file/stream info issues and audio processing logic confusion
Posted: Thu Feb 11, 2016 2:58 pm
Hi Дима!
I would like to sort out some mess with file/stream information displayed under the seek-bar (when 3 dots are clicked under controls).
When 64-bit processing is enabled the following info is being showed: AAC@16 bits, MPC@64, OPUS@32, MPEG@64, OGG@32- all that is very confusing since non-PCM formats have no associated bit depths and most decoders usually output data @ 32 float-point. So what decoder is used in Neutron for all these formats? Is it 32 bit float-point or 64-bit or varies
I believe, there should be no bit depth specified for non-PCM formats! But it should be clear what bit depth decoder produces for each format. in 32-bit mode or 64-bit (if decoder outputs in such bit depth).
Also i don't understand the point of using dithering as a quality option. If output device is limited to 16-bit it should be enabled automatically for anything higher than 16-bit by default (32/64-bit non-pcm decoder out, 24 bit files, enabled 64-bit processing). Since Android 5< allows for 32bit output, there's no need for dithering at all and no need for option to enable 32-bit output as well, imo, it should be outputting 32bit as default.
Btw OGG shouldn't be used as description of an audio stream, since it's a container. Vorbis would be the correct indication (in case it's Vorbis)
Also it would be nice to have option to completely turn off internal sample rate conversion.
Дим, clarify this stuff, пожалуйста.
Very kind regards.
I would like to sort out some mess with file/stream information displayed under the seek-bar (when 3 dots are clicked under controls).
When 64-bit processing is enabled the following info is being showed: AAC@16 bits, MPC@64, OPUS@32, MPEG@64, OGG@32- all that is very confusing since non-PCM formats have no associated bit depths and most decoders usually output data @ 32 float-point. So what decoder is used in Neutron for all these formats? Is it 32 bit float-point or 64-bit or varies
I believe, there should be no bit depth specified for non-PCM formats! But it should be clear what bit depth decoder produces for each format. in 32-bit mode or 64-bit (if decoder outputs in such bit depth).
Also i don't understand the point of using dithering as a quality option. If output device is limited to 16-bit it should be enabled automatically for anything higher than 16-bit by default (32/64-bit non-pcm decoder out, 24 bit files, enabled 64-bit processing). Since Android 5< allows for 32bit output, there's no need for dithering at all and no need for option to enable 32-bit output as well, imo, it should be outputting 32bit as default.
Btw OGG shouldn't be used as description of an audio stream, since it's a container. Vorbis would be the correct indication (in case it's Vorbis)
Also it would be nice to have option to completely turn off internal sample rate conversion.
Дим, clarify this stuff, пожалуйста.
Very kind regards.