Re: [NTLK] Newt and Internet Telephony

On May 23, 2006, at 4:08 AM, Morgan Aldridge wrote:

> Performance is really the main issue at hand, especially the
> performance of NIE (the Newton Internet Enabler). I think Adriano's
> initiative to update NIE (http://u.nie.notwen.com/) is going to be
> more and more important.
Hmmm, Do we know what the mac throughput for NIE on an MP2100 is? I
think a full duplex conversation with the GSM codec over SIP is less
then 30K bits per second. but not much less, so you may be right.

Also I have to point out, that Paul Guyot has already suggested that
with his Einstein project and "Relativity" the Newton emulator will
gain the ability to access native system calls, so the whole NIE thing
would go away moving in that direction. Did Paul's japanese demo
already happen?
>
> Someone could definitely write a SIP client for the Newton OS (I'd
> love one) and the processor should be fast enough for some of the
> lower quality codecs such as GSM, but the latency over ethernet and
> WiFi might be too much of an issue.
MIGHT be.
>
> Also, more processing power would be required to do echo
> cancellation, so it might push the limits of the Newton's processor.
Only true if you need an echo can. I can live without that function on
the Newton ;~)
>
> However, with VoIP becoming more common, this opens up a possibility
> to extend the Newton's usability quite a bit. If someone were to try
> and implement this, I'd love to see it done and would be willing to
> pay as well.

This would be a definite show off item and worthy of another splash of
web media attention. ie "Newton get softphone".

Marty

PS For any interested developers there are many open sourced soft
phones as well...

-- 
This is the NewtonTalk list - http://www.newtontalk.net/ for all inquiries
Official Newton FAQ: http://www.chuma.org/newton/faq/
WikiWikiNewt for all kinds of articles: http://tools.unna.org/wikiwikinewt/