twice fast, the second
1380 note was silent
1381 - fixed crash happening when a pitch bend event arrived at the
1382 same time a new instrument was loading
1384 * LSCP server:
1385 - added new LSCP event "CHANNEL_MIDI" which can be used by frontends to
1386 react on MIDI data arriving on certain sampler channels (so far only
1387 Note-On and Note-Off events are sent via this LSCP event)
1388 - added new LSCP event "DEVICE_MIDI" which can be used by frontends to
1389 react on MIDI data arriving on certain MIDI input devices (so far only
1390 Note-On and Note-Off events are sent via this LSCP event)
1391 - added new LSCP commands: FIND LOST DB_INSTRUMENT_FILES and
1393 - added new LSCP command: SEND CHANNEL MIDI_DATA which can be used by
1394 frontends to send MIDI messages to specific sampler channel
1395 - added two additional fields to GET FILE INSTRUMENT INFO command -
1397 - bugfix: the bank number provided by MIDI_INSTRUMENT_INFO notifications
1398 was incorrect
1399 - Optimized the retrieval of the MIDI instrument mappings
1400 - added new LSCP command "SET VOICES" to globally alter the maximum
1401 amount of voices, added new LSCP event "GLOBAL_INFO:VOICES" which will
1402 be triggered respectively
1403 - added new LSCP command "SET STREAMS" to globally alter the maximum
1404 amount of disk streams, added new LSCP event "GLOBAL_INFO:STREAMS"
1405 which will be triggered respectively
1406 - bugfix: retry if "select" returns EINTR (this fixes a crash when
1407 a gigedit file dialog is opened)
1408 - close all connections when LSCPServer is deleted
1409 - hard close of all sockets on exit
1410 - bugfix: SET CHANNEL MIDI_INPUT_TYPE didn't work with the MME
1411 driver
1413 * Bug fixes:
1414 - fixed a crash which occurs when removing a sampler channel waiting
1415 to start instrument loading after another channel
1416 - fixed a crash which occurs when removing a sampler channel with
1417 instrument loading in progress (bug #113)
1418 - fixed termination caused by uncaught exception when adding MIDI
1419 instrument with PERSISTENT load mode
1420 - fixed possible iterator invalidations when resetting the sampler
1421 - fixed memory leaks when issuing the following LSCP commands:
1426 - fixed possible compilation error when sqlite is not present
1427 - fixed orphaned pointers when setting maximum voices limit (bug #118)
1428 - fixed crash when changing the audio output device of a sampler
1429 channel with loaded instrument and start playing notes
1430 - fixed endless loop in Engine::SuspendAll() (bug #120)
1431 - fixed a low-level atomic load function that was broken on 64-bit
1432 PowerPC, which probably could cause crashes on that platform
1433 - fixed a memory management error which could cause a crash when a
1434 plugin was unloaded
1435 - bugfix: two private structs had the same name, which could cause
1436 problems if the linker chose the wrong constructor
1437 - fixed low-level ConditionServer usage bug that caused lockups on
1438 Windows
1441 Version 0.5.1 (6 December 2007)
1443 * packaging changes:
1444 - added autoconf checks for pthread library
1445 - added autoconf check for pthread bug found on certain NPTL-enabled
1446 glibc versions (see Gentoo bug report #194076)
1447 - added autoconf checks for MS Windows
1449 * general changes:
1450 - bugfix: the thread used by an editor plugin didn't die when the
1451 editor closed
1452 - bugfix: Ringbuffer.h: fill_write_space_with_null() did not zero
1453 out all the space. operator--() did not apply size_mask after
1454 decrementing the read_ptr. DEFAULT_WRAP_ELEMENTS set to 0 in
1455 order to avoid problems with the _NonVolatileReader functions.
1456 - bugfix: Stream.h: added a missing divide by BytesPerSample in
1457 GetWriteSpace(). Since this function is currently only used in
1458 the stream's qsort() compare function, it didn't trigger any
1459 bugs.
