1 |
Version SVN trunk (?) |
2 |
|
3 |
* GigaStudio/Gigasampler format engine: |
4 |
- Use libgig's new IO-per-thread feature to avoid file I/O concurrency |
5 |
issues with .gig file. |
6 |
- Use libgig's new method gig::File::GetSample() with method |
7 |
gig::File::CountSamples() instead of now deprecated |
8 |
gig::File::GetFirstSample() / gig::File::GetNextSample() pair. |
9 |
|
10 |
Version 2.2.0 (9 May 2021) |
11 |
|
12 |
* general changes: |
13 |
- Fixed compiler error in Pool.h. |
14 |
- Require C++14 compiler support. |
15 |
- Autoconf: Added m4/ax_cxx_compile_stdcxx.m4 macro which is used |
16 |
for checking in configure for C++14 support (as mandatory |
17 |
requirement) and automatically adds compiler argument if required |
18 |
(e.g. -std=C++14). |
19 |
- RTMath: Implemented floating point comparison methods |
20 |
fEqual32(float,float) and fEqual64(double,double) which take the |
21 |
expected floating point tolerances into account. |
22 |
- Added public API C++ class "LFO", which is a cluster class encapsulating |
23 |
all the sampler's LFO implementations to be used by 3rd party applications |
24 |
(e.g. by Gigedit). |
25 |
- Added int math square LFO implementation. |
26 |
- Added int math saw LFO implementation. |
27 |
- Added numeric complex nr sine LFO implementation. |
28 |
- Marked class LFOTriangleDiHarmonic as deprecated |
29 |
(will be removed in future). |
30 |
- Added LFOAll.h which includes all LFO implementation's header files. |
31 |
- Refactored LFO class names and their file names (PulseLFO -> LFOPulse, |
32 |
LFOSawIntMath -> LFOSawIntMathNew, SawLFO -> LFOSawIntMathOld, |
33 |
SineLFO -> LFOSineBuiltinFn, LFOSine -> LFOSineNumericComplexNr, |
34 |
SquareLFO -> LFOSquarePulse and separated the latter to its own header |
35 |
file). |
36 |
- Renamed type LFOSigned -> LFOTriangleSigned. |
37 |
- Renamed type LFOUnsigned -> LFOTriangleUnsigned. |
38 |
- Using now cubic interpolation as resampling algorithm by default; and |
39 |
linear interpolation is deprecated now. |
40 |
- Event:Init(): memset() entire Event::Param union to zero instead of trying |
41 |
to do that for its individual members; the latter became error prone due |
42 |
to the union's growing complexity meanwhile. |
43 |
- EngineChannel: Overhaul of RPN/NRPN related methods: Added new |
44 |
*Rpn*Parameter*() methods as replacement for previous *Rpn*Controller*() |
45 |
methods and marked the old ones as deprecated; and likewise added new |
46 |
*Nrpn*Parameter*() methods as replacement for previous *Nrpn*Controller*() |
47 |
methods and marked those old ones deprecated as well. |
48 |
- Added support for MIDI CC #38 (data entry LSB): this allows higher value |
49 |
resolution for certain (e.g. hard coded) RPN and NRPN parameters. |
50 |
- Implemented support for compressed RPN/NRPN message sequences (that is to |
51 |
allow processing MIDI data entry messages without always having to be |
52 |
preceded strictly by RPN/NRPN parameter selection messages on each value |
53 |
change). |
54 |
- Added support for MIDI CC #96 (data increment) and MIDI CC #97 (data |
55 |
decrement). |
56 |
- Added support for 'null function RPN/NRPN' (which is RPN/NRPN parameter |
57 |
MSB=127 and LSB=127) and which is disabling subsequent data |
58 |
entry/increment/decrement CC messages. |
59 |
- Fixed crash when editing key groups (a.k.a. exclusive groups) with an |
60 |
instrument editor. |
61 |
- Fixed playing noise at the end of a disk stream under certain conditions. |
62 |
- Fixed potential crash with high pitch at end of RAM playback. |
63 |
- Added configure checks whether C++ compiler supports so called |
64 |
'designated initializers'. |
65 |
- Added global backtraceAsString() function for debugging purposes |
66 |
(POSIX systems only, no Windows implementation yet). |
67 |
- Thread class: Added new methods pushCancelable(), popCancelable(), name(), |
68 |
nameOfCaller() and setNameOfCaller(). |
69 |
- configure: Always check for availability of pthread_testcancel() and use |
70 |
it if so (previously it was only used if explicitly enabled by configure |
71 |
option). |
72 |
- Thread class: raise compiler warning if pthread_testcancel() is not |
73 |
available, because it can lead to deadlocks. |
74 |
- Mutex class: Implemented optional bug detection and deadlock debugging |
75 |
features. |
76 |
- Ref<> class is now thread safe (as lock-free and wait-free |
77 |
implementation). |
78 |
- Fixed compilation error with Bison >= 3.6.0. |
79 |
- Fixed rare crash when switching between instruments that had instrument |
80 |
scripts loaded. |
81 |
- Allow EffectChain::InsertEffect() as alternative to |
82 |
EffectChain::AppendEffect() (i.e. with index equal to current chain size). |
83 |
- ConstCapacityArray class: Fixed bug in remove() method. |
84 |
- Fixed crash on certain MIDI RPN and NRPN messages that occurred if no |
85 |
real-time instrument script was loaded. |
86 |
|
87 |
* Real-time instrument scripts: |
88 |
- Added method ScriptVM::setExitResultEnabled() which allows to |
89 |
explicitly enable the built-in exit() function to optionally accept |
90 |
one function argument; the value of the passed exit() function |
91 |
argument will then become available by calling |
92 |
VMExecContext::exitResult() after script execution. |
93 |
- 64 bit support for NKSP integer scripts variables (declare $foo). |
94 |
- Variable names, function names and preprocessor condition names must start |
95 |
with a regular character (a-z or A-Z); starting them with a digit or |
96 |
underscore is not allowed. |
97 |
- NKSP parser fix: equal comparison operator "=" and not equal comparison |
98 |
operator "#" must only accept integer operands. |
99 |
- NKSP language: Implemented support for standard units like Hertz, seconds, |
100 |
Bel including support for metric unit prefixes; so one can now e.g. |
101 |
conveniently use numbers in scripts like "5us" meaning "5 microseconds", |
102 |
or e.g. "12kHz" meaning "12 kilo Hertz", or e.g. "-14mdB" meaning |
103 |
"minus 14 Millidecibel", or e.g. "28c" meaning "28 cents" (for tuning). |
104 |
- NKSP language: Introduced "final" operator "!" which is specifically |
105 |
intended for synthesis parameter values to denote that the synthesis |
106 |
parameter value is intended to be the "final" value for that synthesis |
107 |
parameter that should explicitly be used by the engine and thus causing |
108 |
the sampler engine to ignore all other modulation sources for the same |
109 |
synthesis parameter (like e.g. LFO, EG); by simply prefixing a value, |
110 |
variable or formula with this new "!" operator the expression is marked as |
111 |
being "final". |
112 |
- NKSP script editor API: Added support for detecting standard unit tokens |
113 |
and their potential metric prefix token. |
114 |
- NKSP language: Added support for NKSP real number literals and |
115 |
arithmetic operations on them (e.g. "(3.9 + 2.9) / 12.3 - 42.0"). |
116 |
- NKSP language: Added support for NKSP real number (floating point) script |
117 |
variables (declare ~foo). |
118 |
- NKSP language: Added support for NKSP real number (floating point) array |
119 |
script variables (declare ?foo[]). |
120 |
- Built-in script function "message()" accepts now real number argument as |
121 |
well. |
122 |
- Added built-in script function "real_to_int()" and its short hand form |
123 |
"int()" for casting from real number to integer in NKSP scripts. |
124 |
- Added built-in script function "int_to_real()" and its short hand form |
125 |
"real()" for casting from integer to real number in NKSP scripts. |
126 |
- Allow built-in exit() function to potentially accept real number type |
127 |
argument as well. |
128 |
- Built-in script functions may have a different return type depending on |
129 |
the arguments passed to the function. |
130 |
- Built-in script function "abs()" optionally accepts and returns real |
131 |
number. |
132 |
- Built-in script functions "min()" and "max()" optionally accept real |
133 |
number arguments and return real number as result in that case. |
134 |
- NKSP VM API: Allow units and 'final'ness to be returned as result from |
135 |
built-in functions (added methods VMFunction::returnUnitType() and |
136 |
VMFunction::returnsFinal() for that purpose which must be implemented by |
137 |
built-in function implementations). |
138 |
- NKSP language: Allow metric unit prefixes of numeric scalar and array |
139 |
variables to be changed freely at runtime (unlike unit types like Hz etc. |
140 |
which are still sticky, parse-time features of variables which cannot be |
141 |
changed at runtime for the intentional sake of determinism). |
142 |
- NKSP language: 'final' values are prohibited for array variables for now |
143 |
(attempt causes a parsers error). |
144 |
- NKSP language: expressions with unit types (e.g. Hz) are prohibited for |
145 |
conditions of runtime control structures like if(), while(), select() |
146 |
(attempt causes a parser error). |
147 |
- NKSP VM API: Allow built-in functions to perform their own, individual |
148 |
parse time checks of arguments going to be passed to the function at |
149 |
runtime (added method VMFunction::checkArgs() for that purpose). |
150 |
- NKSP language: raise parser warning if only one operand of binary |
151 |
operators (like logical 'or' comparison) contain a 'final' value (because |
152 |
it would always yield in a 'final' result in such cases). |
153 |
- NKSP language: Allow comparison (=, #, <, >, <=, >=) of values with |
154 |
different metric unit prefixes, which will behave as expected (e.g. |
155 |
result of expression '1000us < 2ms' is true). |
156 |
- NKSP language: Allow adding values with different metric unit prefixes |
157 |
(e.g. result of expression '100Hz + 5kHz' is '5100Hz'). |
158 |
- NKSP language: Allow subtracting values with different metric unit |
159 |
prefixes (e.g. result of expression '1ms - 20us' is '980us'). |
160 |
- NKSP language: Allow multiplying with any metric unit prefixes |
161 |
(e.g. result of expression '2k * 3ms' is '6s'), however multiplications |
162 |
with unit types on both sides (e.g. '2s * 2s') is still prohibited since |
163 |
we don't have any considerable practical use for a term like '4s^2' |
164 |
(hence any attempt multiplying two unit types still causes parser error). |
165 |
- NKSP language: Allow dividing by any metric unit prefixes and allow |
166 |
division of same unit type on both sides (e.g. expression '8kHz / 1000Hz' |
167 |
yields in unit free result '8'). So this is now a way to cast units away |
168 |
e.g. for passing the result to other expressions, certain function calls |
169 |
or variables which are not accepting any units (or that specific unit). |
170 |
- NKSP language: integer arrays and real number arrays can now be converted |
171 |
to strings (e.g. for dumping their content with message() calls for |
172 |
script debugging purposes). |
173 |
- NKSP language: expressions and variables with units are now correctly |
174 |
casted to strings (e.g. with message() calls). |
175 |
- NKSP language: comparing real numbers for equalness (e.g. '~foo = 3.1') or |
176 |
unequalness (e.g. '~foo # 3.1') is now less strict and takes the expected |
177 |
floating point tolerances into account. |
178 |
- NKSP VM API: Added methods VMScalarNumberExpr::evalCastInt() and |
179 |
VMScalarNumberExpr::evalCastReal(). |
180 |
- NKSP VM API: Added base class 'VMNumberArrayExpr' for classes |
181 |
'VMIntArrayExpr' and 'VMRealArrayExpr'. |
182 |
- NKSP VM API: replaced all unitPrefix() (parse time) methods by |
183 |
unitFactor() (runtime) methods. |
184 |
- Built-in function "exit()" supports now returning units and 'final'ness |
185 |
for test cases. |
186 |
- The following built-in functions support now units as well: "abs()", |
187 |
"random()", "inc()", "dec()", "in_range()", "min()", "max()", |
188 |
"real_to_int()", "int()", "int_to_real()" and "real()". |
189 |
- Built-in functions "array_equal()", "search()" and "sort()" support now |
190 |
real number arrays (correctly) as well. |
191 |
- Added individual parse time checks of arguments to be passed to built-in |
192 |
functions "random()", "inc()", "dec()", "in_range()", "min()", "max()", |
193 |
"array_equal()" and "search()" specific for their individual purposes. |
194 |
- NKSP VM refactoring: Renamed all methods, functions and classes matching |
195 |
pattern *ScalarNumber* to simply *Number* (that is i.e. classes |
196 |
VMScalarNumberExpr -> VMNumberExpr, ScalarNumberExpr -> NumberExpr, |
197 |
ScalarNumberVariable -> NumberVariable, ScalarNumberBinaryOp -> |
198 |
NumberBinaryOp, VMScalarNumberResultFunction -> VMNumberResultFunction, |
199 |
method VMExpr::asScalarNumberExpr() -> VMExpr::asNumber(), function |
200 |
isScalarNumber() -> isNumber()). |
201 |
- NKSP VM API: Added 4 overridden methods to class VMNumberExpr: |
202 |
evalCastInt(MetricPrefix_t), evalCastInt(MetricPrefix_t,MetricPrefix_t), |
203 |
evalCastReal(MetricPrefix_t), evalCastReal(MetricPrefix_t,MetricPrefix_t) |
204 |
as convenient methods for automatically converting values to expected |
205 |
metric value basis. |
206 |
- Built-in function "wait()" accepts now both integers and real numbers as |
207 |
argument. |
208 |
- NKSP VM API cleanup: Get rid of legacy method |
209 |
VMFunction::argType(vmint iArg) which was already superseded by its new |
210 |
replacement VMFunction::acceptsArgType(vmint iArg, ExprType_t type). |
211 |
- NKSP parser: if wrong argument type was passed to a built-in function and |
212 |
that built-in function accepts more than one data type for the argument, |
213 |
