Ascend Archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

(ASCEND) Quake and latency on Ascend




As a result of the discussion about LQM here, I was looking at some of the
behavior of the MAXs when they are under load.  Not overall load, but per
session load.

Long ago I had noticed that MAXs have a tendency to have ping times rise
ever higher before they start losing packets.  This reminded me of an
issue where Quake players would see a fair ping (300ms) from "Qspy", but a
yucky 3000ms ping from Quake when actually playing.  I had the theory that
the difference could have been from the RAS buffering packets instead of
dropping them.  Whether I was right isn't my question.

The question.  Which is better:  Dropping overflow packets immediately, or
buffering them for God knows how many seconds?

I understand the MAX may find buffering better because of the possibility
of having more bandwidth just a few B-channels away.  But is that the
reason?  Are there better ways?

----------------- Brian  Curnow ----------------

++ Ascend Users Mailing List ++
To unsubscribe:	send unsubscribe to ascend-users-request@bungi.com
To get FAQ'd:	<http://www.nealis.net/ascend/faq>


Follow-Ups: