[MUD-Dev] Quake II has gone GPL

shren shren at io.com
Wed Mar 6 08:12:18 CET 2002


On Tue, 5 Mar 2002, Joshua Judson Rosen wrote:
> On Thu, Feb 28, 2002 at 11:31:18AM -0600, shren wrote:
>> On Tue, 22 Jan 2002, Travis Nixon wrote:
>>> From: "Vincent Archer" <archer at frmug.org>

>>> Never trust the client.
 
>> You did a lot of snipping there.  Part of the text indicates why,
>> for the sake of constant movement, the server has to let the
>> client have some say over where he is.  It's not a short cut,
>> it's a requirement.  Otherwise, in a 3d enviornment, you can only
>> move as fast as the packet time.
 
> The server can still do validation of client-originated data and
> take accordant action if it looks bogus, like:
 
>         The client says that the character just moved 1 mile in 2
>         seconds, on foot, without any super-speed or teleportation
>         abilities. That looks bogus--reset his position to
>         something sane and alert the client.
 
> ... right?

Right.  If you don't do this then teleportation hacks are possible.
Even if you do this, however, you can still take advantage of the
behavior.  Note that UO, EQ, and AC (I think all three of them) all
had, at one point or another, cheat programs for them that gave you
fastwalk.

I even think I remember how the UO fastwalk worked.  UO used some
kind of query response system for walking.  Your client would say,
"I'm walking 4 steps", and the server would say, "you walk 4 steps",
and your client would wait for the server's response before sending
another walk request.  Fastwalk programs would send "you walk 4
steps" messages whenever it felt like, thus gaining the walk rate of
someone with a ping time of zero.  I could be wrong about this.
There are people here who would know much better than I would how it
worked.

Movement is hard to handle.  I keep expecting to see games appear
where the players play in some kind of enviornment where normal
movement rules don't apply, thus dodging this whole problem, like a
'cyberpunk decking' game.

_______________________________________________
MUD-Dev mailing list
MUD-Dev at kanga.nu
https://www.kanga.nu/lists/listinfo/mud-dev



More information about the mud-dev-archive mailing list