["Volker Steinmann" <steinmann@xxxxxx>]

Hi Lars,

in the attached capture file the RTP analysis for packet SSRC 1512435544
shows in Ethereal v0.9.13a:
packet nr. 42, delay 0.000072, jitter 0.112496
packet nr. 46, delay 0.087740, jitter 0.107016
packet nr. 54, delay 0.122276, jitter 0.104858
and in v0.10.3
packet nr. 42, delay 0.000072, jitter 0.112496
packet nr. 46, delay 0.087740, jitter 0.212481
packet nr. 54, delay 0.122276, jitter 0.304058
the jitter calculation for all following packets differs from v0.9.13.
How do you calculate the time of arrival in RTP timestamp units? 

Volker


> -----Original Message-----
> From: ethereal-users-bounces@xxxxxxxxxxxx 
> [mailto:ethereal-users-bounces@xxxxxxxxxxxx] On Behalf Of Lars Ruoff
> Sent: Tuesday, November 09, 2004 12:03 PM
> To: Ethereal user support
> Subject: Re: [Ethereal-users] RTP Stream Analysis
> 
> 
> Hi Volker,
> 
> J' starts as zero for the first iteration.
> At the end of each iteration, J' is replaced with the current 
> value of J. Can you post a capture file or a .csv-output of 
> values where you believe there is a calculation error?
> 
> regards,
> Lars
> 
> ----- Original Message ----- 
> From: "Volker Steinmann" <steinmann@xxxxxx>
> To: <ethereal-users@xxxxxxxxxxxx>
> Sent: Tuesday, November 09, 2004 11:41 AM
> Subject: Re: [Ethereal-users] RTP Stream Analysis
> 
> 
> > Hello Lars,
> >
> > thanks for your fast reply.
> > I know this formula, however I do not comprehend the real 
> values. Do 
> > you know which value does the RTP analysis use for (i 
> renamed it) J' 
> > in the formula: J=J'+(|D(i-1,i)|-J')/16
> >
> > Furthermore Ethereal calculates two different results for 
> the jitter 
> > using version 0.9.13 and the current version. It seems that the 
> > results of the old version are more realistic.
> >
> > Volker
> >
> >
> >
> >
> >
> >
> >
> > ----- Original Message -----
> > RTP delay and jitter are calculated as described in RFC 1889 
> > (http://www.ietf.org/rfc/rfc1889.txt),
> > that is:
> >
> > <quote>
> >    If Si is the RTP timestamp from packet i, and Ri is the time of
> >    arrival in RTP timestamp units for packet i, then for 
> two packets i
> >    and j, D may be expressed as
> >
> >                  D(i,j)=(Rj-Ri)-(Sj-Si)=(Rj-Sj)-(Ri-Si)
> >
> >    The interarrival jitter is calculated continuously as each data
> >    packet i is received from source SSRC_n, using this 
> difference D for
> >    that packet and the previous packet i-1 in order of arrival (not
> >    necessarily in sequence), according to the formula
> >
> >                     J=J+(|D(i-1,i)|-J)/16
> > </quote>
> >
> > regards,
> > Lars Ruoff
> >
> >
> > ----- Original Message -----
> > From: "Volker Steinmann" <steinmann@xxxxxx>
> > To: <ethereal-users@xxxxxxxxxxxx>
> > Sent: Thursday, November 04, 2004 5:39 PM
> > Subject: [Ethereal-users] RTP Stream Analysis
> >
> >
> > > Hello,
> > >
> > > does anyone know how jitter and delay is calculated in the 
> > > Statistics
> > -
> > > RTP Stream Analysis menu?
> > > We measured IP jitter using hardware measurement device and got 
> > > different results. While the device measured 150ms IP jitter, 
> > > Ethereal measured 500ms RTP jitter which does even not match with 
> > > the arrival time. As any RTP packet is mapped into an IP 
> packet the 
> > > jitter should
> > be
> > > similar.
> > >
> > > Thanks for your help.
> > >
> > > Volker
> > >
> >
> >
> > _______________________________________________
> > Ethereal-users mailing list
> > Ethereal-users@xxxxxxxxxxxx 
> > http://www.ethereal.com/mailman/listinfo/ethereal-users
> 
> _______________________________________________
> Ethereal-users mailing list
> Ethereal-users@xxxxxxxxxxxx 
> http://www.ethereal.com/mailman/listinfo/ether> eal-users
> 
>

Attachment: mp4_isma0_ethernet_client_2.eth
Description: Binary data