Wireshark-users: [Wireshark-users] Interpreting the difference graph for a VoIP call
>From what I understand, the difference is the time that elapsed between sending two packets.
When I analyze the RTP packets, looking at the difference of each individual packet that a person sends out, the difference seems to be about 20ms for each packet.
Now, when I look at the difference graph, (I set tick interval=.001sec, pixelspertick=1), the red vertical lines appear at intervals of about 20ms on the time as expected. However, along the vertical (difference) axis, only about 1 in every 8 red lines reach the values of about of 20ms. The rest are very low (1-5ms). I thought that each red line's vertical value should be 20ms.
So what I am asking is what does the vertical axis of the difference graph represent since it does not seem to be the difference?
I am even more confused because when I look at the jitter graph, the vertical axis is the jitter and reflects the data.
Here is a second question:
There is something that confuses me a bit.
I asked about why when a person is not speaking, he is still sending packets to the other person. The explanation was that the microphone continuously produces a voltage signal regardless of whether the person is talking and so that voltage signal is continuously packaged into packets and sent every 20ms to the recipient.
And I asked about why that there are times when a person is not sending any packets over a while. The answer was:
"Clever suppression techniques avoid 'empty' packets from being
transmitted, but have to be accounted for at both ends."
From this answer, I reasoned that sometimes packets are not sent because of suppression techniques that avoid sending "empty" packets. But I do not understand the part about having to be accounted for at both ends." How is it accounted for?
and my third question:
Since there are times that a person does not speak and still sends out packets, and there are times when that person does not send any packet, is this because the suppression techniques are triggered only at certain times (for example when a person does not speak for a minimum specified duration)?
What I am trying to find out here is why there are times that "empty" packets are sent and other times not.
