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

Contents of /linuxsampler/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1039 - (show annotations) (download)
Sat Feb 3 20:46:44 2007 UTC (17 years, 1 month ago) by persson
File size: 10699 byte(s)
* EG fix: a release value of zero could cause noises or crash
* fix of previous thread safety fix, which in some cases locked the
  engine

1 Version CVS HEAD (?)
2
3 * general changes:
4 - replaced the old, confusing MIDI program change mechanism by a
5 flexible MIDI instrument mapper which allows to map arbitrary
6 (MIDI bank, MIDI program) pairs with arbitrary
7 (engine type, instrument file, file index) triplets which will be
8 loaded on the respective channel when such MIDI program change
9 messages arrive, beside that, each entry allows to define a life-time
10 strategy for the instrument, so the user can define whether the
11 instrument should i.e. be permanently loaded (to be able to switch
12 quickly among sounds) or i.e. loaded just on demand when the
13 respective program change arrives (to safe RAM space), as well as a
14 global volume factor for each entry, so the user can adjust the volume
15 dynamic of his mapped instrument collection without having to modify
16 the instrument files, also one can manage arbitrary amount of such
17 MIDI instrument maps and assign each sampler channel individually
18 a certain map, so that i.e. separation of normal instruments and
19 drumkits is possible
20 - new notification events for tracking audio/MIDI device changes,
21 MIDI instrument map changes and MIDI instrument changes
22 - sampler was limited to load max. 200 instruments in the background
23 due to a constant size RingBuffer FIFO which is now replaced by a
24 dynamic (unlimited) size std::list FIFO
25 - added FX sends, these allow to route audio signals to arbitrary audio
26 output channels for being processed by external effect processors
27 (i.e. jack-rack), the send levels are controllable via arbitrary MIDI
28 controllers
29 - global (sampler wide) volume can now be controlled at runtime
30
31 * MIDI driver:
32 - dispatch bank select (MSB and LSB) messages
33
34 * Gigasampler format engine:
35 - if a filter is used and EG2 finishes before EG1, let the voice
36 die when EG2 has finished (this fixes a problem with clicks and
37 voice starvation for some gigs)
38 - playback is no longer disabled during instrument loading
39 - all notes playing on a channel that changes its instrument keep
40 playing with the old instrument until they get a note off
41 command
42 - EG fix: a release value of zero could cause noises or crash
43
44 * Bug fixes:
45 - fixed crash occurring on certain LSCP scripts (Bug 39)
46 - another thread safety fix for lscp "load engine" and "set
47 channel audio output device"
48
49 Version 0.4.0 (24 November 2006)
50
51 * packaging changes:
52 - changed deprecated copyright attribute to license; added ldconfig
53 to post-(un)install steps; added devel package for liblinuxsampler;
54 to linuxsampler.spec (RPM)
55 - install necessary development header files for allowing 3rd party
56 applications to link against liblinuxsampler
57 - liblinuxsampler's API documentation can be generated with 'make docs'
58 (Doxygen required)
59 - added benchmark to automatically detect the best triangle LFO
60 implementation (currently either an integer math solution or a
61 di-harmonic approximation), automatic detection can be overriden
62 with --enable-signed-triang-algo=x and --enable-unsigned-triang-algo=x
63 configure script argument though (mandatory for cross-compilation)
64 - do not automatically pick optimized gcc flags if the user already
65 provided some on his own (as CXXFLAGS)
66 - added compile time option to disable processing of All-Notes-Off MIDI
67 messages
68 - added compile time options to allow disabling the various audio and
69 MIDI drivers
70 - fixed automatic GCC CPU switch detection on PPC
71 (patch by Ebrahim Mayat)
72
73 * Gigasampler format engine:
74 - extensive synthesis optimization
75 (reimplementation of EGs and LFO(s), removed synthesis parameter
76 prerendering and the synthesis parameter matrix in general, splitting
77 each audio fragment into subfragments now where each subfragment uses
78 constant pitch and filter coefficients. The volume coefficient is
79 linearly interpolated inside a subfragment, unless
80 --disable-interpolate-volume is set.)
81 - fine tuning of the EG modulation parameters
82 - improved filter cutoff calculation by adding support for the
83 following gig parameters: Cutoff freq (used when no cutoff
84 controller is defined), Control invert, Minimum cutoff, Velocity
85 curve and Velocity range. The keyboard tracking now scales
86 cutoff frequency instead of resonance.
87 - added support for gig parameter Resonance.
88 - fixed bug in sysex handling (patch by Juan Linietsky)
89 - added global volume attenuation of -9 dB (0.35f) to prevent clipping
90 which can be overridden with --enable-global-attenuation
91 - EG fixes: made the length of "attack hold" stage more
92 accurate. Release stage can now start before attack stage
93 ends. Cancel release didn't work when sustain was zero. Attack
94 time now has a minimal value to prevent clicks.
95 - fixed pitch changes larger than one octave
96 - fixed EG3 (pitch envelope) synthesis which was neutral all the time
97 - implemented portamento mode and solo mode (a.k.a 'mono mode'):