1460 - Resampler.h, Synthesizer.h: cubic interpolation now works in
1461 24bit mode too. Faster method to read 24bit words on little
1462 endian machines (x86): replaced 3 byte reads + shifts with a 1
1463 unaligned 32bit read and shift
1464 - experimental support for MS Windows (MIDI input via MME, AUDIO
1465 output via ASIO)
1466 - made handling of SIGINT signal (Ctrl-C) a bit more robust
1467 - support for monitoring the total number of active disk streams
1468 (new LSCP commands: GET TOTAL_STREAM_COUNT,
1471 * AUDIO driver:
1472 - added Windows ASIO low latency audio driver
1474 * MIDI driver:
1475 - added MME Windows MIDI driver
1477 * LSCP server:
1478 - added support for Windows style path / filenames, however with
1479 forward slash path separators instead of backslash
1480 (i.e. "C:/foo/bar.gig")
1481 - allow naughty liblscp to send non-string device parameters within
1482 apostrophes as well
1483 - added new LSCP commands: "GET FILE INSTRUMENTS <file>",
1484 "LIST FILE INSTRUMENTS <file>" and
1485 "GET FILE INSTRUMENT INFO <file> <index>" for retrieving informations
1486 about an arbitrary instrument file on the system where the sampler is
1487 running on
1489 Version 0.5.0 (15 October 2007)
1491 * packaging changes:
1492 - config.h is not going to be installed along with liblinuxsampler's
1493 API header files anymore
1494 - only the API relevant header (and source) files will be exported to
1495 the Doxygen API documentation (explicitly listed in Doxyfile.in)
1496 - added completely new XCode project files for Mac OSX which is now
1497 capable to execute our autoconf environment, thus no need anymore
1498 to maintain the compile time configuration file (osx/version.h) for
1499 OSX manually (patch by Toshi Nagata)
1500 - fixed buggy boolean --enable-foo / --disable-foo configure script
1501 parameters
1502 - global.h now only covers global definitions that are needed for the
1503 C++ API header files, all implementation internal global definitions
1504 are now in global_private.h
1505 - atomic.h is not exposed to the C++ API anymore
1506 - no need to include config.h anymore for using LS's API header files
1507 - fixed warnings in API doc generation
1509 * general changes:
