jamulus/ChangeLog

506 lines
13 KiB
Text

3.4.8git
* refresh server list if the Central Server address type is changed
TODO fix UpdateCurLevel() for new 64 samples frame size
TODO mixer faders linear -> log gain is better
TODO auto jitter buffer performance not good at 64 samples frame size -> to be checked again
TODO for different frame sizes (64/128) the start value of 6 for the jitter buffer might be too low
TODO offer the Jamulus ASIO settingspanel in case of an ASIO ERROR to fix, e.g., incorrect sample rate (https://sourceforge.net/p/llcon/discussion/533517/thread/777663cf94/#035f)
TODO client: larger sound card buffers are managed by conversion buffer, not in processintern for loop (note iSndCrdFrameSizeFactor!)
maybe let the use select all possible sound card buffer sizes, this should solve Ticket #53
TODO store Central Server jamulus.server start scripts on the Repo (one for Central Server, one for Central Server North America)
TODO should LHostAddr be QHostAddress instead of CHostAddress? See https://github.com/corrados/jamulus/pull/48#issuecomment-612594981
3.4.7 (2020-04-11)
* added support for alternative Central Servers to solve the 200 server registration limit (Ticket #50)
* added support for 64 samples frame size in the server (if server runs in 64 or 128 samples
mode it is still compatible to both, 64 and 128 samples frame size clients)
* added multichannel CoreAudio support, coded by emlynmac (#44)
* fixed server not visible if in same local network, coded by pljones (Ticket #27)
3.4.6 (2020-04-09)
- added support for channel meters, coded by pljones
- added show licence switch in the server GUI
- store fader mute state in the ini file, coded by doloopuntil
- fixed low-res icon issue (Ticket #28)
3.4.5 (2020-04-04)
- audio fade-in at the server if a new client connects
- added a scroll bar to the mixer board to support large numbers of
mixer faders (a thank you to doloopuntil for his help)
- changed the maximum number of clients supported by the server from 20 to 50
- Windows installer now contains a 32 and 64 bit version of Jamulus (the version to
be installed is selected automatically according to the detected operating system)
- bug fix: server list ping times may not be accurate and client list may not be retrieved
3.4.4 (2020-03-25)
- added support for controlling the audio mixer faders with a MIDI controller (MacOS and Linux)
- added command line argument for disabling auto jack connection (Ticket #49)
- audio recording for the server, coded by pljones
- SVG server history graph, coded by pljones
3.4.3 (2018-05-17)
- for ASIO and 4 channel input, support mixing channels 1&2 with 3&4
- bug fix: fixed a crash, running Jamulus on MacOS version "High Sierra"
3.4.2 (2017-08-11)
- removed old CELT library (minimum compatible version is now 3.3.1)
- show server name in the server list in bold font if it is a permanent server
- Jack can be used instead of CoreAudio on MacOS (using qmake "CONFIG+=jackonmac")
3.4.1 (2016-02-10)
- show the number of connected clients in the MacOS task bar if more
than one client is connected
- avoid a single jitter buffer for the auto detection
- the Musicians value in the server list shows a warning if the server is full
- automatic server setting for permanent server flag in the protocol
- bug fix: ping time measurement may be invalid for the Linux OS
3.4.0 (2015-12-10)
- show the names of the connected clients in the server list
3.3.11 (2015-11-26)
- added a new client fader level setting
- changed the MacOS audio interface to be future proof (do not use
the Carbon Component Manager anymore)
- added support for audio channel selection for MacOS
3.3.10 (2015-10-20)
- changed the default central server URL
- added support for server disconnection
3.3.9 (2015-07-12)
- another improvement of auto jitter buffer detection in very bad
network conditions
- support client operation without using a GUI front end
- bug fix: fixed a crash in the MacOS audio interface
3.3.8 (2015-03-15)
- improved audio quality in bad network conditions
- more realistic overall delay estimation
- improvement of auto jitter buffer detection in bad network conditions
- show info about sound card buffer size if not one of the standard sizes
3.3.7 (2015-02-05)
