[e2e] 150ms - tolerable latency for quakeIII

Christian Huitema huitema at windows.microsoft.com
Fri May 25 08:47:07 PDT 2001

> -----Original Message-----
> From: Lloyd Wood [mailto:l.wood at eim.surrey.ac.uk]
> Sent: Friday, May 25, 2001 5:42 AM
> To: Jon Crowcroft
> Cc: Christian Huitema; end2end
> Subject: Re: [e2e] 150ms - tolerable latency for quakeIII
> On Fri, 25 May 2001, Jon Crowcroft wrote:
> > if you look at the game source code you would delete the idea that
> > game programmers "design" the game for any kind of network.
> >
> > network awareness in game programmres (at the level of detail we are
> > talking about, and the sort of syatmes understanfding that the X
> > had) is not widespread yet (there are exceptions, sure but not in
> > games we (tristan, manuel) are looking at or grenville ...
> As an example of this, see:
> http://www.gamasutra.com/features/19990903/lincroft_01.htm

Very interesting piece. It is actually straight in line with the
end-to-end argument and with "application level framing." The only thing
that is still missing is a feedback loop that would restrict sending to
the most important data when congestion is detected.

-- Christian Huitema

More information about the end2end-interest mailing list