then show all supported data types as parser error message. |
214 |
- Built-in function "play_note()" accepts now real numbers and seconds as |
215 |
unit type as well for its 3rd and 4th function arguments. |
216 |
- The following built-in functions accept now real numbers as well for their |
217 |
2nd function argument: "change_vol()", "change_tune()", "change_cutoff()", |
218 |
"change_attack()", "change_decay()", "change_release()", |
219 |
"change_sustain()", "change_cutoff_attack()", "change_cutoff_decay()", |
220 |
"change_cutoff_sustain()", "change_cutoff_release()", |
221 |
"change_amp_lfo_freq()", "change_cutoff_lfo_freq()", |
222 |
"change_pitch_lfo_freq()", "change_vol_time()", "change_tune_time()", |
223 |
"change_pan_time()", "fade_in()", "fade_out()", "change_play_pos()". |
224 |
- Fixed built-in function "change_play_pos()" not having accepted metric |
225 |
prefixes at all. |
226 |
- Fixed the following built-in functions having misinterpreted values given |
227 |
with unit type (for their 2nd argument) as if they were relative values |
228 |
(that is as if they were passed without a unit type): "change_attack()", |
229 |
"change_decay()", "change_release()", "change_cutoff_attack()", |
230 |
"change_cutoff_decay()", "change_cutoff_release()". |
231 |
- Fixed the following built-in functions having applied completely wrong |
232 |
'final' values: "change_sustain()", "change_cutoff_sustain()" (since the |
233 |
respective EGs being their modulation sink assume uint data type with |
234 |
value range 0..1000 instead of 0.0..1.0. |
235 |
- Added individual parse-time checks of function arguments for the following |
236 |
built-in functions: "play_note()", "note_off()", "set_event_mark()", |
237 |
"delete_event_mark()", "by_marks()", "change_cutoff()", "change_attack()", |
238 |
"change_decay()", "change_release()", "change_cutoff_attack()", |
239 |
"change_cutoff_decay()", "change_cutoff_release()", |
240 |
"change_amp_lfo_freq()", "change_cutoff_lfo_freq()", |
241 |
"change_pitch_lfo_freq()", "change_vol_time()", "change_tune_time()" and |
242 |
"change_pan_time()". |
243 |
- Don't abort function call if unit type was used and at the same time |
244 |
'final' operator was omitted for the primary value argument of the |
245 |
following built-in functions: "change_cutoff()", "change_attack()", |
246 |
"change_decay()", "change_release()", "change_cutoff_attack()", |
247 |
"change_cutoff_decay()", "change_cutoff_release()", |
248 |
"change_amp_lfo_freq()", "change_cutoff_lfo_freq()", |
249 |
"change_pitch_lfo_freq()", "change_vol_time()", "change_tune_time()", |
250 |
"change_pan_time()", instead imply 'final'ness at runtime and raise an |
251 |
appropriate parser warning at parse time. |
252 |
- Added built-in real number functions "round()", "ceil()", "floor()", |
253 |
"sqrt()", "log()", "log2()", "log10()", "exp()", "pow()", "sin()", |
254 |
"cos()", "tan()", "asin()", "acos()", "atan()". |
255 |
- Added built-in script real number constant "~NI_MATH_PI". |
256 |
- Added built-in script real number constant "~NI_MATH_E". |
257 |
- NKSP language: Allow unary '+' operator. |
258 |
- Added built-in script functions "msb()" and "lsb()". |
259 |
- NKSP language: Added support for RPN event handler |
260 |
("on rpn ... end on" in instrument scripts). |
261 |
- NKSP language: Added support for NRPN event handler |
262 |
("on nrpn ... end on" in instrument scripts). |
263 |
- Added built-in read-only variables "$RPN_ADDRESS" and "$RPN_VALUE" which |
264 |
may be read from the new RPN/NRPN script handlers to get the (N)RPN |
265 |
parameter that had been changed and its new value. |
266 |
- Added built-in const variables "$NI_CB_TYPE_RPN" and "$NI_CB_TYPE_NRPN" |
267 |
which are identifying the new (N)RPN handlers as such at script runtime. |
268 |
- Implemented built-in script function "set_rpn()". |
269 |
- Implemented built-in script function "set_nrpn()". |
270 |
- Fix: built-in functions returning an array type were broken and caused a |
271 |
crash when the built-in function's (supposed to be array) result value was |
272 |
used by the script VM. |
273 |
- Removed upper value constraint for 2nd argument of built-in functions |
274 |
"change_amp_lfo_freq()", "change_cutoff_lfo_freq()", |
275 |
"change_pitch_lfo_freq()","change_reso()", "change_cutoff()", |
276 |
"change_amp_lfo_depth()", "change_cutoff_lfo_depth()" and |
277 |
"change_pitch_lfo_depth()", to allow e.g. passing value 2000000 to double |
278 |
the LFO frequency / resonance / cutoff frequency. |
279 |
- Reduced code involved for parsing NKSP variable declarations by using 2 |
280 |
generalized grammar rules with optional components (i.e. optional value |
281 |
assignment and optional variable qualifiers) instead of previous linear |
282 |
combinations of grammar rules, which also fixes some edge case. |
283 |
- NKSP language parser: track code block locations also by a) raw byte |
284 |
position in NKSP source code string and b) length of that code block in |
285 |
bytes (along to the already existing code block tracking by line/column); |
286 |
and added appropriate new members to public C++ API classes like |
287 |
'firstByte' and 'lengthBytes' to class VMSourceToken and struct CodeBlock. |
288 |
- NKSP language: Added support for 'patch' variable qualifier (capture their |
289 |
locations and allow replacement of their assignment expression via VM C++ |
290 |
API). |
291 |
- Fixed incorrect EG times if standard measuring unit was passed to built-in |
292 |
functions "change_attack()", "change_decay()" or "change_release()". |
293 |
- Fixed intermediate function result values never having reflected any |
294 |
standard measuring unit type. |
295 |
- Fixed re-entrant issue with function calls which caused wrong result |
296 |
values if the same function was called multiple times in a term |
297 |
(specifically if metric prefixes were used). |
298 |
- NKSP language: Just throw a warning, not an error if an array variable of |
299 |
size zero was declared. |
300 |
- NKSP language: Allow omitting explicit array size on array variable |
301 |
declaration if combined with immediate value assignment |
302 |
(e.g. declare %foo[] := ( 1, 2, 3 ) ). |
303 |
- NKSP language: emit warning if an array variable was declared with bigger |
304 |
array size than amount of initial values been assigned, and initialize the |
305 |
missing array elements with zero in this case. |
306 |
- Fixed built-in NKSP exit() function to behave as return statement. |
307 |
- NKSP parser: Fixed crash if unary '-' operator was used on a non-number |
308 |
data type. |
309 |
- NKSP VM: Fixed the "release" handler being executed too often under |
310 |
certain situations if polyphonic data was passed from "note" handler to |
311 |
"release" handler. |
312 |
- Fixed memory leak in NKSP parser: string tokens were allocated as C |
313 |
strings and never freed. |
314 |
- Fixed incorrect polyphonic data transfer from wrong note handler to |
315 |
release handler; which also fixes no release handler being executed |
316 |
sometimes, and due to the latter it also fixes potential crashes as some |
317 |
polyphonic script events were never released and the engine hence was |
318 |
running out of free script events. |
319 |
- Fixed crash when handling array results of built-in functions |
320 |
(e.g. "num_elements(by_marks($MARK_1)"). |
321 |
- NKSP VM: Fixed dead note events not been removed from user's event groups |
322 |
(i.e. causing incorrect behaviour of built-in NKSP functions by_marks() |
323 |
and set_event_mark() after max. size of event group list was exhausted). |
324 |
|
325 |
* test cases: |
326 |
- Fixed compiler errors in test cases. |
327 |
- Updated README for how to compile & run test cases. |
328 |
- Updated test case |
329 |
MutexTest::testDoubleLockStillBlocksConcurrentThread() to latest |
330 |
expected behaviour of the Mutex class implementation (recursive |
331 |
mutex type). |
332 |
- Added test cases for NKSP core language aspects and core built-in |
333 |
functions. |
334 |
- Fixed thread tests segfaulting on Linux. |
335 |
- NKSP: Added real number test cases for built-in functions exit(), |
336 |
int_to_real(), real(), real_to_int() and int(), as well as for the |
337 |
plus, minus and negate language operators. |
338 |
- Added massive amount of NKSP test cases for standard measuring units and |
339 |
'final' operator usage cases. |
340 |
- Added NKSP test cases for (floating point tolerance aware) real number |
341 |
equalness / unequalness comparison. |
342 |
- Added NKSP int array and real array tests for value assignment and |
343 |
initialization of arrays. |
344 |
- Added NKSP test cases for built-in functions "round()", "ceil()", |
345 |
"floor()", "sqrt()", "log()", "log2()", "log10()", "exp()", "pow()", |
346 |
"sin()", "cos()", "tan()", "asin()", "acos()", "atan()". |
347 |
- Added NKSP test cases for unary '+' operator. |
348 |
- Added NKSP test cases for core built-in variables. |
349 |
- Added thorough NKSP test cases for variable declarations. |
350 |
|
351 |
* GigaStudio/Gigasampler format engine: |
352 |
- LFOTriangleIntMath and LFOTriangleIntAbsMath: Fixed FlipPhase=true |
353 |
behaviour for start_level_mid. |
354 |
- Changed LFO start levels: LFO1 and LFO2 both to mid, LFO3 to max. start |
355 |
level (see discussion "GigaStudio LFO compatibility" on mailing list |
356 |
from 2019-09-26 for details). |
357 |
- Changed default wave form for all 3 LFOs to sine (instead of triangle; |
358 |
see discussion "GigaStudio LFO compatibility" on mailing list from |
359 |
2019-09-26 for details). |
360 |
- Format extension: Added support for different LFO wave forms (currently |
361 |
either sine [default], triangle, saw or square). |
362 |
- Format extension: Added support for LFO phase displacement (0��..360��). |
363 |
- Format extension: Added support for flipping LFO polarity on LFO 3 |
364 |
(in the original gig format this was only available for LFO 1 and LFO 2). |
365 |
- Format extension: Support for additional filter types: |
366 |
lowpass 1/2/4/6-pole, highpass 1/2/4/6-pole, bandpass 2-pole, |
367 |
bandreject 2-pole. |
368 |
- Fixd potential crash with polynomial interpolator and max. pitch. |
369 |
- If minimum cutoff is defined, remap cutoff controller CC values to really |
370 |
span the entire possible cutoff range (and avoiding a dead controller |
371 |
zone). |
372 |
- Implemented support for loading instrument scripts with 'patch' variables |
373 |
(by patching these script variables with potentially overridden ones by |
374 |
gig Instrument). |
375 |
|
376 |
* SFZ format engine: |
377 |
- Fixed support for regions with loccN/hiccN conditions on more than one |
378 |
MIDI controller. |
379 |
- Fixed internal opcode 'xfin_hivel' falsely being set by user opcode |
380 |
'xfin_lovel' (patch by Christian Czezatke). |
381 |
|
382 |
* Benchmarks: |
383 |
- Fixed benchmarks/triang.cpp falsely having favoured "int math abs" |
384 |
algorithm (since result of 2nd run was not accumulated). |
385 |
- Added benchmark for saw wave (benchmarks/saw.cpp). |
386 |
- Added benchmark for sine wave (benchmarks/sine.cpp). |
387 |
- Added benchmark for square wave (benchmarks/square.cpp). |
388 |
- Increased amount of benchmarks runs by factor 6 to achieve benchmark times |
389 |
which are large enough on modern systems. |
390 |
|
391 |
Version 2.1.1 (27 Jul 2019) |
392 |
|
393 |
* Real-time instrument scripts: |
394 |
- Fixed behavior of built-in NKSP functions change_sustain(), |
395 |
change_cutoff_attack(), change_cutoff_decay(), change_cutoff_sustain() |
396 |
and change_cutoff_release(). |
397 |
|
398 |
* general changes: |
399 |
- Only play release trigger samples on sustain pedal up if this behaviour |
400 |
was explicitly requested by the instrument (otherwise only on note-off). |
401 |
- Fixed compiler warnings. |
402 |
- Fixed compilation error when cross-compiling to Mac. |
403 |
- FX Sends: Provide more useful error messages on routing problems |
404 |
(see bug #169). |
405 |
- LSCP doc: Be more clear describing the two distinct approaches |
406 |
of using external vs. internal effects (see bug #169). |
407 |
- "optional" class: Fixed comparison operators. |
408 |
|
409 |
* Gigasampler/GigaStudio format engine: |
410 |
- Format extension: If requested by instrument then don't play release |
411 |
trigger sample on note-off events. |
412 |
|
413 |
* SFZ format engine: |
414 |
- Fixed memory leak when releasing samples |
415 |
(fixes bug #307, patch by Jacek Roszkowski) |
416 |
- Fixed potential crash when a sample is shared by more than one region |
417 |
(fixes bug #308, patch by Jacek Roszkowski). |
418 |
- Opcode 'sample': Added support for built-in sample '*silence' |
419 |
(fixes bug #310, patch by Jacek Roszkowski). |
420 |
|
421 |
Version 2.1.0 (25 Nov 2017) |
422 |
|
423 |
* SFZ format engine: |
424 |
- added support for <global>, <master> and #define (patch by Alby M) |
425 |
- Removed code duplication in SFZ file loading code. |
426 |
- Added support for sfz extension opcode 'script' which may be used to |
427 |
load real-time instrument script file (NKSP script language). |
428 |
- Implemented opcode set_ccN (initial patch by Giovanni Senatore). |
429 |
- Fixed unintended volume fade-in of voices under certain conditions. |
430 |
- sfz parser: allow missing space between header and opcode |
431 |
|
432 |
* Gigasampler/GigaStudio format engine: |
433 |
- Fixed clicks and pumping noise with Lowpass Turbo filter on very low |
434 |
cutoff settings. |
435 |
- Got rid of resembling an ancient GSt misbehavior which did not pitch at |
436 |