- added a musician profile dialog (some settings in the main window
were removed)
- a tool tip for the fader tag shows the complete musician profile
- the city and skill level can be set in the musician profile
- added new instrument picture for "Guitar+Vocal"
3.3.6 (2015-01-25)
- support for a country flag icon on the fader tag
- a licence agreement dialog can be requested by the server
3.3.5 (2014-07-30)
- new compile config options for disabling old CELT, use OPUS in a shared
library and change the executable name
- added a Linux jamulus.desktop file
3.3.4 (2014-02-25)
- true stereo reverberation effect (previously it was a mono reverberation
effect on both stereo channels)
- added a mono-in/stereo-out mode to support special sound cards which have
mono inputs for the instrument and a microphone but have stereo outputs
- store fader solo state in the ini file
- improved stability of the audio stream by reducing audio drop outs (by
using a separate socket thread)
- removed unnecessary settings and LED indicators
- bug fix: the fader level could not be changed if the fader was on solo
3.3.3 (2013-12-30)
- support for storing/recovering the window positions
- added new instrument pictures for "Recorder", "Streamer" and "Listener"
- the solo state of a mixer fader is not exclusive any more and the solo
state is preserved if the number of mixer faders changes
- more precise sound card latency estimation under Windows
- integrated a new OPUS version: v1.1
- bug fix: support for correct utf-8 storage of names in the ini-file
3.3.2 (2013-08-24)
- the connection setup dialog can now be opened during a connection
- support for three audio quality settings: low, normal and high
3.3.1 (2013-03-24)
- enabled the OPUS codec
- store fader level settings
- improved server performance under Linux
- changed the network buffer for improved OPUS PLC performance
- added protocol overhead for DSL line for upload rate calculation
- fixed outstanding renaming from llcon to Jamulus
- bug fix: the overall delay was not correctly calculated
3.3.0 (2013-02-24)
- renamed the software from llcon to Jamulus
- support for an instrument picture on the fader tag
- a server welcome message can be shown on a connect to the
server (the message is shown in the chat dialog of the client)
- the source code is now compatible to Qt5
- bug fix: strings in the protocol are now utf-8 coded
3.2.2 (2012-07-15)
- changed auto jitter buffer property to reduce audio drop outs
- bug fix: mechanism to keep port in NAT of slave server open works now correctly
- bug fix: slave server unregistering may not be successful
3.2.1 (2012-02-02)
- support for audio device selection under Mac OS
- connect dialog list is sorted by the ping time
- software icon changed
- bug fix: fixed distorted audio with USB sound cards on Mac OS
- bug fix: fixed crash on slave server unregistering
3.2.0 (2011-07-23)
- new GUI style of the main window, added switch for selecting the GUI style
in the settings window
- a list of available servers is shown on pressing the connect button, the list
is managed by a central server, any private server is added automatically if
the registering setting is enabled
- the jitter buffer size can be independently set for client and server
- improved auto jitter buffer algorithm
- the Qt project file is now used for Linux, too
- ini-file is stored in the home directory instead of the application directory
- added server settings in the GUI for the server list and added ini file
support to store the settings
- bug fix: fixed Jack audio interface instability
3.1.2 (2011-03-02)
- show warning in main window if audio delay is too long
- added download link in help menu
- bug fix: solo switch did not work correctly
3.1.1 (2010-07-01)
- added stereo audio channel support
- added input/output audio channel mapping for ASIO audio interface
3.1.0 (2010-03-16)
- support for MacOS (using CoreAudio audio interface)
- sound card frame size support for frame sizes other than 128, 256 and 512
samples
- improvement of network buffer (jitter buffer) in case of small buffer sizes
- all available ASIO sample formats supported
- bug fix for M-Audio Delta ASIO driver
3.0.3 (2009-12-05)
- accessibility improvements
