On Thu, 13 Dec 2001, John Mariotti wrote: >While this is hard to implement on a starting mud do to poor player turn >out, I completely support this type of system and have already put the code >in for my MUD, however to compensate for lack of players (No the MUD isn't >up yet) when it goes live you have to have a contingency plan ready to go >active based on your average player count and have the economy adjust >accordingly.
I'm still not sure about methods for fine-tuning or making a self-tuning system, but a start might be to base extraction timers on how many players are active, or rate of new resources coming in. Both fairly easy to track. As for worst-case scenarios (no player activity & me wanting to watch to system run), figure solution there can just be a reboot. At boot (or repop, to allow for players to do mean things) time, mobs should have enough resources to keep themselves busy for a short while. >Also to make things a bit more realistic, I put in the ability for shop mobs >to have 'free will' when it comes to buying or selling an object based on a >lot of different factors and also if they do decide to buy/sell will base >the price on a wide variety of factors, for example Faction, Race, >Occupation, Luck, Charm, Economy, Is there a War going on and if so is >what's being sold/bought a high wartime commodity or a low end civilian >fruit toy, time of year, whether the item is currently in demand in another >part of the world (For my import/export code), and a few others as well. >However, while _no_ economy can ever be perfect, simply for the fact that >it's far to hard to emulate everything that could possibly happen, so you >have to either have a hard and strict economy that stays the same regardless >of what happens, or hope your player base works towards altering their >economy. I like. Dunno if I want to get that complicated on individual transactions (especially mob-to-mob sales, since they should be frequent when players active...) but might take some ideas from you there. >Well I addressed some of this above, but game balance is always a highly >debatable topic. IMHO it's easy to achieve balance in ALL aspects of your >MUD.....if you take the time to figure it out and have a few good writers by >your side. (Fortunately I have one =D ) Don't tell me who they are, I might try to co-opt them. (kidding!) >I feel for game balance you need to take your entire game world into >perspective and then break it all down to the smallest level, figure out >what would make the world tick, then add responsibly from there. Have a partially completed storyline I want to use as the background. Much work to do from there... >Just try not to get carried away because like, you and I said above, you >have to code what you create. Everything sounds perfect on paper, it's a >completely different story when your sitting at your computer staring at >a file and not have a clue as to what you should be doing (I know I've >been there many a time, and still frequent that imbecilic mode quite often) >But I learn best by trial and error, so that works for me *shrugs* Yes, yes, planning vs. reality. More effort in planning can help avoid these things, but overplanning ... think pretty much anyone learning to code (and a few with too much experience) drop into idiot mode that from time to time :) Don't know for sure if it's be better to have the entire system try to be self tuning, or everything hard & fast, with the polayers having to ry & play by the rules. At the moment, leaning towards self-tuning, but suspect that the end result of attempts towards tuning won't be completely successful. Then there's more consideration... robustness : If you only have a few locations where each type of product comes from, or a limited number of traders (all good, so far as CPU & memory usage are concerned), a small group of dedicated players could keep the economy in chaos & ruin fairly easily. Hrm... perhaps that could be the backing premise for a guild or two :P (storyline mentioned previously doesn't lend itself to defining what guilds should be, outside of religious grounds) >John Thanks for the feedback. Helps me clarify a few things already :) Closing with a few things that occurred to me after I posted (maybe should have waited a day or three to post). --- Not mentioned before, but MOB to MOB intereactions should have some flavor to them. Trader greets merchant as enters, etc. Probably best done with mprogs, current mprog system should be just fine with few mods... looser checking on mob-to-mob interactions... Corpse scavengers - one per MOB death is overkill. Keep counter for corpses in an area. When counter gets too high (threshold set per area in AEDIT?), send out scavenger to clean up the area. PC deaths should warrent higher priority, so either one scavenger per, or a unique scavenger who only handles PC corpses. Think along the lines of a MOB pushing a cart moving down the trail screaming "Bring out your dead!" and ringing a gong.. Scavenger for NPC corpses should likely be big and ugly... mean enough that it'd take a small group of hero-level characters to take it down. Maybe invis/hidden/sneaking as well. Should have 'pass door' affect, and ignore EX_NOPASS flag. "No doors are barred to the Agents of Death". Try think under what circumstances will players will (benifit from | try to) clean an area out and lay an ambush for the scavenger... MOB tracking room to room - Instead of running track for each move, run track when MOB chooses a new destination room & store path. simply take a step higher along path each pulse. This lowers CPU use per pulse. Tracking to players/mobs (and objects if/when added) should, of course, stay the same and recheck the path at each step. Later, d.c. | dwight a campbell - [EMAIL PROTECTED] - http://rats.darktech.org - KD7KJG | | yet another linux guru in training & reality avoidance therapist at large | | "When you live in a sick society, just about everything you do is wrong." |