all if an up-pitch of more than 40 semi tones was requested (I don't |
437 |
think there is any stock gig sound that requires this behavior to |
438 |
resemble its original sound). |
439 |
- Added support for controlling whether the individual EGADSR stages may |
440 |
be aborted (as LinuxSampler extension to the original GigaStudio 4 |
441 |
format). |
442 |
|
443 |
* general changes: |
444 |
- fixed printf type errors (mostly in debug messages) |
445 |
- use unique_ptr instead of auto_ptr when building with C++11 |
446 |
- Added RTAVLTree class which is a real-time safe ordered multi-map, thus |
447 |
allowing to sort data efficiently in real-time safe manner. |
448 |
- RTList class: added methods for moving/inserting elements to arbitrary |
449 |
position within a list. |
450 |
- RTList class: added method fromPtr() for creating an Iterator object from |
451 |
a raw element pointer. |
452 |
- Attempt to partly fix resetting engine channels vs. resetting engine, an |
453 |
overall cleanup of the Reset*(), ConnectAudioDevice(), |
454 |
DisconnectAudioDevice() API methods would still be desirable though, |
455 |
because the current situation is still inconsistent and error prone. |
456 |
- Active voices are now internally grouped to "Note" objects, instead of |
457 |
being directly assigned to a keyboard key. This allows more fine graded |
458 |
processing of voices, which is i.e. required for certain instrument |
459 |
script features. |
460 |
- Fix: Release trigger voices were not spawned on sustain pedal up |
461 |
(CC #64) events. |
462 |
- Fix: Release trigger voices can now also distinguish correctly between |
463 |
note-off and sustain pedal up events. |
464 |
- All engines: Increased ramp speed of volume smoother and pan smoother |
465 |
(while slow rate of crossfade smoother is preserved) to allow quick |
466 |
volume and pan changes by instrument scripts for instance. |
467 |
- gig/sf2/sfz: Fixed aftertouch channel pressure events being processed |
468 |
properly and thus were ignored for certain uses. |
469 |
- Added new C++ API method |
470 |
VirtualMidiDevice::SendChannelPressureToSampler(). |
471 |
- windows, 32-bit: fixed potential crashes by making sure the stack in |
472 |
sub threads is 16-byte aligned |
473 |
- fixed numerous compiler warnings |
474 |
- Fixed invalid (note-on) event ID being assigned to new Note objects. |
475 |
- Revised fundamental C++ classes "Thread", "Mutex" and "Condition" which |
476 |
fixes potential undefined behavior. |
477 |
- Fixed Note object leak when triggering notes on keys which did not |
478 |
have a valid sample mapped (fixes bug #252). |
479 |
- Fixed compilation errors when compiling with CONFIG_DEVMODE enabled. |
480 |
- linuxsampler binary fix: option --create-instruments-db ignored |
481 |
subsequent optional argument due to glibc's implementation oddity |
482 |
which expects a "=" sign, but no space between them. |
483 |
|
484 |
* packaging changes: |
485 |
- removed unnecessary dependency to libuuid |
486 |
(originated by libgig's usage of it) |
487 |
- Automake: set environment variable GCC_COLORS=auto to allow GCC to |
488 |
auto detect whether it (sh/c)ould output its messages in color. |
489 |
- Debian: Fixed packaging error about invalid "Source-Version" |
490 |
substitution variable. |
491 |
- Debian: Raised Debian compatibility level to Debian 9 "Stretch". |
492 |
- Debian: Added build dependency to libsqlite3-dev for building |
493 |
linuxsampler with instruments DB support. |
494 |
|
495 |
* Real-time instrument scripts: |
496 |
- Implemented scheduler for delayed MIDI events and for suspended scripts. |
497 |
- Built-in script function "wait()": implemented support for function's |
498 |
"duration-us" argument, thus scripts using this function are now |
499 |
correctly resumed after the requested amount of microseconds. |
500 |
- Built-in script function "play_note()": implemented support for |
501 |
function's "duration-us" argument, thus notes triggered with this |
502 |
argument are now correctly released after the requested amount of |
503 |
microseconds. |
504 |
- Fix: script events were not cleared when engine channel was reset, |
505 |
potentially causing undefined behavior. |
506 |
- Fixed crash which happened when trying to reference an undeclared |
507 |
variable. |
508 |
- Built-in script function "play_note()": Added support for passing |
509 |
special value -1 for "duration-us" argument, which will cause the |
510 |
triggered note to be released once the original note was released. |
511 |
- Instrument script classes now exported with the liblinuxsampler C++ API. |
512 |
- Added new API method ScriptVM::syntaxHighlighting() which provides |
513 |
a convenient syntax highlighting backend for external instrument |
514 |
script editor applications. |
515 |
- Added new C++ API class "ScriptVMFactory". |
516 |
- Extended parser issues to provide not only first line and first |
517 |
column, but also last line and last column of issue (thus marking |
518 |
the precise span of the issue within the source code). |
519 |
- Reload script automatically after being modified by an instrument |
520 |
editor. |
521 |
- NKSP language grammar correction: allow empty event handler bodies |
522 |
like "on note end on". |
523 |
- Implemented built-in script function "change_vol()". |
524 |
- Implemented built-in script function "change_tune()". |
525 |
- Implemented built-in script function "change_pan()". |
526 |
- Implemented built-in script function "change_cutoff()". |
527 |
- Implemented built-in script function "change_reso()". |
528 |
- Implemented built-in script function "event_status()". |
529 |
- Added built-in script constants "$EVENT_STATUS_INACTIVE" and |
530 |
"$EVENT_STATUS_NOTE_QUEUE" both for being used as flags for |
531 |
"event_status()" function. |
532 |
- NKSP language: Added support for bitwise operators ".or.", ".and." |
533 |
and ".not.". |
534 |
- NKSP language scanner: Fixed IDs matching to require at least one |
535 |
character (i.e. when matching function names or variable names). |
536 |
- NKSP language scanner: disabled unusued rules. |
537 |
- Fixed behavior of play_note() and note_off() functions which must |
538 |
be distinguished engine internally from "real" MIDI note on/off |
539 |
events in order to avoid misbehaviors like hanging notes. |
540 |
- Implemented built-in script variable "$KSP_TIMER". |
541 |
- Implemented built-in script variable "$NKSP_REAL_TIMER". |
542 |
- Implemented built-in script variable "$NKSP_PERF_TIMER". |
543 |
- Implemented built-in script variable "$ENGINE_UPTIME". |
544 |
- Implemented built-in script function "inc()". |
545 |
- Implemented built-in script function "dec()". |
546 |
- NKSP language fix: division expressions were evaluated too often. |
547 |
- NKSP language fix: string concatenation operator was right |
548 |
associative instead of left (to right). |
549 |
- NKSP language correction: allow empty statements in entire language |
550 |
(i.e. "if end if"). |
551 |
- Implemented built-in script function "stop_wait()". |
552 |
- Implemented built-in script variable "$NI_CALLBACK_ID". |
553 |
- Implemented built-in script variable "$NI_CALLBACK_TYPE". |
554 |
- Implemented built-in script variable "$NKSP_IGNORE_WAIT". |
555 |
- Added support for read-only built-in variables (respectively |
556 |
handled by the script parser). |
557 |
- Added built-in script constant "$NI_CB_TYPE_INIT". |
558 |
- Added built-in script constant "$NI_CB_TYPE_NOTE". |
559 |
- Added built-in script constant "$NI_CB_TYPE_RELEASE". |
560 |
- Added built-in script constant "$NI_CB_TYPE_CONTROLLER". |
561 |
- NKSP Language: Added support for user defined script functions. |
562 |
- Implemented built-in script function "change_attack()". |
563 |
- Implemented built-in script function "change_decay()". |
564 |
- Implemented built-in script function "change_release()". |
565 |
- Fixed all change_*() built-in script functions to apply their |
566 |
synthesis parameter changes immediately in case the respective note |
567 |
was triggered at the same time, instead of scheduling the parameter |
568 |
change, especially because it would cause some parameter types's |
569 |
changes either to be ramped (i.e. change_vol()) or other types even |
570 |
to have not effect at all (i.e. change_attack()). |
571 |
- Implemented built-in script function "sh_left()". |
572 |
- Implemented built-in script function "sh_right()". |
573 |
- Implemented built-in script function "min()". |
574 |
- Implemented built-in script function "max()". |
575 |
- NKSP Fix: "init" event handler was not always executed when a script |
576 |
was loaded. |
577 |
- NKSP built-in wait() script function: abort script execution if |
578 |
a negative or zero wait time was passed as argument (since this is |
579 |
a common indication of a bug either of the script or even of the |
580 |
engine, which could lead to RT instability or even worse). |
581 |
- ScriptVM: Implemented automatic suspension of RT safety |
582 |
threatening scripts. |
583 |
- Provide more user friendly error messages on syntax errors. |
584 |
- Fixed NKSP parser warning "Not a statement" when assigning an |
585 |
initializer list to an array variable. |
586 |
- Implemented built-in script array variable "%ALL_EVENTS". |
587 |
- Implemented built-in script function "in_range()". |
588 |
- Implemented built-in script function "change_amp_lfo_depth()". |
589 |
- Implemented built-in script function "change_amp_lfo_freq()". |
590 |
- Implemented built-in script function "change_pitch_lfo_depth()". |
591 |
- Implemented built-in script function "change_pitch_lfo_freq()". |
592 |
- Implemented built-in script function "change_vol_time()". |
593 |
- Implemented built-in script function "change_tune_time()". |
594 |
- Implemented built-in script function "fade_in()". |
595 |
- Implemented built-in script function "fade_out()". |
596 |
- Fixed acceptance of wrong data type of parameters passed to built-in |
597 |
script functions "change_vol()", "change_tune()", "change_pan()", |
598 |
"change_cutoff()", "change_reso()", "change_attack()", "change_decay()", |
599 |
"change_release()", "change_amp_lfo_depth()", "change_amp_lfo_freq()", |
600 |
"change_pitch_lfo_depth()" and "change_pitch_lfo_freq()". |
601 |
- Added built-in script function "get_event_par()" and implemented some |
602 |
of its possible parameter selections. |
603 |
- Added built-in script function "set_event_par()" and implemented some |
604 |
of its possible parameter selections. |
605 |
- Fixed a bunch of scheduler time related bugs. |
606 |
- Fixed polyphonic variables not being reset to zero after usage. |
607 |
- Built-in "ignore_event()" function: argument is now optional, like with |
608 |
built-in function "ignore_controller()". |
609 |
- Implemented built-in script function "change_velo()". |
610 |
- Implemented built-in script function "change_note()". |
611 |
- Adjusted behavior of "change_vol()" and "change_tune()" to a more |
612 |
intuitive behavior if used in combination with "change_vol_time()" or |
613 |
"change_tune_time()" respectively: now tuning/volume changes are only |
614 |
assigned (without delay) immediately to a new note if the respective |
615 |
timing function has not been called before, otherwise the volume/tuning |
616 |
changes are automatically faded (before, only the event's time stamp was |
617 |
relevant). |
618 |
- Implemented built-in script function "array_equal()". |
619 |
- Implemented built-in script function "search()". |
620 |
- Implemented built-in script function "sort()". |
621 |
- NKSP Fix: Never suspend "init" event handlers. |
622 |
- Implemented built-in script function "same_region()" (currently only |
623 |
available for gig format engine). |
624 |
- Added built-in script constant "$NKSP_LINEAR". |
625 |
- Added built-in script constant "$NKSP_EASE_IN_EASE_OUT". |
626 |
- Implemented built-in script function "change_vol_curve()". |
627 |
- Implemented built-in script function "change_tune_curve()". |
628 |
- built-in "play_note()" function now supports a sample playback start |
629 |
offset with argument 3, where special value -1 means to use the regular |
630 |
sample offset as defined by the instrument file. |
631 |
- Built-in array variable %KEY_DOWN[] is now a read-only variable. |
632 |
- Built-in variable $EVENT_NOTE is now a read-only variable. |
633 |
- Built-in variable $EVENT_VELOCITY is now a read-only variable. |
634 |
- built-in "play_note()" function now accepts -2 for its fourth argument |
635 |
(note duration) which means the life time of the note shall be sticked |
636 |
to the requested note number of argument 1. |
637 |
- Fix: built-in "play_note()" function now returns 0 as result value if |
638 |
-1 was passed for its fourth argument (note duration) and the respective |
639 |
parent note is already gone. |
640 |
- Implemented built-in script function "change_play_pos()". |
641 |
- NKSP language: Added support for user declared const array variables. |
642 |
- NKSP language: Raise parser warning if array variable is accessed with |
643 |
an index that exceeds the array's size. |
644 |
- NKSP language fix: Unknown characters were not handled correctly. |
645 |
- NKSP language: Added support for "synchronized .. end synchronized" |
646 |
code blocks. |
647 |
- Implemented built-in script function "abort()" which allows to abort |
648 |
another script handler by passing its callback ID. |
649 |
- Fixed potential memory access bug and potential undefined behavior of |
650 |
"init" event handlers. |
651 |
- Print a time stamp along to each call of built-in function "message()". |