- show number of connected clients in window title (and therefore in OS task
bar)
- added "Whats this" help text to the GUI controls in the general settings
dialog, added Tool Tips to some GUI controls
- server logging history grid lines of weekend days are now plotted with
different line width
- removed ALSA support since the ALSA interface implementation in llcon was
buggy and will not be maintained in the future
- bug fix: fix for server logging history graph x-axis
- bug fix: fix for disconnect issue at the server
3.0.2 (2009-09-21)
- new server features: for chat messages the time stamp is also shown, parsing
of existing log file supported
- updates for help texts
- new design for fader tag
- change for Jack Linux audio interface: ports are only registered and
connected once when the software is started up
- bug fix: under bad network conditions chat messages were randomly repeated
- bug fix: in case the server was shutdown and restarted during a connection,
the channel name was not updated correctly at the server
3.0.1 (2009-08-23)
- use low complexity CELT encoder profile, this lowers audio dropout
probability on slow computers
- in case "Open Chat on New Message" is not enabled, a hint in the status bar
is shown when a message is received
- bug fix: buzzing occurred when audio stream was interrupted (e.g. in case
of network trouble)
3.0.0 (2009-08-19)
- introduced new audio codec "CELT", not compatible to old versions
- only the sound card frame sizes 128, 256 and 512 are allowed (since other
frame sizes require additional conversion buffers which introduce delay)
- IMA-ADPCM, MS-ADPCM removed
- since CELT works on 48 kHz sample rate, resampling was removed
- various bug fixes (e.g. disconnecting did not work reliably)
2.3.0 (2009-07-09)
- new system sample rate of 33 kHz to improve audio quality, not compatible
to old versions
- added history graph for server logging
- added command line argument for connecting a server at software start-up
2.2.2 (2009-05-14)
- "Mute" and "Solo" check boxes for each connected client fader
- store previous server URLs
- changes to the main GUI (grouped "local" and "server" settings)
- LED status lights and LED input level meter
- better behaviour on disconnection (introduced disconnection protocol message)
2.2.1 (2009-03-29)
- bug fix and improvements for automatic jitter buffer size setting
2.2.0 (2009-03-13)
- added Jack audio interface (Linux)
- simplified settings dialog, complete redesign (removed sound card
buffer settings, network block sizes settings, added upload rate display)
- improved audio stability (audio interface is not callback based, removed
intermediate audio buffers, client audio buffer size equals network
buffer size now)
- added upload rate limitation for server (server decides which network
parameters to use depending on the upload limit and the number of connected
clients)
2.1.4 (2009-02-15)
- added automatic jitter buffer size setting
- speed optimizations to improve audio interface stability
- new default settings (e.g. turn off Reverb by default since it requires
significant CPU)
2.1.3 (2008-11-02)
- added sound card selection
- improved ASIO configuration
- added total delay display
2.1.2 (2008-08-15)
- audio compression type can be selected (IMA ADPCM, MS ADPCM, None)
- security checks for protocol messages (wrong messages could crash the
software)
- bug fix: ping time measurement problems on Windows OS fixed
2.1.1 (2008-08-03)
- added a chat window
- the client can select a port number of server additional to the URL
- the server can be operated on an arbitrary port number
- ping time information in general settings dialog added
2.1.0 (2008-07-17)
- ASIO support
2.0.0 (2008-03-29)
- first QT4 compatible release
0.9.9 (2007-09-10)
- new client settings dialog
- at each client a separate audio mix can be generated for all connected clients
at the server
0.9.4 (2006-03-10)
- added a protocol mechanism, now it is possible to set the jitter buffer in the
server according to the setting in the client (they are coupled now)
- removed sample rate offset estimation since it was not used anyway
- internal audio processing in the server is now based on the minimum block
size (improves latency)
0.9.1 (2006-01-28)
- initial version