On 10/13/05, Whisper <[EMAIL PROTECTED]> wrote: > -- > [ Picked text/plain from multipart/alternative ] > Well my point was, there are some what are in retrospect quite easy to > detect/block cheats out there.
No, it looks easy, thats my point :) You are looking at it from a purely geometic point of view, but in fact, it's only robust within the confines of the engine (remember, game levels can vary immensely, as can their effects on a player) if you fix the problem from a player point of view. Example: portals leading to kicks due to rapid level traversal. What about teleports? ok, you code those in, but now can you tell me off the top of your head, does mani teleport like beetlesmod? and does that teleport like nemod? the answer isn't important, the principle is, you're looking at the wrong data to try and diagnose a cheat. thusly, it's not as easy as you think. > If a cheat has to normalise itself to the level where it plays as well as a > normal human, well it almost defeats the purpose of using cheats in the > first place. It's not about normalisation, it's about fooling the inference engine or logical rules. > The thought of offloading all cheat detection onto another server, was > something that was brought up in the HackCam forums (R.I.P. HackCam - We > never really knew you but you promised so much) was to off-load all cheat > detection processing onto a completely different server, thus allowing the > actual game server to process game data only and allow the cheat detection > server to deal with cheat detection / player data calculations since it is > not entirely necessary for cheat detection to occur in real-time, unlike the > actual game, as cheats can be dealt with at a later stage once you have > concrete proof that a player is in fact cheating. I was actually talking about stats processing, player data handling, log processing, the works, but ok. > BTW, if it was up to me, I'd force cl_interpolate 1 & cl_interp 0.1 on > Source for ALL players, I play with those settings and I get plenty of kills > on pub servers. I don't like watching things die 5 or 6 frames late, or regulating over that time period. Changes to cl_interp are only as significant as having a better or worse ping. Player A with cl_interp 0.08 and a ping of 40ms is not dissimlar from player B with cl_interp 0.1 and a ping of 20ms. command packets should arrive at the server at around the same time. The source documentation suggests that cmd packet arrival time is unimportant however, so it *shouldn't* matter provided that sv_maxunlag is at a reasonable level. A min and max interp should be available server side though I would say, as it should be in a common range accross competing players. I get far more enjoyment out of the game when running cl_interp 0.05 on a nearby server. model and gameworld "stutter" have unpleasant effects on my ability to play properly. (too much quake). _______________________________________________ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds
