[VoIP] DP-compatible ATAs

[Steph Kerman]

Shane Young wrote:
> Quoting Steph Kerman <stfkerman at jps.net>:
>> True but I believe that machines that inherently have metallic 2W 
>> trunks should be connected directly together and not go back and 
>> forth between 2W analog and 4W digital any more times than necessary. 
>> As long as my DID trunk is delivered on analog copper, I won't be 
>> inserting an * between the PSTN and my switches for this reason. I 
>> have no objection to inserting an * between an inherently 4W digital 
>> medium like CNET and my switches.
>> In the context of the switching environment under discussion, that's 
>> true as long as you don't object to going back and forth 2W analog 
>> and 4W digital N times.
> So, let me see if I can summerize your thoughts into this sentance.
> "If you are going to do it, do it right."
Correct.  And rather than be too doctrinaire, I'll say this is my own 
personal perception of what is right.  I make no declarations about what 
others should do.  Everyone is dealing with a different situation: 
different toys available to play with and has a different satisfaction 
threshhold.  However the CNET calls I've made so far through Greg's 212 
PSTN portal don't sound all that wonderful to me. 

So while I know that it was the policy of the Bell System to inject 
noise into early CX systems to mask Xtalk, and they strove to equalize 
levels so that user's would not encounter large level differences from 
call to call since that would generate dissatisfaction with the lossier 
calls, I prefer not to drag my internal network down to Voip quality as 
it exists over the public internet circa 2007.  I certainly want to keep 
my real PSTN connections 2W metallic since they arrive that way.  If 
they arrived via ISDN BRI in 4W digital form I might see things quite 
differently.  There's no chance they will arrive over PRI any time soon.
> That is how I feel.
I had that perception about you...
>> I thought * had some built-in echo canc. capability. Is it too lame 
>> to be useful?
> I've had problems with it. It is supposed to be better now in 1.4, 
> I'll have to try it :)
>
> I don't recall what it's echo canceling ability is on an all-IP call 
> (internet to ATA is all IP and the ATA's are bad at echo canceling)
If there is no 2W segment in the call because it is 4W-IP end-end, echo 
should be non-existent.  Echo results from the 2W-4W junctions.  
Multiple 2W-4W junctions creates multiple echo paths.  It would not 
surprise me if that made the echo cancelers less effective but I have no 
expertise on that.

Another reason for my choice not to put hundreds of $ into a T1 card at 
this time is that the price is bound to plummet.  There is absolutely no 
point in stockpiling stuff like this for future use, even 6 months out, 
IMO.  Stockpile stuff that is becoming more scarce, not less.

Steph
>
>
>
>