Commit graph

2528 commits

Author SHA1 Message Date
derselbst
25b0503ba7 Avoid audible clicks when rapidly chaning panning 2021-02-07 11:05:33 +01:00
Tom M
0d76403f9f
Parallelize SF3 loading (#746) 2021-02-06 20:32:17 +01:00
Tom M
4836ff15c4
Speedup Android CI a bit (#769) 2021-02-06 19:28:52 +01:00
jjceresa
616fca3d8d Elaborate out-of-polyphony error message 2021-02-06 16:57:32 +01:00
Tom M
8745f542c2
Merge pull request #747 from FluidSynth/oboe-phil
Update Oboe driver
2021-02-06 10:43:07 +01:00
derselbst
ad16f40325 Speed up build + tests for TravisCI 2021-02-03 21:48:59 +01:00
derselbst
641605294c Introduce audio.oboe.sample-rate-conversion-quality setting 2021-01-30 19:31:51 +01:00
derselbst
977fe31ef4 Document the impact of synth.sample-rate on audio drivers 2021-01-30 18:45:21 +01:00
Chris Xiong
712707fe87
Add WASAPI driver. (#754)
This driver is currently tested and verified to work on:
 - Windows Vista x64 VM
 - Windows 7 x64 VM
 - Windows 10 1909 x64 (VM and Laptop)
 - Windows 10 21296 x64 on a ThinkPad X1 Yoga 1st Gen with 3 different sound cards (Conexant CX20753/4, Scarlett Solo Gen 2, Aureon 7.1 USB)

This driver is capable of reaching very low latency in exclusive mode (~6ms on Scarlett Solo with 48kHz).
2021-01-29 18:11:17 +01:00
Tom M
d0e90be5b5
Update TravisCI badge 2021-01-29 16:43:47 +01:00
derselbst
9e0c401aeb Document new WASAPI settings and other missing ones 2021-01-29 14:57:25 +01:00
derselbst
b84e8b83e0 Merge branch '2.1.x' into master 2021-01-29 14:42:04 +01:00
derselbst
e2d67ea772 Bump to 2.1.7 2021-01-29 14:37:27 +01:00
derselbst
dc3d7b8ffa Fix possible NULL deref
when allocation of drybuf fails
2021-01-29 14:07:01 +01:00
derselbst
4d76a6ad85 Restore int16 sample format for waveout and dsound
Addresses #760
2021-01-29 13:59:21 +01:00
jjceresa
8322d95425
Fix waveout driver crash (#759)
This PR addresses issue https://github.com/FluidSynth/fluidsynth/issues/758.

It ensures that the `internal buffer size` used by waveout device driver and the `extra buffers size` required by fluid_synth_process() are both coherent (i.e they should be of same size). To ensure this, both kind of buffers are now dependent of`audio.period` and `audio.period-size settings`.
2021-01-29 13:56:57 +01:00
rncbc
cd28701d52 Allow the MIDI file player restart on-demand (#755)
Resets the default sample timer properly, allowing the internal
MIDI file player to restart its playlist on any other time but
the initial first.

Lets Qsynth play any MIDI files that are drag-n-dropped, anytime
after the first, following synth engine initialization.
2021-01-28 11:25:13 +01:00
derselbst
c7878dec74 Update API docs of fluid_settings_getstr_default() 2021-01-28 11:15:23 +01:00
derselbst
4b5afca76c Fix a double-free when issuing the info command on string setting
Fixes #756.
2021-01-28 10:34:39 +01:00
derselbst
70abf1953f Add a comment for issue #751 2021-01-23 17:50:40 +01:00
derselbst
4fd7899d61 Merge branch '2.1.x' into master 2021-01-23 17:16:26 +01:00
derselbst
e887a6369c Update new_fluid_audio_driver2() support list 2021-01-23 16:28:56 +01:00
derselbst
fff51822eb Disable floating point underflow exceptions
They often occur in reverb, chorus and IIR and break our CI for no good reason.
2021-01-23 16:00:42 +01:00
derselbst
b9820362f4 Update API docs about bug #751 2021-01-23 15:57:28 +01:00
Chris Xiong
8e9d361651 chorus: avoid overwriting input buffer when storing samples in the delay line.
In `fluid_rvoice_mixer.c`:`fluid_rvoice_mixer_process_fx()`:

If an audio processing callback is used, `mix_fx_to_out` would be `FALSE`. As a result, `in_ch` and `out_ch_l` points to the same buffer.

