Age | Commit message (Collapse) | Author |
|
This is needed in order for us to allow using a different
keyboard layout. The idea in the future is to move libinput
to a peripheral add-on but that won't be until V19. So this
is needed in the meantime.
I can also add more options in the future related to mice and keyboards
but I have only heard about keyboard layouts so far.
|
|
previous commit, making them unavailable in kodi.
Add default audio device to Pi settings
|
|
|
|
settings: Add settings option to control atempo resampling
|
|
Atempo rsampling does increase cpu usage so may want to be disabled (e.g. on Pi 1).
The threshold when atempo resampling is used compared to normal resampling could be usefully changed.
This adds a setting for controlling this and defaults to disabled on Pi 1
|
|
The group got renumbered in #10184
|
|
|
|
|
|
|
|
|
|
The 3D settings got lost in the refactor.
Also fix a couple of cosmetics.
|
|
|
|
|
|
|
|
|
|
|
|
[rbp/settings] Allow av sync type to be enabled
|
|
It works for dvdplayer
|
|
The Pi supports dual audio in omxplayer, where the gpu outputs audio to
both hdmi and analogue connections.
With dvdplayer being more usable on Pi now, there are frequent requests
for this functionality through the Pi Sink.
This removes the gui setting for "dual audio" and creates a new sink
output for dual audio. It add support to the Pi sink and makes omxplayer
use the new setting.
|
|
|
|
While omxplayer is the best performing player on the Pi (and the only option for raw Blu-Rays)
there are some reasons to use dvdplayer on a Pi:
DVDs with menus do not work well with omxplayer
support of software decode for SD codecs like divx3, msmpeg, sorenson spark etc
omxplayer doesn't support alsa, so can't be used with USB and I2S sound cards
dvdplayer may behave better with dodgy timestamps and packet loss, e.g. with live TV
So this PR adds hardware acceleration to dvdplayer using the MMAL api.
There is also a MMAL based video renderer. Compared to using GL, this:
saves the processing cost of YUV->RGB conversion
saves the memory cost of the RGB textures for video frames (1080p = 3MB for YUV and 8MB for RGB)
allows high quality (cubic) scaling
|
|
|
|
On a 256M Pi, it's safest to default to 720p gui as memory is quite limited.
|
|
(ALWAYS) and quirk it on Linux and FreeBSD
|
|
by moving them to their own settings category
|
|
|
|
Also, it is too expensive for the Pi and never keeps up,
so mention that in Pi specific help.
|
|
The hardware can't support HD audio, so remove them from settings
|
|
|
|
To save memory we default to 16bpp textures,
but provide a GUI option to increase that to 32bpp.
|
|
|
|
[rbp] Disable frame multi threaded decoded settings option
|
|
This was done in #2388, but got lost in the setting refactor and is currently disabled
|
|
Makes no sense on Pi
|
|
|
|
Requires updated firmware (from next branch)
|
|
|
|
|
|
|
|
This adds a DRC scheme. A number of dBs of amplification can be chosen.
The audio_decode component is setup to do the usual downmix operation early
before most of the audio buffering. This means we get a second or two to
detect if the downmix would overflow. This allows the amplification to be
attenuated in time to avoid the clipping.
The attenuation gradually decays when overflow condition stops.
Also adds a gui setting to boost centre channel.
Requires updated firmware
|
|
Thanks to stupid-boy for the original implementation.
|
|
I've been through the settings carefully and have made any settings that are not implemented on Pi be not-visible.
If and when support is added for probing capabilities of the media player then the hiding of unsupported options can be removed.
Thanks to montellese for making this configuration so simple.
|
|
|