HP3000-L Archives

August 2003, Week 2

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
"Atwood, Tim (DVM)" <[log in to unmask]>
Reply To:
Atwood, Tim (DVM)
Date:
Mon, 11 Aug 2003 13:25:30 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (71 lines)
Yep. that's me. Always too much of a dreamer/idealist to believe in the laws
of physics.

I want to live on an island on the BC gulf coast. Want to live in the middle
of a forest without annoying ugly power and phone lines. Live amongst all
this beauty and still try to make money doing computer work. I want it all!

So I get annoyed when I spend hours travelling to clients. Dang, why don't
they hurry up and invent that instant teleportation Star Trek keeps
promising! And this pesky speed-of-light thing. When will we have those
subspace communications so we can have really fast Internet?

Come on - If Bill Gates can convince the world Windows is a great operating
system we should be able to get around the space-time continuum.

-----Original Message-----
From: Gavin Scott [mailto:[log in to unmask]]
Sent: Monday, August 11, 2003 1:01 PM
To: [log in to unmask]
Subject: Re: response time problem using Reflections or qcterm


Tim writes:
> I wonder if it is a similar problem to what I have experienced with VPN
> through Hughes' Direcway two-way satellite broadband.

It's that damn slow-as-molasses speed-of-light.  Gets you every time.

> "Running a VPN client over a satellite network is not an ideal
> configuration. Although most VPN clients will work, your speeds will be
> affected significantly. While average download speeds are slightly better
> than dial-up, they will be reduced from typical DIRECWAY speeds by as much
> as 50 to 75 percent. Average upload speeds are comparable to dial-up
> performance. "

I don't know precisely how things like PPTP etc. handle acknowledgements,
but I'd be very surprised if there was any simple request-response stuff
going on that wouldn't be able to take advantage of TCP's sliding window
acknowledgement protocol.  I typically see very little difference between
VPN and non-VPN throughput.

It's possible that the satellite link is so slow that even the normal TCP
"window" isn't large enough, and the connection effectively becomes latency
limited as it has to stop and wait for ACKs to catch up.

Of course this should be another ideal application of Wirt's Advanced Telnet
with QCterm, since that will give you the benefits of local-echo without
losing the character-mode interactivity.

Sounds like Hughes is playing some tricks to help eliminate the problem, but
in general I would expect throughput to be reasonable (even over VPN) but
for any request-response protocol like full-duplex echo, the physics just
isn't on your side:

> "Our communication satellite is located over 22,000 miles from Earth.

So up and down for each character to reach the host then up and back for the
echo to reach you, equals an 88,000 mile round trip.  88,000 miles divided
by the speed-of-light, 186,000 miles/second, gives you 0.47 seconds of delay
just waiting for the signal to propagate.

Clearly you need a faster Universe.

G.

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

ATOM RSS1 RSS2