Re: 1 frame/10 minutes on a ref

Jason Williams (streak@ccwf.cc.utexas.edu)
Mon, 27 Oct 1997 17:54:59 -0600 (CST)


On Mon, 27 Oct 1997, Rolf Hemmerling wrote:
> But.. any technical aspect of a ref is not important if it is "busy" and
> the frame rates are 1 frames per 10 minutes.
=2E..
> So the bottle=B4s neck is just my 28.8 bps connection and the state of th=
e
> ref, not the traffic at my provider=B4s server and not the heavy traffic =
on
> the data highway between Germany/Europe and USA, like from=20
> 9.00..12.00..18.00..02.00 o=B4clock CET, and not the traffic at US noon=
=20
> (18.00..22.00 CET) !=20

No..my guess is the bottleneck is the state of your connection between you
and the reflectors you go to. It's either that or possibly a problem with
your rates (see my response to Ross' inquiry). If things work fine for
you sometimes and horrible for you at others, it's the fluctuations in the
net. If you get horrible reception on CU at ALL times, it's not
necessarily the net.

If the problems were at the reflector, others would complain as well. As
I did with the RichSys reflector. Me and a majority of others got
horrible rates from it, but at the same time others saw everyone else just
fine. I think I eventually chalked that one up to a bad spring router in
between me and the reflector. Since the reflector was multihomed, others
didn't come in from that router and weren't affected.

> I am personally updating my own private reflist, with a rating system
> concerening the number of people connected to it when I contacted to the
> ref.

Technically, I have a means to do that automatically for the 20 or so
reflectors authorized on my scanner..but it'd become a huge resource
drain.

> I would like to point at an important feature of CU: The text chat is
> *not* affected by any bandwidth problems (either audio or video)

That's not entirely true. If I recall, the chat window is the lowest
priority. First is audio, then video, then aux-data (chat).
I believe I remember reading somewhere that aux-data can only eat up a
maximum of 10% of the bandwidth.

> Btw, Gary, how about asking Streak to allow You to include *his* reflist
> in the shipped archive of Your newest WP CU-SEEME client ?

It is included in the 3.0 cdrom though rather indirectly. It's included
as part of Richard Cekal's cusdial CU dialer complete with a link to the
scanner.

White Pine hasn't been on top of the reflector scene in that aspect. I
can only assume the reflectors on their "Video Chat Directory" page are
the ones involved in their Public Reflector Network. I talked with
a woman at White Pine over a year ago about this and she said a web page
with the reflectors participating in the reflector network was a ways off
because she was tied down with work.

I'm also not sure White Pine would want to stay on top of it. It would
seem to me they'd only be interesting in advertising reflectors running
the White Pine reflector though I could be wrong.

> I think that Streak=B4s work by collecting all refs by his refscanner is
> really great ! But.. I heard from the WP staff, on the Hannover Fair
> CEBIT 1997, that they are *not* interested to do Streak=B4s work instead,
> they don=B4t think that this is *their* duty.

There's several ways to look at that. I'm sure White Pine is busy enough
with creating and marketing CU-SeeMe and their MeetingPoint reflector.
They do provide a means to connect and see others through their quite
limited PhoneBook. But they aren't obligated to provide a huge list of
reflectors to go to.

Of course, I wouldn't turn down an offer by White Pine to host my scanner
on their web server but as of yet, haven't received an offer like that
from them :)

--
streak@ccwf.cc.utexas.edu    * Jason Williams -- Austin, Tx.  |     |
streak@mail.utexas.edu       * University of Texas at Austin  | ___ |
streak@cs.utexas.edu         * BS Computer Science             \_|_/
*************** http://ccwf.cc.utexas.edu/~streak/ **************|