652 |
- ScriptVM API: Added VMParserContext::preprocessorComments() which allows |
653 |
to retrieve all code blocks filtered out by the preprocessor. |
654 |
- Added built-in script function "fork()". |
655 |
- Added built-in array variable %NKSP_CALLBACK_CHILD_ID[]. |
656 |
- Added built-in variable $NKSP_CALLBACK_PARENT_ID. |
657 |
- Fixed potential crash when accessing dynamic built-in array variables. |
658 |
- Added built-in script function "callback_status()". |
659 |
- Added built-in constant $CALLBACK_STATUS_TERMINATED. |
660 |
- Added built-in constant $CALLBACK_STATUS_QUEUE. |
661 |
- Added built-in constant $CALLBACK_STATUS_RUNNING. |
662 |
- Removed max. value limitation of built-in functions "change_attack()", |
663 |
"change_decay()" and "change_release()" to i.e. allow passing 2000000 |
664 |
for doubling the respective time. |
665 |
- NKSP script editor syntax highlighting API: Fixed app termination due |
666 |
to a lexer start condition stack underrun. |
667 |
- NKSP preprocessor: Fixed wrong behavior on nested USE_CODE_IF() and |
668 |
USE_CODE_IF_NOT() preprocessor statements. |
669 |
- NKSP: Added built-in preprocessor condition NKSP_NO_MESSAGE, which |
670 |
can be set to disable all subsequent built-in "message()" function calls |
671 |
on preprocessor level. |
672 |
- Implemented built-in script function "change_sustain()". |
673 |
- NKSP script editor syntax highlighting API: catch all fatal lexer errors, |
674 |
to avoid the editor app to crash on ill-formed text input. |
675 |
- Added built-in script function "change_pan_time()". |
676 |
- Added built-in script function "change_pan_curve()". |
677 |
- Added built-in script function "change_cutoff_attack()". |
678 |
- Added built-in script function "change_cutoff_decay()". |
679 |
- Added built-in script function "change_cutoff_sustain()". |
680 |
- Added built-in script function "change_cutoff_release()". |
681 |
- Added built-in script function "change_cutoff_lfo_depth()". |
682 |
- Added built-in script function "change_cutoff_lfo_freq()". |
683 |
|
684 |
* Instruments DB: |
685 |
- Fixed memory access bug of general DB access code which lead to |
686 |
undefined behavior. |
687 |
- Cleanup of instruments DB file creation and opening code. |
688 |
- The instrument DB path of linuxsampler's --create-instruments-db argument |
689 |
is now optional, if it is missing, then a default location is used. |
690 |
- Added support for scanning SFZ (.sfz) files. |
691 |
- Added support for scanning Sound Font (.sf2) files. |
692 |
- Fixed undefined DB transaction behavior. |
693 |
|
694 |
Version 2.0.0 (15 July 2015) |
695 |
|
696 |
* packaging changes: |
697 |
- fixed building with newer MinGW-w64 |
698 |
- Mac OS X: support the new dir for Core Audio SDK |
699 |
- Mac OS X: fixed building outside source directory |
700 |
- made sure all source files for hostplugins are included when |
701 |
doing "make dist" |
702 |
- removed empty directories left from the cvs to svn migration |
703 |
- fixed compilation with gcc 4.6.1 |
704 |
- another "make dist" fix, for the LV2 plugin |
705 |
- made --enable-pthread-testcancel default on Mac OS X |
706 |
- minor Makefile fix for building lscpparser when build and source |
707 |
directories are separate |
708 |
- Mac OS X: made it possible to specify plugin installation dir to |
709 |
configure |
710 |
- Mac OS X: Makefile fix for the install-strip target |
711 |
- fixed compilation with gcc 4.7 |
712 |
- fixed configure script error with old autoconf versions |
713 |
- lsatomic.h: use gcc provided atomic functions if building with |
714 |
gcc 4.7 and C++11 |
715 |
- modernized configure script |
716 |
- fixed linkage error when building with |
717 |
LDFLAGS="-Wl,--no-undefined" (#190) |
718 |
- fixed compilation with Clang 3.2 |
719 |
- removed usage of deprecated Automake variable INCLUDES |
720 |
- fixed building with C++11 |
721 |
- build fix: ChangeFlagRelaxed.h was missing in makefile |
722 |
- build fix: libsndfile compiler flags were missing in some |
723 |
makefiles |
724 |
- fix for building with bison 3.0 (#202) |
725 |
- Mac OS X: added temporary hack allowing to spawn gigedit as callback |
726 |
on the process's main thread |
727 |
- fixed build error on newer MinGW |
728 |
- support building with older jack versions |
729 |
- support building with spaces in vst sdk path |
730 |
- enabled automake 'subdir-objects' option and moved external |
731 |
source references (vst, au, asio) from makefiles to cpp files, |
732 |
in order to get rid of warnings from automake 1.14 |
733 |
|
734 |
* general changes: |
735 |
- Refactoring: moved the independent code from |
736 |
the Gigasampler format engine to base classes |
737 |
- command line option '--profile' is currently disabled, since the |
738 |
respective profiling code is currently broken |
739 |
- Introduced support for optional environment variable |
740 |
"LINUXSAMPLER_PLUGIN_DIR" which allows to override the directory |
741 |
where the sampler shall look for instrument editor plugins |
742 |
(patch by Luis Garrido, slightly modified). |
743 |
- implemented Roland GS NRPN 1ArrH which allows to set volume per note |
744 |
- implemented Roland GS NRPN 1CrrH which allows to set pan per note |
745 |
- implemented Roland GS NRPN 1DrrH which allows to set reverb send per |
746 |
note (in this implementation of the sampler its simply hard routed to |
747 |
the 1st effect send of the sampler channel, no matter what the actual |
748 |
effect type is) |
749 |
- implemented Roland GS NRPN 1ErrH which allows to set chorus send per |
750 |
note (in this implementation of the sampler its simply hard routed to |
751 |
the 2nd effect send of the sampler channel, no matter what the actual |
752 |
effect type is) |
753 |
- implemented support for internal LADSPA effects |
754 |
- raised limit of program change queue from 100 to 512 |
755 |
(as suggested by Alex Stone) |
756 |
- implemented sine LFO, pulse LFO and saw LFO |
757 |
- experimental support for per voice equalization |
758 |
- added command line option --exec-after-init |
759 |
- Introduced new C++ API method: |
760 |
EngineChannel::InstrumentFileName(int index) |
761 |
allowing to retrieve the whole list of files used for the loaded |
762 |
instrument on an engine channel (a.k.a. part). Some GigaStudio |
763 |
instruments for example are splitted over several files like |
764 |
"Foo.gig", "Foo.gx01", "Foo.gx02", ... |
765 |
- Added new C++ API method Sampler::GetGlobalMaxVoices(). |
766 |
- Added new C++ API method Sampler::GetGlobalMaxStreams(). |
767 |
- Added new C++ API method Sampler::SetGlobalMaxVoices(). |
768 |
- Added new C++ API method Sampler::SetGlobalMaxStreams(). |
769 |
- Various "const" and "restrict" optimizations. |
770 |
- all engines: add pan CC value to instrument pan parameter before |
771 |
applying panning, instead of using two separate pan functions in |
772 |
series (#182) |
773 |
- added a lock guard class for exception safe mutex handling and |
774 |
used it everywhere appropriate |
775 |
- Immediately apply scale tuning changes to active voices. |
776 |
- Exposed scale tuning to C++ API (along to the already existing standard |
777 |
SysEx way). |
778 |
- lsatomic.h fixes: seq_cst load and store were suboptimal for x86 |
779 |
and broken for ppc64. (Seq_cst loads and stores are actually not |
780 |
used in LS, so the bug wasn't noticable.) |
781 |
- lsatomic.h: added ARMv7 support |
782 |
- Added support for multiple MIDI input ports per sampler channel (and |
783 |
added various new C++ methods for this new feature / design change, old |
784 |
C++ API methods for managing SamplerChannel's MIDI inputs are now marked |
785 |
as deprecated but are still there and should provide full behavior |
786 |
backward compatibility). |
787 |
- AbstractEngine::GSChecksum(): don't allocate memory on the stack (was |
788 |
unsafe and caused compilation error with clang 2.x). |
789 |
- Bugfix: only process the latest MIDI program change event. |
790 |
- Introducing the LSCP shell, which provides convenient control of the |
791 |
sampler from the command line by providing LSCP aware features. |
792 |
- VirtualMidiDevice: Added support for program change. |
793 |
- VirtualMidiDevice: Added support for bank select (MSB & LSB). |
794 |
- VirtualMidiDevice: Added support for pitch bend. |
795 |
- Aftertouch: extended API to explicitly handle channel pressure and |
796 |
polyphonic key pressure events (so far polyphonic pressure was not |
797 |
supported at all, and channel pressure was rerouted as CC128 but not |
798 |
used so far). |
799 |
- Added initial support for real-time instrument scripts. The script VM |
800 |
code is shared by all sampler engine implemementations, however only the |
801 |
gig file format currently provides support for storing instrument scripts |
802 |
(as LinuxSampler extension to the original GigaStudio 4 format). |
803 |
|
804 |
* Gigasampler format engine: |
805 |
- implemented the "round robin keyboard" dimension |
806 |
- fixed round robin and random dimensions for cases when number of |
807 |
dimension zones is not a power of two |
808 |
- made round robin use a counter for each region instead of each |
809 |
key |
810 |
- bugfix: pitch LFO controller "internal+aftertouch" was broken |
811 |
- bugfix: filter keyboard tracking was broken |
812 |
- filter performance fix (an unnecessary copy was made of the |
813 |
filter parameters in each sub fragment) |
814 |
- handle special case when pan parameter in gig file has max or |
815 |
min value |
816 |
- Exclusive Groups: don't ever stop voices of the same note, |
817 |
doesn't sound naturally with a drumkit |
818 |
- fixed EG1 modulation when attack or release is zero |
819 |
- Fixed support for 'aftertouch' attenuation controller. |
820 |
- Fixed crash that happened with velocity split sounds under certain |
821 |
conditions (see also previous commit on libgig). |
822 |
- fixed behaviour of filter LFO |
823 |
|
824 |
* SFZ format engine: |
825 |
- Initial implementation (not usable yet) |
826 |
- added support for v2 multiple stage envelope generators |
827 |
- added a fine-tuned v1 envelope generator instead of using the |
828 |
one from the gig engine |
829 |
- fixed memory leak and memory handling errors |
830 |
- added support for trigger=first, trigger=legato and sw_previous |
831 |
- allow non-numerical key values ("C#4" for example) |
832 |
- "key" opcode now sets pitch_keycenter too |
833 |
- fixed error when unloading instrument with same sample used by |
834 |
multiple regions |
835 |
- added some opcode aliases, like loopmode for loop_mode, to be |
836 |
more compatible |
837 |
- added support for trigger=release and rt_decay |
838 |
- added support for off_mode=normal |
839 |
- added support for random, seq_position, seq_length and volume |
840 |
- added v1 LFO opcodes to parser (no support in engine yet) |
841 |
- added support for amp_veltrack and amp_velcurve_N |
842 |
- fine-tuned the default velocity curve |
843 |
- added support for transpose |
844 |
- fixed crash when using sw_down/up |
845 |
- improved logic for sw_lokey/hikey/up/down/last |
846 |
- added more v1 aliases to parser |
847 |
- reduced memory usage for sfz data |
848 |
- RT-safeness: avoid malloc in audio thread |
849 |
- fixed a bug that could cause voice stealing to fail |
850 |
- optimized sample lookup |
851 |
- improved support for exclusive groups (group, off_by and |
852 |
off_mode) |
853 |
- added support for controller triggered regions (on_locc/on_hicc) |
854 |
- added support for loop_mode=one_shot |
855 |
- fixed playback of 16 bit wav files on big endian CPUs |
856 |
- added support for Ogg Vorbis sample files |
857 |
- use loop markers from sample file if loop_start and loop_end are |
858 |
not set in sfz file |
859 |
- implemeted filters. Filter types: lowpass, bandpass, bandreject |
860 |
and highpass. 1, 2, 4 and 6 pole filters. Opcodes: fil_type, |
861 |
cutoff, resonance, fil_veltrack, fil_keytrack, fil_keycenter, |
862 |
cutoff_cc, cutoff_chanaft. |
863 |
- bugfix: zero ampeg_sustain didn't work |
864 |
- added support for velocity effect on amplifier envelope time |
865 |
(ampeg_vel2attack, ampeg_vel2decay, ampeg_vel2sustain and |
866 |
ampeg_vel2release) |
867 |
- added support for EG hold (ampeg_hold) |
868 |
- added support for sample offset (offset) |
869 |
- use the newly introduced signal units model |
870 |
- implemented opcodes lfoN_delay, |
871 |
lfoN_freq, lfoN_pan, lfoN_cutoff, lfoN_resonance |
872 |
- implemented opcodes ampeg_delay, ampeg_vel2delay |
873 |
pitcheg_delay, pitcheg_start, pitcheg_attack, pitcheg_hold, |
874 |
pitcheg_decay, pitcheg_sustain, pitcheg_release, pitcheg_vel2delay, |
875 |
pitcheg_vel2attack, pitcheg_vel2hold, pitcheg_vel2decay, |
876 |
pitcheg_vel2sustain, pitcheg_vel2release, pitcheg_depth |
877 |
- implemented opcodes pitchlfo_delay, pitchlfo_freq, |
878 |
pitchlfo_depth, fillfo_delay, fillfo_freq, fillfo_depth, |
879 |
amplfo_delay, amplfo_freq, amplfo_depth |
880 |
- implemented opcodes fileg_delay, fileg_start, fileg_attack, |
881 |
fileg_hold, fileg_decay, fileg_sustain, fileg_release, fileg_vel2delay, |
882 |
fileg_vel2attack, fileg_vel2hold, fileg_vel2decay, fileg_vel2sustain, |
883 |
fileg_vel2release |
884 |
- implemented opcode lfoN_wave |
885 |
- implemented opcode pitchlfo_depthccN |
886 |
- implemented opcodes lfoN_phase, lfoN_phase_onccX, |
887 |
lfoN_pitch, lfoN_pitch_onccX |
888 |
- implemented opcodes pitchlfo_fade, |
889 |
fillfo_fade, amplfo_fade, lfoN_fade, lfoN_fade_onccX |
890 |
- implemented opcodes fillfo_freqccN, |
891 |
pitchlfo_freqccN, amplfo_freqccN, lfoN_freq_onccX |
892 |
- implemented opcodes ampeg_delayccN, ampeg_startccN, |
893 |
ampeg_attackccN, ampeg_holdccN, ampeg_decayccN, ampeg_sustainccN, |
894 |
ampeg_releaseccN, egN_timeX_onccY, egN_levelX_onccY |
895 |
- lfoN_* and egN_* opcodes defined in group sections |
896 |
are now taken into account |
897 |
- implemented curves |
898 |
- implemented opcodes volume_onccN, volume_curveccN |
899 |
- implemented opcode volume_smoothccN |
900 |
- implemented opcodes fillfo_depthccN, amplfo_depthccN, |
901 |
lfoN_volume, lfoN_volume_onccX, lfoN_volume_smoothccX, |
902 |
lfoN_freq_smoothccX, lfoN_pitch_smoothccX, lfoN_pan_onccX, |
903 |
lfoN_pan_smoothccX, lfoN_cutoff_onccX, lfoN_cutoff_smoothccX, |
904 |
lfoN_resonance_onccX, lfoN_resonance_smoothccX, lfoN_delay_onccX |
905 |
- implemented opcode end |
906 |
- implemented opcodes egN_amplitude_onccX, |
907 |
egN_volume, egN_volume_onccX, egN_cutoff, egN_cutoff_onccX, |
908 |
egN_pitch, egN_pitch_onccX, egN_resonance, egN_resonance_onccX |
909 |
- implemented opcodes xfin_lokey, xfin_hikey, |
910 |
xfout_lokey, xfout_hikey, xf_keycurve, xfin_lovel, xfin_hivel, |
911 |
xfout_lovel, xfout_hivel, xf_velcurve, xfin_loccN, xfin_hiccN, |
912 |
xfout_loccN, xfout_hiccN, xf_cccurve |
913 |
- implemented opcodes pan_onccN, pan_smoothccN, |
914 |
pan_curveccN, egN_pan, egN_pan_curve, egN_pan_onccX, |
915 |
egN_pan_curveccX, pitch_veltrack |
916 |
- when failed to parse a sfz file |
917 |
print the line number on which the error occurs |
918 |
- use common pool of CC objects to minimize RAM usage |
919 |
- implemented opcodes amplfo_delay_onccN, |
920 |
amplfo_fade_onccN, fillfo_delay_onccN, fillfo_fade_onccN, |
921 |
pitchlfo_delay_onccN, pitchlfo_fade_onccN |
922 |
- implemented opcodes fileg_delay_onccN, |
923 |
fileg_start_onccN, fileg_attack_onccN, fileg_hold_onccN, |
924 |
fileg_decay_onccN, fileg_sustain_onccN, fileg_release_onccN, |
925 |
fileg_depth_onccN, pitcheg_delay_onccN, pitcheg_start_onccN, |
926 |
pitcheg_attack_onccN, pitcheg_hold_onccN, pitcheg_decay_onccN, |
927 |
pitcheg_sustain_onccN, pitcheg_release_onccN, pitcheg_depth_onccN |
928 |
- implemented automatic aliasing *ccN <-> *_onccN |
929 |
- *lfo_freqccN wasn't working when the respective *lfo_freq |
930 |
was not set or was set to zero |
931 |
- lfoN_freq_onccX wasn't working when lfoN_freq |
932 |
was not set or was set to zero |
933 |
- implemented opcodes resonance_onccN, resonance_smoothccN, |
934 |
resonance_curveccN, cutoff_smoothccN, cutoff_curveccN |
935 |
- implemented opcodes fillfo_depthchanaft, |
936 |
fillfo_freqchanaft, amplfo_depthchanaft, amplfo_freqchanaft, |
937 |
pitchlfo_depthchanaft, pitchlfo_freqchanaft |
938 |
- implemented opcodes pitch_onccN, |
939 |
pitch_curveccN, pitch_smoothccN, pitch_stepccN |
940 |
- implemented opcodes volume_stepccN, pan_stepccN, |
941 |
cutoff_stepccN, resonance_stepccN, lfoN_freq_stepccX, |
942 |
lfoN_volume_stepccX, lfoN_pitch_stepccX, lfoN_pan_stepccX, |
943 |
lfoN_cutoff_stepccX, lfoN_resonance_stepccX |
944 |
- implemented opcodes eq1_freq, eq2_freq, eq3_freq, |
945 |
eq1_freqccN, eq2_freqccN, eq3_freqccN, eq1_bw, eq2_bw, eq3_bw, |
946 |
eq1_bwccN, eq2_bwccN, eq3_bwccN, eq1_gain, eq2_gain, eq3_gain, |
947 |
eq1_gainccN, eq2_gainccN, eq3_gainccN |
948 |
- implemented opcodes delay, delay_onccN, delay_random, |
949 |
delay_samples, delay_samples_onccN |
950 |
- implemented opcodes egN_eq1gain, egN_eq2gain, egN_eq3gain, |
951 |
egN_eq1gain_onccX, egN_eq2gain_onccX, egN_eq3gain_onccX, egN_eq1freq, |
952 |
egN_eq2freq, egN_eq3freq, egN_eq1freq_onccX, egN_eq2freq_onccX, |
953 |
egN_eq3freq_onccX, egN_eq1bw, egN_eq2bw, egN_eq3bw, egN_eq1bw_onccX, |
954 |
egN_eq2bw_onccX, egN_eq3bw_onccX, lfoN_eq1gain, lfoN_eq2gain, |
955 |
lfoN_eq3gain, lfoN_eq1gain_onccX, lfoN_eq2gain_onccX, lfoN_eq3gain_onccX, |
956 |
lfoN_eq1gain_smoothccX, lfoN_eq2gain_smoothccX, lfoN_eq3gain_smoothccX, |
957 |
lfoN_eq1gain_stepccX, lfoN_eq2gain_stepccX, lfoN_eq3gain_stepccX, |
958 |
lfoN_eq1freq, lfoN_eq2freq, lfoN_eq3freq, lfoN_eq1freq_onccX, |
959 |
lfoN_eq2freq_onccX, lfoN_eq3freq_onccX, lfoN_eq1freq_smoothccX, |
960 |
lfoN_eq2freq_smoothccX, lfoN_eq3freq_smoothccX, lfoN_eq1freq_stepccX, |
961 |
lfoN_eq2freq_stepccX, lfoN_eq3freq_stepccX, lfoN_eq1bw, lfoN_eq2bw, |
962 |
lfoN_eq3bw, lfoN_eq1bw_onccX, lfoN_eq2bw_onccX, lfoN_eq3bw_onccX, |
963 |
lfoN_eq1bw_smoothccX, lfoN_eq2bw_smoothccX, lfoN_eq3bw_smoothccX, |
964 |
lfoN_eq1bw_stepccX, lfoN_eq2bw_stepccX, lfoN_eq3bw_stepccX |
965 |
- implemented opcodes eq1_vel2freq, eq2_vel2freq, |
966 |
eq3_vel2freq, eq1_vel2gain, eq2_vel2gain, eq3_vel2gain |
967 |
- sfz parser: allow double spaces in sample filenames |
968 |
- sfz parser: allow absolute paths for sample filenames |
969 |
- use linear decay and release for filter and pitch EG |
970 |
- bugfix: only the first amp_veltrack definition in a file was |
971 |
used |
972 |
- bugfix: looping was disabled if loop_start was set to 0 |
973 |
- allow regions with end=-1 to turn off other regions using the |
974 |
group and off_by opcodes (#168) |
975 |
- made end=0 play the whole sample |
976 |
- fixed support for lochan and hichan opcodes (#155) |
977 |
- fixed crash when using lochan/hichan opcodes (#187) |
978 |
- sfz parser: allow -200 to 200 for pan_oncc opcode (#182) |
979 |
- added FLAC support (#191) |
980 |
- sfz parser bugfix: lines starting with whitespace were ignored |
981 |
- added amplitude opcode |
982 |
- added support for "#include" instruction |
983 |
(modified patch which was originally posted by Sergey on LS mailing list) |
984 |
- bugfix: generation of velocity curves etc should not be done |
985 |
after each "#include", only after the main file is parsed |
986 |
- bugfix: line numbers in error messages were wrong after "#include" |
987 |
- added support for float and 32 bit sample files |
988 |
|
989 |
* SoundFont format engine: |
990 |
- Initial implementation (not usable yet) |
991 |
- RT-safeness: avoid malloc in audio thread |
992 |
- fixed a bug that could cause voice stealing to fail |
993 |
- fine-tuned amplitude EG (by switching from gig to sfz EG) |
994 |
- initial implementation of Vibrato LFO and Modulation LFO |
995 |
- initial implementation of cutoff filter |
996 |
- use linear decay and release for filter and pitch EG |
997 |
|
998 |
* Host plugins (VST, AU, LV2, DSSI): |
999 |
- AU bugfix: failed to destroy its audio/MIDI devices |
1000 |
- Listen to all interfaces on Mac OS X (INADDR_ANY) |
1001 |
- VST bugfix: If the host called resume() before and after |
1002 |
changing sample rate or block size, the number of channels was |
1003 |
incorrectly set to two. This caused silence in Cubase 5. |
1004 |
- save engine type (gig, sfz or sf2) in plugin state |
1005 |
- VST: when opening Fantasia, look for both 32 and 64 bit Java on |
1006 |
64 bit Windows |
1007 |
- AU: changed number of output channels from one stereo to 16 |
1008 |
stereo |
1009 |
- VST: made it possible to build the VST plugin for Mac |
1010 |
- AU: link AU plugin dynamically if --disable-shared isn't |
1011 |
specified |
1012 |
- LV2 "state" extension support (patch by David Robillard) |
1013 |
- VST bugfix: instrument loading hang and crashed the host when |
1014 |
the plugin was loaded a second time (#174) |
1015 |
- plugin bugfix: instrument loading hang when the plugin was |
1016 |
loaded a second time (this time it's for Linux and Mac, previous |
1017 |
similar fix was for Windows) |
1018 |
- thread safety fixes for the instrument loading thread |
1019 |
- LV2: use the new lv2 package if present |
1020 |
- VST: try to open Fantasia automatically on Linux and Mac too (on |
1021 |
Linux, the Fantasia jar should be placed in <prefix>/share/java) |
1022 |
- VST: fixed crashes on Linux Ardour and EnergyXT |
1023 |
- DSSI bugfix: it wasn't possible to change engine type. The MIDI |
1024 |
port and audio channel routing for DSSI plugins are now visible. |
1025 |
- LV2: use urid and atom extensions instead of deprecated uri-map |
1026 |
and event |
1027 |
- LV2: lv2 package 1.0 is now required to build the LV2 plugin |
1028 |
- LV2: changed number of output channels to 16 stereo, just like |
1029 |
the VST and AU plugins |
1030 |
- LV2: fixed save/restore of SFZ state (patch by David Robillard) |
1031 |
- LV2: made LV2 plugin buildable on Windows and Mac |
1032 |
- VST: implemented retrieval and switching of programs using the |
1033 |
sampler's internal MIDI instrument mapping system |
1034 |
|
1035 |
* Instrument editor interface: |
1036 |
- Changed instrument editor plugin interface, providing additional |
1037 |
informations like the EngineChannel for which the instrument editor was |
1038 |
spawned for. This allows the instrument editors to interact more actively |
1039 |
with the sampler. |
1040 |
|
1041 |
* MIDI driver: |
1042 |
- ALSA MIDI driver supports now "NAME" device parameter, for overriding |
1043 |
the ALSA sequencer client name |
1044 |
- removed limit of maximum amount of MIDI ports per MIDI device, since |
1045 |
there is no reason for this limit |
1046 |
- MME: fixed memory handling bug found with cppcheck |
1047 |
- MME: removed compiler warning |
1048 |
- CoreMIDI: implemented driver specific port parameter "CORE_MIDI_BINDINGS", |
1049 |
which allows to retrieve the list of CoreMIDI clients / ports and to connect |
1050 |
to them a la JACK, via the usual sampler APIs |
1051 |
- CoreMIDI: added driver specific port parameter "AUTO_BIND", if enabled |
1052 |
the driver will automatically connect to other CoreMIDI clients' ports |
1053 |
(e.g. external MIDI devices being attached to the Mac) |
1054 |
- added support for MIDI note on velocity filter |
1055 |
- CoreMIDI: fixed memory deallocation error |
1056 |
- Fixed variable underflow in VirtualMidiDevice, which caused graphical |
1057 |
virtual keyboards in frontends / instrument editors being stuck. |
1058 |
- Bugfix in VirtualMidiDevice: process note on with velocity 0 as note off. |
1059 |
- Implemented missing handling of MIDI "running status". |
1060 |
- CoreMIDI fix: a MIDIPacket can contain more than one event per packet. |
1061 |
- MME bugfix: driver wasn't closed properly |
1062 |
- CoreMIDI: automatically connect to all input sources by default (driver |
1063 |
parameter "AUTO_BIND"). |
1064 |
- CoreMIDI: fixed auto bind feature to CoreMIDI ports that go online |
1065 |
- CoreMIDI: fixed minor memory leak |
1066 |
|
1067 |
* audio driver: |
1068 |
- ASIO driver fixes for newer gcc versions (fix from PortAudio) |
1069 |
- JACK audio: react on sample rate changes. |
1070 |
- JACK audio: react on buffer size changes. |
1071 |
- JACK audio: jack_port_get_buffer() was cached and called outside |
1072 |
RT context. |
1073 |
- ASIO driver: removed compiler warnings |
1074 |
- CoreAudio: fixed minor error handling bug |
1075 |
- ASIO driver: be more verbose when no ASIO card could be found (fixes #203) |
1076 |
- JACK audio: return the JACK server's current sample rate as default value |
1077 |
for audio device parameter "SAMPLERATE" (fixes #166). |
1078 |
|
1079 |
* LSCP server: |
1080 |
- added support for sending MIDI CC messages via LSCP command |
1081 |
"SEND CHANNEL MIDI_DATA CC <sampler-chan> <ctrl> <val>" |
1082 |
- added LSCP command "GET AVAILABLE_EFFECTS" |
1083 |
- added LSCP command "LIST AVAILABLE_EFFECTS" |
1084 |
- added LSCP command "GET EFFECT INFO <effect-index>" |
1085 |
- added LSCP command "CREATE EFFECT_INSTANCE <effect-index>" |
1086 |
- added LSCP command |
1087 |
"CREATE EFFECT_INSTANCE <effect-system> <module> <effect-name>" |
1088 |
- added LSCP command "DESTROY EFFECT_INSTANCE <effect-instance>" |
1089 |
- added LSCP command "GET EFFECT_INSTANCES" |
1090 |
- added LSCP command "LIST EFFECT_INSTANCES" |
1091 |
- added LSCP command "GET EFFECT_INSTANCE INFO <effect-instance>" |
1092 |
- added LSCP command |
1093 |
"GET EFFECT_INSTANCE_INPUT_CONTROL INFO <effect-instance> <input-control>" |
1094 |
- added LSCP command "SET EFFECT_INSTANCE_INPUT_CONTROL VALUE |
1095 |
<effect-instance> <input-control> <value>" |
1096 |
- added LSCP command "GET SEND_EFFECT_CHAINS <audio-device>" |
1097 |
- added LSCP command "LIST SEND_EFFECT_CHAINS <audio-device>" |
1098 |
- added LSCP command "ADD SEND_EFFECT_CHAIN <audio-device>" |
1099 |
- added LSCP command |
1100 |
"REMOVE SEND_EFFECT_CHAIN <audio-device> <effect-chain>" |
1101 |
- added LSCP command |
1102 |
"GET SEND_EFFECT_CHAIN INFO <audio-device> <effect-chain>" |
1103 |
- added LSCP command "APPEND SEND_EFFECT_CHAIN EFFECT <audio-device> |
1104 |
<effect-chain> <effect-instance>" |
1105 |
- added LSCP command "INSERT SEND_EFFECT_CHAIN EFFECT <audio-device> |
1106 |
<effect-chain> <effect-chain-pos> <effect-instance>" |
1107 |
- added LSCP command "REMOVE SEND_EFFECT_CHAIN EFFECT <audio-device> |
1108 |
<effect-chain> <chain-pos>" |
1109 |
- added LSCP command "SET FX_SEND EFFECT <sampler_channel> |
1110 |
<fx_send_id> <effect_chain> <chain_pos>" |
1111 |
- added LSCP command "REMOVE FX_SEND EFFECT <sampler_channel> <fx_send_id>" |
1112 |
- added LSCP commands "SUBSCRIBE EFFECT_INSTANCE_COUNT", |
1113 |
"SUBSCRIBE EFFECT_INSTANCE_INFO", "SUBSCRIBE SEND_EFFECT_CHAIN_COUNT", |
1114 |
"SUBSCRIBE SEND_EFFECT_CHAIN_INFO" |
1115 |
- provide comprehensive error messages on LSCP syntax errors |
1116 |
(suggesting expected next non-terminal symbols) |
1117 |
- Fixed client connection not being closed after network errors. |
1118 |
|
1119 |
* LSCP shell: |
1120 |
- Added support for auto correction of obvious and trivial syntax mistakes. |
1121 |
- Added support for auto completion by tab key. |
1122 |
- Show currently available auto completion while typing. |
