Hi,
Couldn't do the math details as there are also unknown vars/values
(e.g: hardware/driver layer buffering), however, as mentioned in the
wiki (combined items on sound device & end-to-end latency test):
----
This test will measure the total latency introduced by:
* microphone and speaker device buffering (both in application
layer, driver layer, and in the hardware itself)
* conference bridge buffering
* jitter buffering
* codec buffering
* and everything else (e.g: processing)
---
One of the reason why conference bridge does buffering is because
there is sound device burst.
In case you haven't seen this
http://trac.pjsip.org/repos/wiki/FAQ#audio-latency (also mentioned in
the same wiki page).
Regards,
nanang
On Mon, Feb 2, 2009 at 3:33 PM, Antoine Junod <toto at tots-ns.net> wrote:
> Hello, list!
>
> I'm trying to improve the latency of my end to end system, starting
> with the local loop of one of the UA. I'm measuring the latency as
> described in [0]. With the following parameters, I'm able to go down
> to 74ms in average:
>
> --no-tones \
> --ec-tail 0 --no-vad --jb-max-size=1 \
> --clock-rate 8000 --snd-clock-rate 48000 \
> --rec-file rec1.wav --play-file tock8.wav \
> --ptime=10 \
> --playback-lat=1 --capture-lat=1
>
> Is there a way to know were these 74ms are spent? It is of course not
> a lot but I would be very happy to cut it down by 2 or more. (I'm
> trying to use the pjsip stack for a mic application).
>
> Any link, pointer or explanations are welcome.
>
> Thanks a lot for your reply,
> -AJ
>
> [0] http://trac.pjsip.org/repos/wiki/MeasuringSoundLatency
>
>
>
> _______________________________________________
> Visit our blog: http://blog.pjsip.org
>
> pjsip mailing list
> pjsip at lists.pjsip.org
> http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org
>
