Explain “core show translation”

Home » Asterisk Users » Explain “core show translation”
Asterisk Users 3 Comments

Hi,

I’ve read in http://www.voip-info.org/wiki/view/Asterisk+codecs but I still
have questions about “core show translation”.

How are values replied by “core show translation” computed in the the first
place ?
I’ve got 2 machines using a 3GHz Intel CPU : 1 is a Xeon, 1 is a Pentium 4
(gathered with cat /proc/cpuinfo)

The Xeon machine is showing, for instance:
gsm
ulaw – 1601

The other shows:
gsm
ulaw – 2

Why are these values so different ?
Is it correct to say “if “core show translation” is showing a 4 digits value
in its matrix, then the translation path between corresponding codecs is
unusable.

Regards

3 thoughts on - Explain “core show translation”

  • _____

    href=”mailto:asterisk-users-bounces@lists.digium.com”>asterisk-users-bounces@lists.digium.com
    [mailto:asterisk-users-bounces@lists.digium.com] On Behalf Of Olivier
    Sent: Thursday, October 14, 2010 10:43 AM

    Hi,

    I’ve read in http://www.voip-info.org/wiki/view/Asterisk+codecs but I still
    have questions about “core show translation”.

    How are values replied by “core show translation” computed in the the first
    place ?
    I’ve got 2 machines using a 3GHz Intel CPU : 1 is a Xeon, 1 is a Pentium 4
    (gathered with cat /proc/cpuinfo)

    The Xeon machine is showing, for instance:
    gsm
    ulaw – 1601

    The other shows:
    gsm
    ulaw – 2

    Why are these values so different ?
    Is it correct to say “if “core show translation” is showing a 4 digits value
    in its matrix, then the translation path between corresponding codecs is
    unusable.

    Regards

    Perhaps the Xeon machine needs some optimization since a “ulaw-to-gsm”
    translation take 1.6 seconds for 1 second of data as opposed to .002 seconds
    on the P4 machine.

  • _____

    href=”mailto:asterisk-users-bounces@lists.digium.com”>asterisk-users-bounces@lists.digium.com
    [mailto:asterisk-users-bounces@lists.digium.com] On Behalf Of Olivier
    Sent: Thursday, October 14, 2010 11:11 AM

    2010/10/14 Danny Nicholas

    _____

    href=”mailto:asterisk-users-bounces@lists.digium.com”>asterisk-users-bounces@lists.digium.com
    [mailto:asterisk-users-bounces@lists.digium.com] On Behalf Of Olivier
    Sent: Thursday, October 14, 2010 10:43 AM

    Hi,

    I’ve read in http://www.voip-info.org/wiki/view/Asterisk+codecs but I still
    have questions about “core show translation”.

    How are values replied by “core show translation” computed in the the first
    place ?
    I’ve got 2 machines using a 3GHz Intel CPU : 1 is a Xeon, 1 is a Pentium 4
    (gathered with cat /proc/cpuinfo)

    The Xeon machine is showing, for instance:
    gsm
    ulaw – 1601

    The other shows:
    gsm
    ulaw – 2

    Why are these values so different ?
    Is it correct to say “if “core show translation” is showing a 4 digits value
    in its matrix, then the translation path between corresponding codecs is
    unusable.

    Regards

    Perhaps the Xeon machine needs some optimization since a “ulaw-to-gsm”
    translation take 1.6 seconds for 1 second of data as opposed to .002 seconds
    on the P4 machine.

    Where this data comes from in the first place ?
    Is it computed each time “core show translation” is typed ?

    What does “core show translation recalc 60” add to “core show translation” ?

    Ahh – questions that make me read. The “core show translation” invokes the
    translate.c module. If you do “c s t” it does a “one shot” display of the
    current values. If you do “c s t r 60” it recalculates and redisplays the
    values every 60 seconds. If your machine is a “variable load” state, you
    could get significantly different output. If your machine is running at a
    “20%” load, it’s probably not going to vary very much.