1123 |
- Added support for browsing command history with up / down keys. |
1124 |
- Show all possible next symbols immediately right to the current command |
1125 |
line while typing (no double tab required for this feature, as it would |
1126 |
be the case in other shells). |
1127 |
- Added support for moving cursor left/right with arrow keys. |
1128 |
- Added support for built-in LSCP reference documentation, which will |
1129 |
automatically show the relevant LSCP reference section on screen as soon |
1130 |
as one specific LSCP command was detected while typing on the command |
1131 |
line. |
1132 |
|
1133 |
* Real-time instrument scripts: |
1134 |
- Implemented built-in script array variable %CC. |
1135 |
- Implemented built-in script int variable $CC_NUM. |
1136 |
- Implemented built-in script int variable $EVENT_NOTE. |
1137 |
- Implemented built-in script int variable $EVENT_VELOCITY. |
1138 |
- Implemented built-in script constant variable $VCC_MONO_AT. |
1139 |
- Implemented built-in script constant variable $VCC_PITCH_BEND. |
1140 |
- Implemented execution of script event handler "init". |
1141 |
- Implemented execution of script event handler "controller". |
1142 |
- Implemented execution of script event handler "note". |
1143 |
- Implemented execution of script event handler "release". |
1144 |
- Implemented built-in script function "play_note()" (only two of the |
1145 |
max. four function arguments are currently implemented yet though). |
1146 |
- Implemented built-in script int variable $EVENT_ID. |
1147 |
- Implemented built-in script function "ignore_event()" |
1148 |
- Implemented built-in script function "ignore_controller()" (may have one |
1149 |
or no argument). |
1150 |
- Implemented built-in script function "set_controller()". |
1151 |
- Added extended script VM for the Gigasampler/GigaStudio format sampler |
1152 |
engine, which extends the general instrument script VM with Giga format |
1153 |
specific variables and functions. |
1154 |
- Giga format scripts: added built-in script int constant variables |
1155 |
$GIG_DIM_CHANNEL, $GIG_DIM_LAYER, $GIG_DIM_VELOCITY, $GIG_DIM_AFTERTOUCH, |
1156 |
$GIG_DIM_RELEASE, $GIG_DIM_KEYBOARD, $GIG_DIM_ROUNDROBIN, $GIG_DIM_RANDOM, |
1157 |
$GIG_DIM_SMARTMIDI, $GIG_DIM_ROUNDROBINKEY, $GIG_DIM_MODWHEEL, |
1158 |
$GIG_DIM_BREATH, $GIG_DIM_FOOT, $GIG_DIM_PORTAMENTOTIME, $GIG_DIM_EFFECT1, |
1159 |
$GIG_DIM_EFFECT2, $GIG_DIM_GENPURPOSE1, $GIG_DIM_GENPURPOSE2, |
1160 |
$GIG_DIM_GENPURPOSE3, $GIG_DIM_GENPURPOSE4, $GIG_DIM_SUSTAIN, |
1161 |
$GIG_DIM_PORTAMENTO, $GIG_DIM_SOSTENUTO, $GIG_DIM_SOFT, |
1162 |
$GIG_DIM_GENPURPOSE5, $GIG_DIM_GENPURPOSE6, $GIG_DIM_GENPURPOSE7, |
1163 |
$GIG_DIM_GENPURPOSE8, $GIG_DIM_EFFECT1DEPTH, $GIG_DIM_EFFECT2DEPTH, |
1164 |
$GIG_DIM_EFFECT3DEPTH, $GIG_DIM_EFFECT4DEPTH, $GIG_DIM_EFFECT5DEPTH. |
1165 |
- Giga format scripts: Implemented built-in script function |
1166 |
"gig_set_dim_zone(event_id, dimension, zone)". |
1167 |
- Implemented built-in script int array variable %KEY_DOWN. |
1168 |
- Implemented built-in script function "abs()". |
1169 |
- Implemented built-in script function "random()". |
1170 |
- Implemented built-in script function "num_elements()". |
1171 |
- Implemented built-in script function "note_off()". |
1172 |
- Implemented built-in script function "set_event_mark()". |
1173 |
- Implemented built-in script function "delete_event_mark()". |
1174 |
- Implemented built-in script function "by_marks()". |
1175 |
- Added built-in script int const variables $MARK_1 to $MARK_28. |
1176 |
- Built-in script functions "ignore_event()", "note_off()" and |
1177 |
"gig_set_dim_zone()" now also accept an array of event IDs as argument |
1178 |
(i.e. return value of new script function "by_marks()"). |
1179 |
- Pass/preserve polyphonic variable data from respective "note" event |
1180 |
handler to "release" event handler. |
1181 |
- Fixed crash when using built-in script function "by_marks()". |
1182 |
|
1183 |
* Bug fixes: |
1184 |
- Fixed crash which may occur when MIDI key + transpose is out of range |
1185 |
- minor valgrind fixes |
1186 |
- fixed crash which occurred when changing an already deployed sampler |
1187 |
channel to a different engine type |
1188 |
- fixed crash when deleting a sampler channel or changing engine |
1189 |
type while an instrument load was in progress |
1190 |
- bugfix: playing a note while changing the instrument could cause |
1191 |
a crash, or give "Handing back unknown region" error messages |
1192 |
- bugfix: calling SET VOICES (which Fantasia does on start and |
1193 |
refresh) could cause instruments to be unloaded |
1194 |
- Mac OS X: fixed name collision of enum in EffectControl and |
1195 |
wrong name of destructor in AudioOutputDeviceCoreAudio.cpp |
1196 |
- Mac OS X: fixed hanging threads |
1197 |
- Windows: disabled the previous thread fix on non-Mac systems, as |
1198 |
it caused hanging threads on Windows |
1199 |
- Fixed possible crashes due to corrupted MIDI/audio device list |
1200 |
after MIDI/audio device creation failure |
1201 |
- When creating MIDI instrument map entries with "PERSISTENT" type, the |
1202 |
instruments were uselessly precached with zero samples, however it still |
1203 |
took the full preloading time and on 1st program change the respective |
1204 |
instrument was completely reloaded again. |
1205 |
- fixed handling of rapid bank select and program change messages |
1206 |
sent to the same sampler channel (patch from the Open Octave |
1207 |
project, slightly adjusted) |
1208 |
- fixed crash when trying to create an effect instance with controls |
1209 |
which min and/or max values depend on the sample rate |
1210 |
- fixed bug #162 |
1211 |
- bugfix: LADSPA_PATH was not evaluated correctly when containing |
1212 |
multiple paths (#165) |
1213 |
- thread safety fixes for the instrument loading thread |
1214 |
- bugfix: instrument loading crashed for sfz and sf2 in Ardour |
1215 |
(#176) |
1216 |
- more thread safety fixes for the instrument loading thread |
1217 |
- sfz/sf2 engine: fixed crash when using small audio fragment size |
1218 |
- Mac OS X: fixed crash when unloading plugin on 10.7 and later |
1219 |
- Mac OS X: fixed process hang when unloading 32-bit plugin (bug |
1220 |
introduced in previous fix) |
1221 |
- fixed crash when a channel received a program change while |
1222 |
playing a note in a key group |
1223 |
- fixed erroneous error message piping in VoiceBase.h |
1224 |
(labelled "Disk stream not available in time") |
1225 |
- Update effects on sample rate & period size changes (to avoid |
1226 |
crashes and noise on such transitions). |
1227 |
- ignore missing LADSPA paths without ignoring valid LADSPA paths |
1228 |
(fixes #208) |
1229 |
- Fixed bug in LSCP grammar definition which caused a statement like |
1230 |
"GET SERVER INFOasdf\n" to be accepted as valid statement (was so far |
1231 |
practically irrelevant, however it caused problems with the new LSCP |
1232 |
shell's auto completion feature). |
1233 |
- Fixed MIDI program change messages being ignored if quickly executed |
1234 |
after each other (fixes #231). |
1235 |
|
1236 |
Version 1.0.0 (31 July 2009) |
1237 |
|
1238 |
* packaging changes: |
1239 |
- autoconf bugfix: the PKG_CONFIG variable wasn't initialized properly, |
1240 |
causing e.g. the libgig test to fail when |
1241 |
"./configure --disable-jack-driver" was used |
1242 |
(patch by Alexis Ballier) |
1243 |
- fixed compilation with gcc 4.3 |
1244 |
- fixes for building on OS X (thanks to Ebrahim Mayat for testing) |
1245 |
- fixed configure so it detects x86_64 (#107) |
1246 |
- fixes for building with newer MinGW versions |
1247 |
- fix for building with bison 2.4 (#111) |
1248 |
- fixed building with libgig installed in a non-standard directory |
1249 |
- minor fix in configure for mmsystem.h detection on MinGW |
1250 |
- Windows: look for editor plugins and Fantasia using base |
1251 |
directory of liblinuxsampler dll (look in the same directory and one |
1252 |
directory above) |
1253 |
- configure script fix: removed unconditional use of SSE |
1254 |
- fixed building with sqlite installed in a non-standard directory |
1255 |
- when cross-compiling, don't try to create instruments.db |
1256 |
- fix for new mingw-w64 version, which has usleep |
1257 |
|
1258 |
* general changes: |
1259 |
- bugfix: on some POSIX systems instrument editor plugins refused to |
1260 |
load as we used a non-portable Linux specific struct field |
1261 |
(fixes bug #70, patch by Ronald Baljeu) |
1262 |
- fixed endless loop which occured when loading an instrument editor |
1263 |
plugin DLL on Windows systems failed |
1264 |
- fixed memory leaks that occurred when liblinuxsampler was unloaded |
1265 |
- optimized the SynchronizedConfig class so it doesn't wait |
1266 |
unnecessarily long after an update |
1267 |
- added support for notifying instrument editors on note-on / note-off |
1268 |
events (e.g. to highlight the pressed keys on the virtual keyboard |
1269 |
of gigedit) |
1270 |
- added support for triggering notes by instrument editors (see above) |
1271 |
- be verbose on DLL load errors (on Linux) |
1272 |
- fixed di-harmonic triangle LFO implementation (this LFO implementation |
1273 |
is only used by very few systems, most pick the int math |
1274 |
implementation, which is usually faster) |
1275 |
- fixes for audio drivers with varying buffer sizes |
1276 |
- experimental support for running LinuxSampler as a DSSI, LV2 and |
1277 |
VST plugin |
1278 |
- notification events for stream/voice count statistics are now sent |
1279 |
only when there are actual changes |
1280 |
- added memory ordering constraints to improve stability on |
1281 |
multi-core and multi-cpu systems |
1282 |
- maximum voices and disk streams can now be altered at runtime |
1283 |
- fixed CPU feature detection on x86_64 (maybe fixes #108) |
1284 |
- automatic stacktrace mechanism is now turned off by default and can |
1285 |
be switched on by command line option "--stacktrace" (the automatic |
1286 |
stacktrace mechanism seems to be broken on most systems at the moment) |
1287 |
- C++ API method InstrumentManager::LaunchInstrumentEditor() now returns |
1288 |
a pointer to the launched InstrumentEditor object |
1289 |
- added optional 3rd party user data parameter for following |
1290 |
liblinuxsampler C++ API methods: InstrumentEditor::Main(), |
1291 |
InstrumentEditor::Launch(), |
1292 |
InstrumentManager::LaunchInstrumentEditor() |
1293 |
- theoretical fix: made SynchronizedConfig follow C++0x memory |
1294 |
model more strictly |
1295 |
- fixes for using large audio device buffers |
1296 |
- Windows: add the installation directory to the DLL search path |
1297 |
when loading an editor plugin (solves problems with VST and |
1298 |
gigedit on systems with other GTK versions installed) |
1299 |
- updated linuxsampler man page |
1300 |
|
1301 |
* audio driver: |
1302 |
- removed the nonsense audio channel constraint (which was hard coded to |
1303 |
max. 100 audio channels) for most audio drivers |
1304 |
- JACK audio driver did not offer a device parameter "SAMPLERATE" as |
1305 |
opposed to the LSCP specs |
1306 |
- bugfix: the SAMPLERATE parameter of some drivers (e.g. JACK) |
1307 |
reflected the wrong value |
1308 |
- fixed a memory management bug in ASIO driver |
1309 |
- Makefile fix: JACK_CFLAGS wasn't used |
1310 |
- JACK: use jack_client_open instead of the deprecated |
1311 |
jack_client_new |
1312 |
- added (experimental) CoreAudio driver |
1313 |
- applied old fixes to the ASIO driver that were included in the |
1314 |
previous binary release but accidentally never committed to CVS |
1315 |
(fixes #117) |
1316 |
- fixes for ASIO on mingw-w64 (iasio wrapper is not needed on |
1317 |
win64) |
1318 |
- VST: added support for sample rate and buffer size changes |
1319 |
- VST: close editor (Fantasia) when the VST is removed |
1320 |
- VST: avoid opening Fantasia more than once for each VST instance |
1321 |
- VST: export main function as "main" on Linux too (fix for |
1322 |
energyXT) |
1323 |
- VST: changed number of output channels from one stereo to 16 |
1324 |
stereo |
1325 |
- added channel routing, fxsends and midi maps to the settings |
1326 |
stored in the plugin state |
1327 |
- performance optimization of AudioChannel::MixTo() and |
1328 |
AudioChannel::CopyTo() methods using GCC vector exensions |
1329 |
(if available) |
1330 |
- ASIO fixes: avoid initializing the device twice, avoid throwing |
1331 |
exception when getting parameters from a disconnected device |
1332 |
|
1333 |
* MIDI driver: |
1334 |
- added JACK MIDI driver |
1335 |
- dispatch bank select as ordinary CC as well, the user might seriously |
1336 |
want to (mis)use it for some purpose ("fixed" in all current MIDI |
1337 |
input drivers: ALSA, CoreMIDI, JACK, MidiShare, MME) |
1338 |
- bugfix: pitch bend wasn't working with JackMidi, VST, LV2, MME, |
1339 |
CoreMidi or AU |
1340 |
- fixed mingw-w64 compilation error in MME driver |
1341 |
- made program change handling in MIDI thread real-time safe by |
1342 |
moving the logic to a non-RT thread |
1343 |
- fixed minor memory leak in ALSA driver |
1344 |
|
1345 |
* instruments database: |
1346 |
- avoid time consuming samples scanning when adding instruments |
1347 |
to the instruments database |
1348 |
- added support for handling lost files in the instruments database |
1349 |
- Implemented option for adding instruments in separate directories |
1350 |
in the instruments database |
1351 |
(patch by Chris Cherrett & Andrew Williams, a bit adjusted) |
1352 |
- work-around for missing fnmatch function on Windows to make |