1510 - replaced the old, confusing MIDI program change mechanism by a
1511 flexible MIDI instrument mapper which allows to map arbitrary
1512 (MIDI bank, MIDI program) pairs with arbitrary
1513 (engine type, instrument file, file index) triplets which will be
1514 loaded on the respective channel when such MIDI program change
1515 messages arrive, beside that, each entry allows to define a life-time
1516 strategy for the instrument, so the user can define whether the
1517 instrument should i.e. be permanently loaded (to be able to switch
1518 quickly among sounds) or i.e. loaded just on demand when the
1519 respective program change arrives (to safe RAM space), as well as a
1520 global volume factor for each entry, so the user can adjust the volume
1521 dynamic of his mapped instrument collection without having to modify
1522 the instrument files, also one can manage arbitrary amount of such
1523 MIDI instrument maps and assign each sampler channel individually
1524 a certain map, so that i.e. separation of normal instruments and
1525 drumkits is possible
1526 - new notification events for tracking changes to audio/MIDI devices,
1527 MIDI instrument maps, MIDI instruments, FX sends, global volume.
1528 - sampler was limited to load max. 200 instruments in the background
1529 due to a constant size RingBuffer FIFO which is now replaced by a
1530 dynamic (unlimited) size std::list FIFO
1531 - added FX sends, these allow to route audio signals to arbitrary audio
1532 output channels for being processed by external effect processors
1533 (i.e. jack-rack), the send levels are controllable via arbitrary MIDI
1534 controllers
1535 - global (sampler wide) volume can now be controlled at runtime
1536 - Implemented new, improved notification system
1537 - fixed compilation errors regarding OSX
1538 (patch by Toshi Nagata)
1539 - implemented instruments database
1540 - added support for escape sequences to the instruments database
1541 - added highly experimental support for on-the-fly instrument editing
1542 within the sampler's process (by using instrument editor plugins),
1543 you'll notice the new "Registered instrument editors:" message on
1544 startup, added a new LSCP command:
1545 "EDIT CHANNEL INSTRUMENT <sampler-channel>"
1546 to spawn a matching instrument editor for the instrument on the
1547 given sampler channel, the plugin path can be overridden at compile
1548 time with ./configure --enable-plugin-dir=/some/dir
1549 - added experimental code for synchronizing instrument editors hosted
1550 in the sampler's process to safely edit instruments while playing
1551 without a crash (hopefully) by either suspending single regions
1552 wherever possible or if unavoidable whole engine(s)
1553 - fixed several issues in fundamental "Thread" class: set scheduling
1554 policy and priority on thread level, set a minimum stack size for
1555 thread (TODO: a reasonable value yet to be tested), bugfix: non-RT
1556 threads simply inherited properties of starting thread instead of
1557 setting their own policy and priority
1558 - minor fix in our automatic stack trace mechanism on crashes, the
1559 main process did not wait for the stack trace process to finish
1560 its output
1561 - fixed some minor memory leaks
1562 - reenabled assembly features support, at the moment only for
1563 enabling a fast denormal FPU mode (x86 platforms supporting SSE2)
1564 - minor assembly fix in x86 features detection (don't use the PIC
1565 register, to avoid relocations in the text segment at runtime)
1566 - POSIX callback functions of Thread.h are hidden
1568 * MIDI driver:
1569 - dispatch bank select (MSB and LSB) messages
1571 * audio driver:
1572 - the ALSA audio output driver parameters now reflect the correct
1573 parameter value ranges for the respective selected sound card
1574 (patch by Till Wimmer, a bit fixed and extended)
1576 * Gigasampler format engine:
1577 - if a filter is used and EG2 finishes before EG1, let the voice
1578 die when EG2 has finished (this fixes a problem with clicks and
1579 voice starvation for some gigs)
1580 - playback is no longer disabled during instrument loading
1581 - all notes playing on a channel that changes its instrument keep
1582 playing with the old instrument until they get a note off
1583 command
1584 - EG fix: a release value of zero could cause noises or crash
1585 - handle MIDI coarse tuning messages (MIDI RPN #0 MSB #2 LSB)
1586 - EG fine tuning: when attack is zero the EG starts at a level
1587 above max sustain level, which means that there is a decay phase
1588 even if sustain is 100%
1589 - more EG fixes: the level could sometimes go below zero and cause
1590 noises or crashes
1591 - minor fix of EGDecay (patch by Toshi Nagata)
1592 - fixed compiler error when --enable-override-filter-type was
1593 supplied to the configure script (fixes #46)
1594 - disk thread: queue sizes are now proportional to CONFIG_MAX_STREAMS
1595 instead of a fix value
1596 - behavior fix: on MIDI CC# 65 (portamento on / off), 126 (mono mode),
1597 127 (solo mode) only kill voices if the respective mode really
1598 changed
1600 * LSCP server:
1601 - fixed compile time error for old Bison versions
1602 (i.e. v1.28 found on MacOS 10.4, patch by Toshi Nagata)
1603 - parser now supports extended ASCII character set
1604 (up to ASCII code 255, i.e. includes now umlauts and accents)
1605 - filename arguments in LSCP commands now allow to use escape
1606 sequences, that is directly literal as one of: \', \", \\, \n, \r,
1607 \f, \t, \v, or as octal ASCII code value like \132, or as hex ASCII
1608 code value like \xf2) (fixes bug #24)
1609 - the following LSCP commands now also support escape sequences for at
1610 least one of their text-based parameters (i.e. name, description):
1617 - returns verbose syntax errors (line and column where syntax error
1618 occured, the unexpected character and the actually expected, possible
1619 character(s), the latter only if less than 5 possibilities)
1620 - made sure that LSCP syntax is not affected by gigedit locale
1621 settings
1622 - bugfix regarding strings parameter lists: all comma separated lists
1623 of strings were treated as being one string containing commas
1624 (fixes #57)
1626 * Bug fixes:
1627 - fixed crash occurring on certain LSCP scripts (Bug 39)
1628 - another thread safety fix for lscp "load engine" and "set
1629 channel audio output device"
1630 - fixed a crash which occurs when reassigning the same engine
1631 on a sampler channel with connected MIDI device
1632 - fixed a crash which occurs when changing the number of ports of a MIDI
1633 device connected to a sampler channel to number less then or equal
1634 to the index of the port to which the sampler channel is connected.
