Tony Langdon wrote to Brian Rogers <=-
Cool. Yeah I've been around VoIP too long, where packets are tiny and overheads are big as a result, and you pretty much can't go below
around 20kbps on the wire, because of the overheads, even if the
payload is only 2400bps. :)
On 05-21-21 00:07, Brian Rogers wrote to Tony Langdon <=-
Tony Langdon wrote to Brian Rogers <=-
Cool. Yeah I've been around VoIP too long, where packets are tiny and overheads are big as a result, and you pretty much can't go below
around 20kbps on the wire, because of the overheads, even if the
payload is only 2400bps. :)
I wish VoIP used TCP like IAX instead of UDP/Sip :\
Tony Langdon wrote to Brian Rogers <=-
So a dropped packet would cause it to get stuck, until it was replaced?
;) IAX is 4569/UDP (or I've been doing port forwarding wrong for years
;) ). One of the reasons VoIP uses such small packets, is latency. :)
On 05-22-21 20:39, Brian Rogers wrote to Tony Langdon <=-
Not at all. You just have a better flow. IAX is 4569/TCP however you
can use UDP - why is beyond me though. I have 0 jitter on my line
*most* of the time so I don't have an issue with SIP.
Tony Langdon wrote to Brian Rogers <=-
Every implementation I've seen uses UDP. Maybe IAX _can_ use TCP, but
I haven't seen it used that way.
Yeah jitter is much lower these days.
Sysop: | altere |
---|---|
Location: | Houston, TX |
Users: | 69 |
Nodes: | 4 (0 / 4) |
Uptime: | 10:49:20 |
Calls: | 1,160 |
Calls today: | 4 |
Files: | 8,176 |
Messages: | 300,620 |