1353 |
instrument database compilable |
1354 |
- added instrument database support on Windows |
1355 |
works with both standalone mode and VST plugin. |
1356 |
the instruments DB file is located in |
1357 |
%USERPROFILE%\.linuxsampler\instruments.db which allows different |
1358 |
databases for each windows user |
1359 |
if no DB is present it automatically creates the .linuxsampler subdir |
1360 |
and then creates an empty DB |
1361 |
- fixed recursive import, which was broken on Windows |
1362 |
|
1363 |
* Gigasampler format engine: |
1364 |
- fixed a memory leak that could happen when a channel was deleted |
1365 |
while notes were playing |
1366 |
- made it possible to load an instrument even if the audio thread |
1367 |
isn't running |
1368 |
- added partial support for the "Controller Triggered" MIDI rule, |
1369 |
enough for piano gigs with pedal noise samples |
1370 |
- minor fix: only mark FX sends as being modified if really the |
1371 |
respective FX send MIDI controller was used |
1372 |
- added support for GM global device volume SysEx message |
1373 |
(can be selected at compile time to either apply globally to the |
1374 |
whole sampler [default] or only to the sampler channels that are |
1375 |
connected to the respective MIDI input port the SysEx message arrived |
1376 |
on) |
1377 |
- bugfix: notes triggered at position 0 in the audio buffer were |
1378 |
sometimes wrongly killed in the same buffer, causing no sound to |
1379 |
be played |
1380 |
- added support for chromatic / drumkit mode Roland GS Sysex message |
1381 |
which is usally used to switch a part between chromatic sounds and |
1382 |
drumkit sounds (as opposed to their common default setting of MIDI |
1383 |
part 10 being a drumkit part and all other ones chromatic parts), in |
1384 |
LS however one can switch between the first 16 MIDI instrument maps |
1385 |
defined for the sampler |
1386 |
- made it possible to create multiple sample channels even if the |
1387 |
audio thread isn't running |
1388 |
- fixed crash when removing channel with active voices (#116) |
1389 |
- bugfix: on sample reference changes (instrument editor), only |
1390 |
un-cache the respective sample if it's really not used by any |
1391 |
sampler engine anymore |
1392 |
- re-cache samples in case they were changed by an instrument editor, |
1393 |
e.g. when a sample was added while playing (#82) |
1394 |
- fixed hanging notes which occured when note-off event had the exact |
1395 |
same time stamp as the voice's note-on event and occured both in the |
1396 |
same audio fragment cycle (fixes bug #112) |
1397 |
- added support for the "fine tune" and "pitch bend range" |
1398 |
instrument-level gig parameters |
1399 |
- fixed minor artifacts in pitch bend handling |
1400 |
- added support for GS Reset SysEx message |
1401 |
- allow gig files to use unlimited downward pitch shifting |
1402 |
- added a limit check for upward pitch shifting |
1403 |
- bugfix: sometimes, when playing a note twice fast, the second |
1404 |
note was silent |
1405 |
- fixed crash happening when a pitch bend event arrived at the |
1406 |
same time a new instrument was loading |
1407 |
|
1408 |
* LSCP server: |
1409 |
- added new LSCP event "CHANNEL_MIDI" which can be used by frontends to |
1410 |
react on MIDI data arriving on certain sampler channels (so far only |
1411 |
Note-On and Note-Off events are sent via this LSCP event) |
1412 |
- added new LSCP event "DEVICE_MIDI" which can be used by frontends to |
1413 |
react on MIDI data arriving on certain MIDI input devices (so far only |
1414 |
Note-On and Note-Off events are sent via this LSCP event) |
1415 |
- added new LSCP commands: FIND LOST DB_INSTRUMENT_FILES and |
1416 |
SET DB_INSTRUMENT FILE_PATH |
1417 |
- added new LSCP command: SEND CHANNEL MIDI_DATA which can be used by |
1418 |
frontends to send MIDI messages to specific sampler channel |
1419 |
- added two additional fields to GET FILE INSTRUMENT INFO command - |
1420 |
KEY_BINDINGS and KEYSWITCH_BINDINGS |
1421 |
- bugfix: the bank number provided by MIDI_INSTRUMENT_INFO notifications |
1422 |
was incorrect |
1423 |
- Optimized the retrieval of the MIDI instrument mappings |
1424 |
- added new LSCP command "SET VOICES" to globally alter the maximum |
1425 |
amount of voices, added new LSCP event "GLOBAL_INFO:VOICES" which will |
1426 |
be triggered respectively |
1427 |
- added new LSCP command "SET STREAMS" to globally alter the maximum |
1428 |
amount of disk streams, added new LSCP event "GLOBAL_INFO:STREAMS" |
1429 |
which will be triggered respectively |
1430 |
- bugfix: retry if "select" returns EINTR (this fixes a crash when |
1431 |
a gigedit file dialog is opened) |
1432 |
- close all connections when LSCPServer is deleted |
1433 |
- hard close of all sockets on exit |
1434 |
- bugfix: SET CHANNEL MIDI_INPUT_TYPE didn't work with the MME |
1435 |
driver |
1436 |
|
1437 |
* Bug fixes: |
1438 |
- fixed a crash which occurs when removing a sampler channel waiting |
1439 |
to start instrument loading after another channel |
1440 |
- fixed a crash which occurs when removing a sampler channel with |
1441 |
instrument loading in progress (bug #113) |
1442 |
- fixed termination caused by uncaught exception when adding MIDI |
1443 |
instrument with PERSISTENT load mode |
1444 |
- fixed possible iterator invalidations when resetting the sampler |
1445 |
- fixed memory leaks when issuing the following LSCP commands: |
1446 |
GET AUDIO_OUTPUT_DRIVER INFO |
1447 |
GET AUDIO_OUTPUT_DRIVER_PARAMETER INFO |
1448 |
GET MIDI_INPUT_DRIVER INFO |
1449 |
GET MIDI_INPUT_DRIVER_PARAMETER INFO |
1450 |
- fixed possible compilation error when sqlite is not present |
1451 |
- fixed orphaned pointers when setting maximum voices limit (bug #118) |
1452 |
- fixed crash when changing the audio output device of a sampler |
1453 |
channel with loaded instrument and start playing notes |
1454 |
- fixed endless loop in Engine::SuspendAll() (bug #120) |
1455 |
- fixed a low-level atomic load function that was broken on 64-bit |
1456 |
PowerPC, which probably could cause crashes on that platform |
1457 |
- fixed a memory management error which could cause a crash when a |
1458 |
plugin was unloaded |
1459 |
- bugfix: two private structs had the same name, which could cause |
1460 |
problems if the linker chose the wrong constructor |
1461 |
- fixed low-level ConditionServer usage bug that caused lockups on |
1462 |
Windows |
1463 |
|
1464 |
|
1465 |
Version 0.5.1 (6 December 2007) |
1466 |
|
1467 |
* packaging changes: |
1468 |
- added autoconf checks for pthread library |
1469 |
- added autoconf check for pthread bug found on certain NPTL-enabled |
1470 |
glibc versions (see Gentoo bug report #194076) |
1471 |
- added autoconf checks for MS Windows |
1472 |
|
1473 |
* general changes: |
1474 |
- bugfix: the thread used by an editor plugin didn't die when the |
1475 |
editor closed |
1476 |
- bugfix: Ringbuffer.h: fill_write_space_with_null() did not zero |
1477 |
out all the space. operator--() did not apply size_mask after |
1478 |
decrementing the read_ptr. DEFAULT_WRAP_ELEMENTS set to 0 in |
1479 |
order to avoid problems with the _NonVolatileReader functions. |
1480 |
- bugfix: Stream.h: added a missing divide by BytesPerSample in |
1481 |
GetWriteSpace(). Since this function is currently only used in |
1482 |
the stream's qsort() compare function, it didn't trigger any |
1483 |
bugs. |
1484 |
- Resampler.h, Synthesizer.h: cubic interpolation now works in |
1485 |
24bit mode too. Faster method to read 24bit words on little |
1486 |
endian machines (x86): replaced 3 byte reads + shifts with a 1 |
1487 |
unaligned 32bit read and shift |
1488 |
- experimental support for MS Windows (MIDI input via MME, AUDIO |
1489 |
output via ASIO) |
1490 |
- made handling of SIGINT signal (Ctrl-C) a bit more robust |
1491 |
- support for monitoring the total number of active disk streams |
1492 |
(new LSCP commands: GET TOTAL_STREAM_COUNT, |
1493 |
SUBSCRIBE TOTAL_STREAM_COUNT, UNSUBSCRIBE TOTAL_STREAM_COUNT) |
1494 |
|
1495 |
* AUDIO driver: |
1496 |
- added Windows ASIO low latency audio driver |
1497 |
|
1498 |
* MIDI driver: |
1499 |
- added MME Windows MIDI driver |
1500 |
|
1501 |
* LSCP server: |
1502 |
- added support for Windows style path / filenames, however with |
1503 |
forward slash path separators instead of backslash |
1504 |
(i.e. "C:/foo/bar.gig") |
1505 |
- allow naughty liblscp to send non-string device parameters within |
1506 |
apostrophes as well |
1507 |
- added new LSCP commands: "GET FILE INSTRUMENTS <file>", |
1508 |
"LIST FILE INSTRUMENTS <file>" and |
1509 |
"GET FILE INSTRUMENT INFO <file> <index>" for retrieving informations |
1510 |
about an arbitrary instrument file on the system where the sampler is |
1511 |
running on |
1512 |
|
1513 |
Version 0.5.0 (15 October 2007) |
1514 |
|
1515 |
* packaging changes: |
1516 |
- config.h is not going to be installed along with liblinuxsampler's |
1517 |
API header files anymore |
1518 |
- only the API relevant header (and source) files will be exported to |
1519 |
the Doxygen API documentation (explicitly listed in Doxyfile.in) |
1520 |
- added completely new XCode project files for Mac OSX which is now |
1521 |
capable to execute our autoconf environment, thus no need anymore |
1522 |
to maintain the compile time configuration file (osx/version.h) for |
1523 |
OSX manually (patch by Toshi Nagata) |
1524 |
- fixed buggy boolean --enable-foo / --disable-foo configure script |
1525 |
parameters |
1526 |
- global.h now only covers global definitions that are needed for the |
1527 |
C++ API header files, all implementation internal global definitions |
1528 |
are now in global_private.h |
1529 |
- atomic.h is not exposed to the C++ API anymore |
1530 |
- no need to include config.h anymore for using LS's API header files |
1531 |
- fixed warnings in API doc generation |
1532 |
|
1533 |
* general changes: |
1534 |
- replaced the old, confusing MIDI program change mechanism by a |
1535 |
flexible MIDI instrument mapper which allows to map arbitrary |
1536 |
(MIDI bank, MIDI program) pairs with arbitrary |
1537 |
(engine type, instrument file, file index) triplets which will be |
1538 |
loaded on the respective channel when such MIDI program change |
1539 |
messages arrive, beside that, each entry allows to define a life-time |
1540 |
strategy for the instrument, so the user can define whether the |
1541 |
instrument should i.e. be permanently loaded (to be able to switch |
1542 |
quickly among sounds) or i.e. loaded just on demand when the |
1543 |
respective program change arrives (to safe RAM space), as well as a |
1544 |
global volume factor for each entry, so the user can adjust the volume |
1545 |
dynamic of his mapped instrument collection without having to modify |
1546 |
the instrument files, also one can manage arbitrary amount of such |
1547 |
MIDI instrument maps and assign each sampler channel individually |
1548 |
a certain map, so that i.e. separation of normal instruments and |
1549 |
drumkits is possible |
1550 |
- new notification events for tracking changes to audio/MIDI devices, |
1551 |
MIDI instrument maps, MIDI instruments, FX sends, global volume. |
1552 |
- sampler was limited to load max. 200 instruments in the background |
1553 |
due to a constant size RingBuffer FIFO which is now replaced by a |
1554 |
dynamic (unlimited) size std::list FIFO |
1555 |
- added FX sends, these allow to route audio signals to arbitrary audio |
1556 |
output channels for being processed by external effect processors |
1557 |
(i.e. jack-rack), the send levels are controllable via arbitrary MIDI |
1558 |
controllers |
1559 |
- global (sampler wide) volume can now be controlled at runtime |
1560 |
- Implemented new, improved notification system |
1561 |
- fixed compilation errors regarding OSX |
1562 |
(patch by Toshi Nagata) |
1563 |
- implemented instruments database |
1564 |
- added support for escape sequences to the instruments database |
1565 |
- added highly experimental support for on-the-fly instrument editing |
1566 |
within the sampler's process (by using instrument editor plugins), |
1567 |
you'll notice the new "Registered instrument editors:" message on |
1568 |
startup, added a new LSCP command: |
1569 |
"EDIT CHANNEL INSTRUMENT <sampler-channel>" |
1570 |
to spawn a matching instrument editor for the instrument on the |
1571 |
given sampler channel, the plugin path can be overridden at compile |
1572 |
time with ./configure --enable-plugin-dir=/some/dir |
1573 |
- added experimental code for synchronizing instrument editors hosted |
1574 |
in the sampler's process to safely edit instruments while playing |
1575 |
without a crash (hopefully) by either suspending single regions |
1576 |
wherever possible or if unavoidable whole engine(s) |
1577 |
- fixed several issues in fundamental "Thread" class: set scheduling |
1578 |
policy and priority on thread level, set a minimum stack size for |
1579 |
thread (TODO: a reasonable value yet to be tested), bugfix: non-RT |
1580 |
threads simply inherited properties of starting thread instead of |
1581 |
setting their own policy and priority |
1582 |
- minor fix in our automatic stack trace mechanism on crashes, the |
1583 |
main process did not wait for the stack trace process to finish |
1584 |
its output |
1585 |
- fixed some minor memory leaks |
1586 |
- reenabled assembly features support, at the moment only for |
1587 |
enabling a fast denormal FPU mode (x86 platforms supporting SSE2) |
1588 |
- minor assembly fix in x86 features detection (don't use the PIC |
1589 |