1635 - The previous bindings were not been disconnected when altering
1636 the ALSA_SEQ_BINDINGS parameter. Introduced a NONE keyword for
1637 unsubscribing from all bindings (e.g. ALSA_SEQ_BINDINGS=NONE).
1638 - The active stream/voice count statistic was incorrect.
1639 - notification events were not been sent for some sampler
1640 channel changes
1641 - added default min and max values to restrict the number of allowed
1642 audio output channels and MIDI input ports
1643 - the connection to the PCM interface is now closed when destroying
1644 an audio output device
1645 - files with slash in their path or filename could not be loaded
1647 * test cases:
1648 - updated and fixed (haven't been touched in a while)
1650 Version 0.4.0 (24 November 2006)
1652 * packaging changes:
1653 - changed deprecated copyright attribute to license; added ldconfig
1654 to post-(un)install steps; added devel package for liblinuxsampler;
1655 to linuxsampler.spec (RPM)
1656 - install necessary development header files for allowing 3rd party
1657 applications to link against liblinuxsampler
1658 - liblinuxsampler's API documentation can be generated with 'make docs'
1659 (Doxygen required)
1660 - added benchmark to automatically detect the best triangle LFO
1661 implementation (currently either an integer math solution or a
1662 di-harmonic approximation), automatic detection can be overriden
1663 with --enable-signed-triang-algo=x and --enable-unsigned-triang-algo=x
1664 configure script argument though (mandatory for cross-compilation)
1665 - do not automatically pick optimized gcc flags if the user already
1666 provided some on his own (as CXXFLAGS)
1667 - added compile time option to disable processing of All-Notes-Off MIDI
1668 messages
1669 - added compile time options to allow disabling the various audio and
1670 MIDI drivers
1671 - fixed automatic GCC CPU switch detection on PPC
1672 (patch by Ebrahim Mayat)
1674 * Gigasampler format engine:
1675 - extensive synthesis optimization
1676 (reimplementation of EGs and LFO(s), removed synthesis parameter
1677 prerendering and the synthesis parameter matrix in general, splitting
1678 each audio fragment into subfragments now where each subfragment uses
1679 constant pitch and filter coefficients. The volume coefficient is
1680 linearly interpolated inside a subfragment, unless
1681 --disable-interpolate-volume is set.)
1682 - fine tuning of the EG modulation parameters
1683 - improved filter cutoff calculation by adding support for the
1684 following gig parameters: Cutoff freq (used when no cutoff
1685 controller is defined), Control invert, Minimum cutoff, Velocity
1686 curve and Velocity range. The keyboard tracking now scales
1687 cutoff frequency instead of resonance.
1688 - added support for gig parameter Resonance.
1689 - fixed bug in sysex handling (patch by Juan Linietsky)
1690 - added global volume attenuation of -9 dB (0.35f) to prevent clipping
1691 which can be overridden with --enable-global-attenuation
1692 - EG fixes: made the length of "attack hold" stage more
1693 accurate. Release stage can now start before attack stage
1694 ends. Cancel release didn't work when sustain was zero. Attack
1695 time now has a minimal value to prevent clicks.
1696 - fixed pitch changes larger than one octave
1697 - fixed EG3 (pitch envelope) synthesis which was neutral all the time
1698 - implemented portamento mode and solo mode (a.k.a 'mono mode'):
1699 all modes can be altered via standard GM messages, that is CC5 for
1700 altering portamento time, CC65 for enabling / disabling portamento
1701 mode, CC126 for enabling solo mode and CC127 for disabling solo mode
1702 - fine tuning of the curves for volume (CC7), pan (CC10 and gig
1703 parameter) and crossfade
1704 - added support for the "attenuation controller threshold" gig
1705 parameter
1706 - added smoothing of volume changes caused by control change
1707 messages
1708 - sample loop parameters are now taken from the DimensionRegion
1709 instead of the wave chunk
1710 - fixed keyswitching for v3 gigs with a number of keyswitch splits
1711 not equal to a power of two
1712 - reimplementation of the filter algorithm. The new filter is more
1713 accurate and supports all gig filter types, including bandreject
1714 and lowpass turbo.