98 all modes can be altered via standard GM messages, that is CC5 for
99 altering portamento time, CC65 for enabling / disabling portamento
100 mode, CC126 for enabling solo mode and CC127 for disabling solo mode
101 - fine tuning of the curves for volume (CC7), pan (CC10 and gig
102 parameter) and crossfade
103 - added support for the "attenuation controller threshold" gig
104 parameter
105 - added smoothing of volume changes caused by control change
106 messages
107 - sample loop parameters are now taken from the DimensionRegion
108 instead of the wave chunk
109 - fixed keyswitching for v3 gigs with a number of keyswitch splits
110 not equal to a power of two
111 - reimplementation of the filter algorithm. The new filter is more
112 accurate and supports all gig filter types, including bandreject
113 and lowpass turbo.
114 - real support for 24 bit samples - samples are not truncated to
115 16 bits anymore
116 - support for aftertouch (channel pressure, not polyphonic
117 aftertouch)
118 - LFO1 behaviour fixed (dampening from max volume instead of
119 amplifying from 0)
120
121 * LSCP server:
122 - fixed application exit on broken pipe error (fixes bug #20)
123 - fixed the notification messages delay due to lack of
124 network activity (fixes bug #26)
125 - fixed parser bug which occured on space(s) within device parameters
126 (e.g. "SET AUDIO_OUTPUT_CHANNEL_PARAMETER 0 0 SOMEPARAM='foo bar'")
127
128 * audio driver:
129 - added aRts audio output driver (by no means RT safe)
130
131 * MIDI driver:
132 - fixed legacy sysex code which caused dispatching of MIDI SysEx
133 messages several times instead of once
134 - API extension for MIDI drivers which already supply exact time stamps
135 for events (i.e. for offline rendering based MIDI drivers)
136 - added checks for bad MIDI data
137
138 * linuxsampler application:
139 - show available sampler engine types on startup
140
141 * general changes:
142 - support for muting sampler channels and solo mode of the same, two new
143 LSCP commands ("SET CHANNEL MUTE" and "SET CHANNEL SOLO") and two new
144 fields ("MUTE" and "SOLO") for command "GET CHANNEL INFO" were
145 introduced for this, the behavior is the same like on a mixer console
146 (patch by Grigor Iliev, a bit adjusted). Also added configure option
147 --enable-process-muted-channels which can be used to enable the
148 processing of muted channels.
149 - support for sostenuto pedal
150 - support for monitoring the total number of active voices
151 (new LSCP commands: GET TOTAL_VOICE_COUNT, GET TOTAL_VOICE_COUNT_MAX,
152 SUBSCRIBE TOTAL_VOICE_COUNT, UNSUBSCRIBE TOTAL_VOICE_COUNT)
153 - fixed some memory management errors
154 - fixed some concurrency problems that could lead to crashes when
155 LSCP commands were executed
156 - fixed crash when instrument loading failed previously on the same
157 sampler channel (fixes bug #36)
158
159 Version 0.3.3 (15 July 2005)
160
161 * packaging changes:
162 - fixed compilation with gcc 4.0
163
164 * Gigasampler format engine:
165 - LFO preprocessing was performed even though the respective LFO was
166 disabled by the instrument patch (minor efficiency fix)
167 - if period time of chosen audio device is too small (< MIN_RELEASE_TIME)
168 for volume ramp downs in the same fragment (needed for current voice
169 stealing implementation) then simply inform the user with a warning
170 about possible click sounds and reduce the volume ramp down
171 appropriately instead of cancelling the audio device connection
172
173 Version 0.3.2 (24 June 2005)
174
175 * packaging changes:
176 - updated autotools build files to compile on OS X
177 (conditional compilation of CoreMIDI and MidiShare drivers)
178 - hand-crafted assembly optimization code can be disabled with
179 './configure --disable-asm' (definitely not recommended)
180 - fixed 'make dist' rule to include all necessary files
181 - require automake (>= 1.5) for 'make -f Makefile.cvs'
182 (due to 'dist-bzip2' automake option)
183
184 * Gigasampler format engine:
185 - support for the gig parameters for "release velocity response" curves
186 - fine tuning of the EGADSR envelope
187 - volume of release triggered samples now depends on note-on velocity,
188 note length and gig parameter "release trigger decay" instead of
189 note-off velocity.
190 - revised voice stealing
191 (fixes crash and endless loop caused by voice stealing)
192 - don't reset scale tuning on instrument or audio output device change
193 - handle key group conflicts right at the beginning of each fragment
194 instead of when the respective voice is actually launched
195 (fixes undefined behavior if stolen voices belonged to a key group -
196 this case was followed by a "killed voice survived" error message)
197 - fixed minor issue with null/silence samples
198 (those stole voices even though they don't need a voice at all which
199 resulted in "voice stealing didn't work out" messages)
200 - don't reset volume, pan, pitch and MIDI controller values on
201 instrument or audio output device change
202
203 * LSCP server:
204 - fixed some crashes (patch by Grigor Iliev, fixes #19)
205 - fixed LSCP event "CHANNEL_INFO" notification
206 (e.g. did not notify on volume changes or MIDI program change events)
207
208 * linuxsampler application:
209 - added command line parameters --lscp-addr and --lscp-port to override
210 default IP address and port of LSCP server
211
212 Version 0.3.1 (24 May 2005)
213
214 * initial release

  ViewVC Help
Powered by ViewVC