register, to avoid relocations in the text segment at runtime) |
1590 |
- POSIX callback functions of Thread.h are hidden |
1591 |
|
1592 |
* MIDI driver: |
1593 |
- dispatch bank select (MSB and LSB) messages |
1594 |
|
1595 |
* audio driver: |
1596 |
- the ALSA audio output driver parameters now reflect the correct |
1597 |
parameter value ranges for the respective selected sound card |
1598 |
(patch by Till Wimmer, a bit fixed and extended) |
1599 |
|
1600 |
* Gigasampler format engine: |
1601 |
- if a filter is used and EG2 finishes before EG1, let the voice |
1602 |
die when EG2 has finished (this fixes a problem with clicks and |
1603 |
voice starvation for some gigs) |
1604 |
- playback is no longer disabled during instrument loading |
1605 |
- all notes playing on a channel that changes its instrument keep |
1606 |
playing with the old instrument until they get a note off |
1607 |
command |
1608 |
- EG fix: a release value of zero could cause noises or crash |
1609 |
- handle MIDI coarse tuning messages (MIDI RPN #0 MSB #2 LSB) |
1610 |
- EG fine tuning: when attack is zero the EG starts at a level |
1611 |
above max sustain level, which means that there is a decay phase |
1612 |
even if sustain is 100% |
1613 |
- more EG fixes: the level could sometimes go below zero and cause |
1614 |
noises or crashes |
1615 |
- minor fix of EGDecay (patch by Toshi Nagata) |
1616 |
- fixed compiler error when --enable-override-filter-type was |
1617 |
supplied to the configure script (fixes #46) |
1618 |
- disk thread: queue sizes are now proportional to CONFIG_MAX_STREAMS |
1619 |
instead of a fix value |
1620 |
- behavior fix: on MIDI CC# 65 (portamento on / off), 126 (mono mode), |
1621 |
127 (solo mode) only kill voices if the respective mode really |
1622 |
changed |
1623 |
|
1624 |
* LSCP server: |
1625 |
- fixed compile time error for old Bison versions |
1626 |
(i.e. v1.28 found on MacOS 10.4, patch by Toshi Nagata) |
1627 |
- parser now supports extended ASCII character set |
1628 |
(up to ASCII code 255, i.e. includes now umlauts and accents) |
1629 |
- filename arguments in LSCP commands now allow to use escape |
1630 |
sequences, that is directly literal as one of: \', \", \\, \n, \r, |
1631 |
\f, \t, \v, or as octal ASCII code value like \132, or as hex ASCII |
1632 |
code value like \xf2) (fixes bug #24) |
1633 |
- the following LSCP commands now also support escape sequences for at |
1634 |
least one of their text-based parameters (i.e. name, description): |
1635 |
"ADD MIDI_INSTRUMENT_MAP", "MAP MIDI_INSTRUMENT", |
1636 |
"SET MIDI_INSTRUMENT_MAP NAME", "SET FX_SEND NAME", "CREATE FX_SEND", |
1637 |
"SET DB_INSTRUMENT_DIRECTORY NAME", |
1638 |
"SET DB_INSTRUMENT_DIRECTORY DESCRIPTION", "SET DB_INSTRUMENT NAME", |
1639 |
"SET DB_INSTRUMENT DESCRIPTION", "FIND DB_INSTRUMENTS", |
1640 |
"FIND DB_INSTRUMENT_DIRECTORIES" |
1641 |
- returns verbose syntax errors (line and column where syntax error |
1642 |
occured, the unexpected character and the actually expected, possible |
1643 |
character(s), the latter only if less than 5 possibilities) |
1644 |
- made sure that LSCP syntax is not affected by gigedit locale |
1645 |
settings |
1646 |
- bugfix regarding strings parameter lists: all comma separated lists |
1647 |
of strings were treated as being one string containing commas |
1648 |
(fixes #57) |
1649 |
|
1650 |
* Bug fixes: |
1651 |
- fixed crash occurring on certain LSCP scripts (Bug 39) |
1652 |
- another thread safety fix for lscp "load engine" and "set |
1653 |
channel audio output device" |
1654 |
- fixed a crash which occurs when reassigning the same engine |
1655 |
on a sampler channel with connected MIDI device |
1656 |
- fixed a crash which occurs when changing the number of ports of a MIDI |
1657 |
device connected to a sampler channel to number less then or equal |
1658 |
to the index of the port to which the sampler channel is connected. |
1659 |
- The previous bindings were not been disconnected when altering |
1660 |
the ALSA_SEQ_BINDINGS parameter. Introduced a NONE keyword for |
1661 |
unsubscribing from all bindings (e.g. ALSA_SEQ_BINDINGS=NONE). |
1662 |
- The active stream/voice count statistic was incorrect. |
1663 |
- notification events were not been sent for some sampler |
1664 |
channel changes |
1665 |
- added default min and max values to restrict the number of allowed |
1666 |
audio output channels and MIDI input ports |
1667 |
- the connection to the PCM interface is now closed when destroying |
1668 |
an audio output device |
1669 |
- files with slash in their path or filename could not be loaded |
1670 |
|
1671 |
* test cases: |
1672 |
- updated and fixed (haven't been touched in a while) |
1673 |
|
1674 |
Version 0.4.0 (24 November 2006) |
1675 |
|
1676 |
* packaging changes: |
1677 |
- changed deprecated copyright attribute to license; added ldconfig |
1678 |
to post-(un)install steps; added devel package for liblinuxsampler; |
1679 |
to linuxsampler.spec (RPM) |
1680 |
- install necessary development header files for allowing 3rd party |
1681 |
applications to link against liblinuxsampler |
1682 |
- liblinuxsampler's API documentation can be generated with 'make docs' |
1683 |
(Doxygen required) |
1684 |
- added benchmark to automatically detect the best triangle LFO |
1685 |
implementation (currently either an integer math solution or a |
1686 |
di-harmonic approximation), automatic detection can be overriden |
1687 |
with --enable-signed-triang-algo=x and --enable-unsigned-triang-algo=x |
1688 |
configure script argument though (mandatory for cross-compilation) |
1689 |
- do not automatically pick optimized gcc flags if the user already |
1690 |
provided some on his own (as CXXFLAGS) |
1691 |
- added compile time option to disable processing of All-Notes-Off MIDI |
1692 |
messages |
1693 |
- added compile time options to allow disabling the various audio and |
1694 |
MIDI drivers |
1695 |
- fixed automatic GCC CPU switch detection on PPC |
1696 |
(patch by Ebrahim Mayat) |
1697 |
|
1698 |
* Gigasampler format engine: |
1699 |
- extensive synthesis optimization |
1700 |
(reimplementation of EGs and LFO(s), removed synthesis parameter |
1701 |
prerendering and the synthesis parameter matrix in general, splitting |
1702 |
each audio fragment into subfragments now where each subfragment uses |
1703 |
constant pitch and filter coefficients. The volume coefficient is |
1704 |
linearly interpolated inside a subfragment, unless |
1705 |
--disable-interpolate-volume is set.) |
1706 |
- fine tuning of the EG modulation parameters |
1707 |
- improved filter cutoff calculation by adding support for the |
1708 |
following gig parameters: Cutoff freq (used when no cutoff |
1709 |
controller is defined), Control invert, Minimum cutoff, Velocity |
1710 |
curve and Velocity range. The keyboard tracking now scales |
1711 |
cutoff frequency instead of resonance. |
1712 |
- added support for gig parameter Resonance. |
1713 |
- fixed bug in sysex handling (patch by Juan Linietsky) |
1714 |
- added global volume attenuation of -9 dB (0.35f) to prevent clipping |
1715 |
which can be overridden with --enable-global-attenuation |
1716 |
- EG fixes: made the length of "attack hold" stage more |
1717 |
accurate. Release stage can now start before attack stage |
1718 |
ends. Cancel release didn't work when sustain was zero. Attack |
1719 |
time now has a minimal value to prevent clicks. |
1720 |
- fixed pitch changes larger than one octave |
1721 |
- fixed EG3 (pitch envelope) synthesis which was neutral all the time |
1722 |
- implemented portamento mode and solo mode (a.k.a 'mono mode'): |
1723 |
all modes can be altered via standard GM messages, that is CC5 for |
1724 |
altering portamento time, CC65 for enabling / disabling portamento |
1725 |
mode, CC126 for enabling solo mode and CC127 for disabling solo mode |
1726 |
- fine tuning of the curves for volume (CC7), pan (CC10 and gig |
1727 |
parameter) and crossfade |
1728 |
- added support for the "attenuation controller threshold" gig |
1729 |
parameter |
1730 |
- added smoothing of volume changes caused by control change |
1731 |
messages |
1732 |
- sample loop parameters are now taken from the DimensionRegion |
1733 |
instead of the wave chunk |
1734 |
- fixed keyswitching for v3 gigs with a number of keyswitch splits |
1735 |
not equal to a power of two |
1736 |
- reimplementation of the filter algorithm. The new filter is more |
1737 |
accurate and supports all gig filter types, including bandreject |
1738 |
and lowpass turbo. |
1739 |
- real support for 24 bit samples - samples are not truncated to |
1740 |
16 bits anymore |
1741 |
- support for aftertouch (channel pressure, not polyphonic |
1742 |
aftertouch) |
1743 |
- LFO1 behaviour fixed (dampening from max volume instead of |
1744 |
amplifying from 0) |
1745 |
|
1746 |
* LSCP server: |
1747 |
- fixed application exit on broken pipe error (fixes bug #20) |
1748 |
- fixed the notification messages delay due to lack of |
1749 |
network activity (fixes bug #26) |
1750 |
- fixed parser bug which occured on space(s) within device parameters |
1751 |
(e.g. "SET AUDIO_OUTPUT_CHANNEL_PARAMETER 0 0 SOMEPARAM='foo bar'") |
1752 |
|
1753 |
* audio driver: |
1754 |
- added aRts audio output driver (by no means RT safe) |
1755 |
|
1756 |
* MIDI driver: |
1757 |
- fixed legacy sysex code which caused dispatching of MIDI SysEx |
1758 |
messages several times instead of once |
1759 |
- API extension for MIDI drivers which already supply exact time stamps |
1760 |
for events (i.e. for offline rendering based MIDI drivers) |
1761 |
- added checks for bad MIDI data |
1762 |
|
1763 |
* linuxsampler application: |
1764 |
- show available sampler engine types on startup |
1765 |
|
1766 |
* general changes: |
1767 |
- support for muting sampler channels and solo mode of the same, two new |
1768 |
LSCP commands ("SET CHANNEL MUTE" and "SET CHANNEL SOLO") and two new |
1769 |
fields ("MUTE" and "SOLO") for command "GET CHANNEL INFO" were |
1770 |
introduced for this, the behavior is the same like on a mixer console |
1771 |
(patch by Grigor Iliev, a bit adjusted). Also added configure option |
1772 |
--enable-process-muted-channels which can be used to enable the |
1773 |
processing of muted channels. |
1774 |
- support for sostenuto pedal |
1775 |
- support for monitoring the total number of active voices |
1776 |
(new LSCP commands: GET TOTAL_VOICE_COUNT, GET TOTAL_VOICE_COUNT_MAX, |
1777 |
SUBSCRIBE TOTAL_VOICE_COUNT, UNSUBSCRIBE TOTAL_VOICE_COUNT) |
1778 |
- fixed some memory management errors |
1779 |
- fixed some concurrency problems that could lead to crashes when |
1780 |
LSCP commands were executed |
1781 |
- fixed crash when instrument loading failed previously on the same |
1782 |
sampler channel (fixes bug #36) |
1783 |
|
1784 |
Version 0.3.3 (15 July 2005) |
1785 |
|
1786 |
* packaging changes: |
1787 |
- fixed compilation with gcc 4.0 |
1788 |
|
1789 |
* Gigasampler format engine: |
1790 |
- LFO preprocessing was performed even though the respective LFO was |
1791 |
disabled by the instrument patch (minor efficiency fix) |
1792 |
- if period time of chosen audio device is too small (< MIN_RELEASE_TIME) |
1793 |
for volume ramp downs in the same fragment (needed for current voice |
1794 |
stealing implementation) then simply inform the user with a warning |
1795 |
about possible click sounds and reduce the volume ramp down |
1796 |
appropriately instead of cancelling the audio device connection |
1797 |
|
1798 |
Version 0.3.2 (24 June 2005) |
1799 |
|
1800 |
* packaging changes: |
1801 |
- updated autotools build files to compile on OS X |
1802 |
(conditional compilation of CoreMIDI and MidiShare drivers) |
1803 |
- hand-crafted assembly optimization code can be disabled with |
1804 |
'./configure --disable-asm' (definitely not recommended) |
1805 |
- fixed 'make dist' rule to include all necessary files |
1806 |
- require automake (>= 1.5) for 'make -f Makefile.cvs' |
1807 |
(due to 'dist-bzip2' automake option) |
1808 |
|
1809 |
* Gigasampler format engine: |
1810 |
- support for the gig parameters for "release velocity response" curves |
1811 |
- fine tuning of the EGADSR envelope |
1812 |
- volume of release triggered samples now depends on note-on velocity, |
1813 |
note length and gig parameter "release trigger decay" instead of |
1814 |
note-off velocity. |
1815 |
- revised voice stealing |
1816 |
(fixes crash and endless loop caused by voice stealing) |
1817 |
- don't reset scale tuning on instrument or audio output device change |
1818 |
- handle key group conflicts right at the beginning of each fragment |
1819 |
instead of when the respective voice is actually launched |
1820 |
(fixes undefined behavior if stolen voices belonged to a key group - |
1821 |
this case was followed by a "killed voice survived" error message) |
1822 |
- fixed minor issue with null/silence samples |
1823 |
(those stole voices even though they don't need a voice at all which |
1824 |
resulted in "voice stealing didn't work out" messages) |
1825 |
- don't reset volume, pan, pitch and MIDI controller values on |
1826 |
instrument or audio output device change |
1827 |
|
1828 |
* LSCP server: |
1829 |
- fixed some crashes (patch by Grigor Iliev, fixes #19) |
1830 |
- fixed LSCP event "CHANNEL_INFO" notification |
1831 |
(e.g. did not notify on volume changes or MIDI program change events) |
1832 |
|
1833 |
* linuxsampler application: |
1834 |
- added command line parameters --lscp-addr and --lscp-port to override |
1835 |
default IP address and port of LSCP server |
1836 |
|
1837 |
Version 0.3.1 (24 May 2005) |
1838 |
|
1839 |
* initial release |