[e2e] 150ms - tolerable latency for quakeIII

grenville armitage gja at ureach.com
Fri May 25 02:33:14 PDT 2001

Christian Huitema wrote:
> Grenville,
> I really wonder whether the delays that you report characterize the
> player's tolerance, or are merely a sampling of the delays observed by
> well connected actors on the Internet.

Fair question, but perhaps they're not so different. I'm assuming
that the dominance of players on my server from the west coast
of North America is not due entirely because the west coast contains
most of the online QuakeIII players in the world. This assumption may
indeed be flawed.

> It would be consistent with players
> investing on high speed connections, and thus getting reasonable
> service; there is actually no need to hypothesize that they are voting
> with their feet.

True, one cause of low ping times would be a prevalence of high
speed connections in various subsets of the online community.
But these aren't independent variables - FPS gamers are motivated
to get broadband service by the potential for better ping times to
game servers. It is the players who are missing that make me think
there's a rough tolerance limit on latency.

Note that my "voting with their feet" notion needs to be understood
in the context of how QIII players select servers. They generally
know nothing of a server's topological or geographical location,
only its apparent ping time. At any given moment on the Internet
there are hundreds of active QIII servers to choose from. A well-
connected player in New Jersey will see topologically 'east coast'
servers as having lower ping (and therefore being more desirable)
than my server, likewise well connected players in Europe would find
European servers more attractive than mine (other things being

> If you remove the "voting" hypothesis, then the hypothesis that 150 ms
> is some kind of magic number for games also goes.

Indeed true.

> In fact, we could draw
> another conclusion: that the tolerable delay is a function of the design
> of the game, and that the designers of the game take into account the
> prevalent conditions on the Internet to make sure that players get a
> reasonable experience. In short, the designers program for the network
> that most of their customers can get.

I was going to say something similar to Jon's observation. Game
designers have optimized to minimize packet size and trasmission
rates, but I don't believe they make any attempt to design the
game's algorithms for some notion of 'prevalent conditions' in
the sense of prevalent e2e latencies. As I understand it, the
history of online FPS gaming actually involved loosening the
synchronization of player actions precisely because the Internet's
prevalent conditions were (a) hard to predict, and (b) differed
from player to player.

Apologies if I entirely missed your point. Certainly the questions
are valid.

Grenville Armitage                http://members.home.net/garmitage/

More information about the end2end-interest mailing list