[AG-TECH] The AGN Minimum Requirements Document

[Markus Buchhorn]

Nice work Ivan!

At 05:01 AM 26/02/2002 -0600, Ivan R. Judson wrote:
>As promised here is the document that described what an AG Node is.  Please
>read it over and provide feedback via ag-tech.

I agree with the other comments on the shopping list section at the end, 
and won't repeat them. The Gentner requirement is one I'd still like to 
test out, since our VC centre here has a superb arrangement for audio, 
which is using a (non Gentner) box I haven't identified yet. I'll get some 
more details.

One point though - the requirements on latency. Without identifying what 
kind of latency you mean, I think it's not helpful. Can I suggest it would 
be better to have "encoding and decoding" latencies, of the order you 
suggest (maybe a factor of two smaller) and separately mention transport 
latencies - which you can't in general do much about. Those of us in Oz are 
stuck with the Pacific rtt to the US no matter how direct we go :-) (until 
at least... http://www.satirewire.com/news/jan02/australia.shtml ). Photons 
in glass will only go so fast :-(

The cumulative latency for each stream impacts on the user-interaction, 
i.e. how smooth conversations go, and at high values this becomes a pain - 
the interaction becomes more formal. However, most users can really "only" 
impact the encode/decode latencies with better CPUs and add-on comp/decomp 
hardware.

Cheers,
         Markus


Markus Buchhorn, Information Infrastructure Services,   | Ph: +61 2 61258810
Markus.Buchhorn at anu.edu.au, mail: CompSci,CSIT Bldg #108|Fax: +61 2 61259805
Australian National University, Canberra 0200, Australia|Mobile: 0417 281429