how-to block ads
an ATA which send Voice Packets (VoIP) via your ISP connection to the Primus network and finally to TBB servers in Toronto; the VoIP path. From there, your call can be connected to the PSTN number you have dialed, unless of course it is a call to another TBB subscriber.
Although you may wonder if the length (distance) of the VoIP path from your location to Toronto Server affects your TBB Call Quality, it is actually the High Latency, JITTER, and Packet Loss on this VoIP path that is more relevant.
As you can see from below, unless of course there is a TBB System wide problem affecting either the TBB servers or the VoIP path on the Primus network - which rarely occurs, it is actually your ISP connection that affects Call Quality the most.
In some cases, using TBB from a particular house/location over a given ISP may provide great quality. However, connecting the ATA to the same ISP, but at a different house/location (next door or across town), may provide terrible quality. This is all due to the particular manner in which the VoIP path is setup by the ISP in each case - and this is where JITTER in some portion(s) of the ISP VoIP path plays a major role in Call Degradation.
Hence, under normal circumstances, the fact that TBB servers are located in Toronto has no bearing on TBB Call Quality. It is generally a HIGH JITTER issue with your local ISP connection (neighbours also affected), an ISP regional network bottleneck affecting many users (city district), or even at the interconnection point to the Primus Network.
For example, EVEN if Primus had a TBB Gateway in each city, or across the street from where you live, it may still take 6 HOPS on your ISP network to get there - and if there is any HIGH JITTER in ANY of these ISP HOPS, you will have Call Degradation issues.
The VoIP Path
Consider the time VoIP packets spend travelling from your ATA to the TBB servers - through some wires and devices (routers etc), aka hops:
•The Medium - The transmission time through the wires (ethernet, coax, copper, fibre optic).
•The Hops - The time spent within the devices (modem, routers, switches).
The Medium - the Wires
For IP packets, distance (across Canada anyway) is not a big factor since an electrical signal - or light on fibre optics - can cross the span from St. John's (NF) to Victoria (BC), 7300km, in less than 30ms!!.
This transmission time is part of latency; any total latency less than 300ms round trip (150ms one way) is OK. In other words the transmission time can be considered to be low overall (see the communications section in »en.wikipedia.org/wiki/Speed_of_light ).
The "Hops" - Devices
IP Packets also spend their time in devices where they are switched or routed onto different paths/cables etc.
Each time packets go through a device, depending on the quality of the router, it can take from .1ms to 1ms to get through. In some cases, there can be special software (data shaping, etc) to slow down packets at these points.
The other factor here is if a particular router has too many inputs that it becomes congested at times, then this can be a major problem and add serious delay (30-100ms+) and JITTER to incoming packets (like having too many onramps onto an expressway). This all comes down to how an ISP builds their network - and some can be cheap in forcing too many wires/packets through some routers to save cost etc.
The end result is that practically speaking, the time it takes to go through HOPS is more important since this can vary widely; transmission time is not really a concern (it is generally a fixed time of <30ms).
So the more HOPS you have between your ATA and the TBB servers is the more important issue - not the distance.
ISPs generally have "agreements" to peer (i.e. interconnect) at certain points. This simplifies the ISP networks in that they do not have to connect absolutely in every city - also minimizes cost. So depending on which ISP you have, they interconnect to the Primus Network somewhere convenient (traceroutes may show you this).