In `fluid_chorus.c`:`fluid_chorus_processreplace()`:
```C
        /* process stereo unit */
        /* store the chorus stereo unit d_out to left and right output */
        left_out[sample_index]  = d_out[0] * chorus->wet1  + d_out[1] * chorus->wet2;
        right_out[sample_index] = d_out[1] * chorus->wet1  + d_out[0] * chorus->wet2;

        /* Write the current input sample into the circular buffer */
        push_in_delay_line(chorus, in[sample_index]);
```

Here the chorus processing code writes to the left output buffer (which will overwrite the input buffer in this case) before the sample from the input buffer is stored into the delay buffer, making the chorus output all zeros. If no audio processing callback is used, `mix_fx_to_out` would be `TRUE` and `in` and `left_out` will not point to the same buffer, therefore the order doesn't matter.

Simply swapping the two steps should be a sufficient fix. This patch also apply the same change to `fluid_chorus_processmix` only for the sake of consistency (since they are almost exact copies of each other).

Resolves #751.
2021-01-23 15:49:26 +01:00
derselbst
1fefa4d2f2 Merge pull request #745 from chirs241097/winadrv-cb
Add support for new_fluid_audio_driver2 to dsound and waveout drivers.
2021-01-23 15:08:24 +01:00
Chris Xiong
64121229ea Add support for multiple stereo outputs when callbacks are used. 2021-01-23 15:06:40 +01:00
derselbst
d547b569ad Leave framesPerCallback unspecified 2021-01-22 19:48:54 +01:00
derselbst
998eeee471 Elaborate API docs of fluid_audio_func_t 2021-01-22 19:46:03 +01:00
Chris Xiong
908494524b Consistency changes and removal of other unnecessary stuff. 2021-01-20 11:55:29 +08:00
derselbst
1e18ee7781 Use oboe 1.5.0 2021-01-19 18:48:52 +01:00
derselbst
c9c80a5fde Oboe driver can have lower latency 2021-01-19 18:48:52 +01:00
derselbst
2cfd56bb10 Modernize Oboe driver 2021-01-19 18:48:52 +01:00
Chris Xiong
772f62fc91 Add support for new_fluid_audio_driver2 while using the waveout driver. 2021-01-19 10:25:24 +08:00
Chris Xiong
9562ae6009 Add support for new_fluid_audio_driver2 while using the dsound driver. 2021-01-19 10:24:40 +08:00
jjceresa
4f2cb370a1
Add shell commands to the MIDI File Player (#713) 2021-01-15 19:04:02 +01:00
Tom M
ca6bcda7d9
Merge pull request #739 from FluidSynth/parse-cmd3
Handle settings-related commands in user command file early
2021-01-10 12:08:19 +01:00
jjceresa
2cada68e02
Fix MIDI player tempo reset issues (#711) 2021-01-10 12:01:28 +01:00
derselbst
20ff3f866c Avoid unknown command errors when parsing settings early 2021-01-09 19:44:18 +01:00
derselbst
9bb048f27f Don't print warning when parsing early 2021-01-09 18:51:46 +01:00
derselbst
40ed09367a Parse set commands from config file early 2021-01-09 18:51:33 +01:00
Mauro Carvalho Chehab
4e8f570d32 Don't warn when parsing config files
When fluid_shell is called from fluid_source(), is is currently
printing this message:

    Received EOF while reading commands, exiting the shell.

Suppress it.
2021-01-08 23:48:40 +01:00
Mauro Carvalho Chehab
3d805e9a03 Only change a set if the value actually changed
That prevents warnings about changing a non-realtime parameter
set command is issued, but the value is identical to the
previous one.
2021-01-08 23:47:39 +01:00
derselbst
e04cd572cb Merge branch '2.1.x' into master 2021-01-03 21:37:32 +01:00
Tom M
fd9bf6a6b0
MinGW CI: Provide Release Build Binaries (#738) 2021-01-03 20:55:48 +01:00
derselbst
272b3b8717 Update API docs on fluid_synth_sfunload() 2021-01-03 18:30:16 +01:00
jjceresa
cf424de5cf delete_fluid_synth() may not unload soundfonts still used by rvoices 2021-01-03 18:30:16 +01:00
jjceresa
0d38823527 SoundFonts cannot be unloaded if polyphony is ever exceeded
If polyphony is exceeded and FluidSynth has to allocate a voice by
calling fluid_synth_free_voice_by_kill_LOCAL(), two problems occur:

 1)The value returned by fluid_synth_get_active_voice_count() never
   returns back to 0.
 2)SoundFont samples are not unref'd properly, and therefore if an attempt is
   made to unload the SoundFont, the deferred unload timer is started, and
   fluid_synth_sfunload_callback() unsuccessfully tries	to unload the SoundFont forever.

These 2 issues are fixed by this commit.
2021-01-03 18:30:16 +01:00
Tom M
5c1cfe6a5f
Regression tests for #727 (#735)
This PR adds regression tests for #727, ensuring that soundfonts are correctly unloaded via the lazy-timer-unloading mechanism.
2021-01-03 09:42:42 +01:00
Tom M
6339feb106 Update API docs on settings recycling 2021-01-03 09:41:46 +01:00