1715 - real support for 24 bit samples - samples are not truncated to
1716 16 bits anymore
1717 - support for aftertouch (channel pressure, not polyphonic
1718 aftertouch)
1719 - LFO1 behaviour fixed (dampening from max volume instead of
1720 amplifying from 0)
1722 * LSCP server:
1723 - fixed application exit on broken pipe error (fixes bug #20)
1724 - fixed the notification messages delay due to lack of
1725 network activity (fixes bug #26)
1726 - fixed parser bug which occured on space(s) within device parameters
1729 * audio driver:
1730 - added aRts audio output driver (by no means RT safe)
1732 * MIDI driver:
1733 - fixed legacy sysex code which caused dispatching of MIDI SysEx
1734 messages several times instead of once
1735 - API extension for MIDI drivers which already supply exact time stamps
1736 for events (i.e. for offline rendering based MIDI drivers)
1737 - added checks for bad MIDI data
1739 * linuxsampler application:
1740 - show available sampler engine types on startup
1742 * general changes:
1743 - support for muting sampler channels and solo mode of the same, two new
1744 LSCP commands ("SET CHANNEL MUTE" and "SET CHANNEL SOLO") and two new
1745 fields ("MUTE" and "SOLO") for command "GET CHANNEL INFO" were
1746 introduced for this, the behavior is the same like on a mixer console
1747 (patch by Grigor Iliev, a bit adjusted). Also added configure option
1748 --enable-process-muted-channels which can be used to enable the
1749 processing of muted channels.
1750 - support for sostenuto pedal
1751 - support for monitoring the total number of active voices
1754 - fixed some memory management errors
1755 - fixed some concurrency problems that could lead to crashes when
1756 LSCP commands were executed
1757 - fixed crash when instrument loading failed previously on the same
1758 sampler channel (fixes bug #36)
1760 Version 0.3.3 (15 July 2005)
1762 * packaging changes:
1763 - fixed compilation with gcc 4.0
1765 * Gigasampler format engine:
1766 - LFO preprocessing was performed even though the respective LFO was
1767 disabled by the instrument patch (minor efficiency fix)
1768 - if period time of chosen audio device is too small (< MIN_RELEASE_TIME)
1769 for volume ramp downs in the same fragment (needed for current voice
1770 stealing implementation) then simply inform the user with a warning
1771 about possible click sounds and reduce the volume ramp down
1772 appropriately instead of cancelling the audio device connection
1774 Version 0.3.2 (24 June 2005)
1776 * packaging changes:
1777 - updated autotools build files to compile on OS X
1778 (conditional compilation of CoreMIDI and MidiShare drivers)
1779 - hand-crafted assembly optimization code can be disabled with
1780 './configure --disable-asm' (definitely not recommended)
1781 - fixed 'make dist' rule to include all necessary files
1782 - require automake (>= 1.5) for 'make -f Makefile.cvs'
1783 (due to 'dist-bzip2' automake option)
1785 * Gigasampler format engine:
1786 - support for the gig parameters for "release velocity response" curves
1787 - fine tuning of the EGADSR envelope
1788 - volume of release triggered samples now depends on note-on velocity,
1789 note length and gig parameter "release trigger decay" instead of
1790 note-off velocity.
1791 - revised voice stealing
1792 (fixes crash and endless loop caused by voice stealing)
1793 - don't reset scale tuning on instrument or audio output device change
1794 - handle key group conflicts right at the beginning of each fragment
1795 instead of when the respective voice is actually launched
1796 (fixes undefined behavior if stolen voices belonged to a key group -
1797 this case was followed by a "killed voice survived" error message)
1798 - fixed minor issue with null/silence samples
1799 (those stole voices even though they don't need a voice at all which
1800 resulted in "voice stealing didn't work out" messages)
1801 - don't reset volume, pan, pitch and MIDI controller values on
1802 instrument or audio output device change
1804 * LSCP server:
1805 - fixed some crashes (patch by Grigor Iliev, fixes #19)
1806 - fixed LSCP event "CHANNEL_INFO" notification
1807 (e.g. did not notify on volume changes or MIDI program change events)
1809 * linuxsampler application:
1810 - added command line parameters --lscp-addr and --lscp-port to override
1811 default IP address and port of LSCP server
1813 Version 0.3.1 (24 May 2005)
1815 * initial release

  ViewVC Help
Powered by ViewVC