Volume Control Bit Loss
#1
Ok,

This is NOT a "bit perfect" question.... all my MASTER source music files are 44.1/16 WAV's ripped from CD's BUT my PLAYBACK files are FLAC's resampled using "sox" with the "ultra" setting to 48k/24bit (i.e. I have two copies of my music files).

Thus I am already not bit perfect and I do this because historically (and even today) many/most computer platforms work at 48k (sure you can do stuff to overcome this) so to make my playback chain consistent, simple and to avoid unwanted intermediate processing, I use 48k. Plus this means a DAC doesnt need to resample (causes its done already).

Also note I have been a Kodi user for 8 years so am not questioning it in a detrimental/trolling way.

My question is based on my use prior use of a SlimDevices Transporter, which worked internally at 24 bit (unlike Kodi which works at 32). It's volume control used db for its volume measure, noting I was feeding it 44.1/16 files as it doesnt resample.

Thus the general Transporter rule with 16bit files was after you reached -48db on the volume control (1 bit lost per 6db) you would start to lose resolution.

So my question is if Kodi is 32bit internally (and yes I understand that above 16 bits in my case, all we have is zero based bit padding) and the volume control uses percent as it measure, does this means (based on the Transporter logic above) that a 75% volume it is working at ~24 bits and 50% volume it is working at ~16 bits?

Thanks,

Bluck
Reply
#2
Decoder and audio processing are done in float 32 bit. The final ouput sink is opened in highest precision available from the device.

For alsa this is most of the time 32 bit int.
If pulseaudio or pipewire are running, the decoded float data is given directly to them.
First decide what functions / features you expect from a system. Then decide for the hardware. Don't waste your money on crap.
Reply

Logout Mark Read Team Forum Stats Members Help
Volume Control Bit Loss0