Re: Improving status timestamp accuracy

[Alan Young <consult.awy@gmail.com>]

On 06/06/16 09:34, Takashi Iwai wrote:
> On Sun, 05 Jun 2016 12:33:20 +0200,
> Alan Young wrote:
>> Regardless of what value of DMA_RESIDUE_GRANULARITY_xxx that a driver
>> claims to support, it is not really defined how fine a burst might
>> be. So the end result is, from the point of view of audio, that the
>> resulting position obtained by the pointer() call is pretty
>> inaccurate. Hence my proposal to attempt to improve the accuracy of
>> the pcm_status() result given the above constraints.
> Well, the subject appears misleading.  What you want isn't the audio
> timestamp accuracy.  From API POV, the accurate position is calculated
> via the (additional) delay.  So, what you want is rather the accurate
> position delay accounting, and the audio timestamp is merely one of
> the ways to achieve that.
>

Well, yes, you could put it that way. Whether an accurate delay, 
combined with the associated timestamp, or an accurate audio delay, I 
would have the data needed to track audio drift from wallclock time.

See my response to Raymond for more detail.

Alan.