RT Pre-emption

I’ve been reading a fair bit lately about using an Rt-Pre-emption kernel being good for audio applications, in terms of reducing latency. I just wondered if this is something that Volumio uses, or could do? There’s an article here on it talkunafraid.co.uk/2014/04/r … pberry-pi/

Could that result in any improvements in audio quality (in Volumio)?

Why should we care about latency? if the music starts 1/10s later when I press the play button, I really don’t care.

I don’t know. That’s why I’m asking the question, in hope that some of you gentle folk will bestow your knowledge on me! I guess you just did that …

RT kernels are useful when you MUST know that something will be executed within a certain time-frame. This implies longer time frames to ensure that and potentially higher cpu requirements. Suitable for time critical applications, like the ones that MUST stop a rotating blade within few milliseconds after detecting you put your finger on it (look for this specific example from labview on youtube).
Audio does not require it because the audio card (and the main memory) has a big buffer of already decoded data. RT is useful when few milliseconds are necessary, buffers push the tolerance to 100s of them.

Preemptive kernels are used when you want responsiveness with many tasks at the same time but in Volumio you have at most two tasks running (audio and database update). No need of it.

Have you searched for the meaning of those features before asking? :wink: