Yeah I also don''t like the fact that EQs clerics and to a lesser extent druids and Paladins and a much lesser extent rangers are the only types of healers.

I fully agree that in a truly magical world, everyone should be able to find some way of healing themselves whether by spells from a cleric (true healer) or taking a bit out of that magical fruit. Just don''t make the items too easy to get or too powerful.

EQ tries to encourage grouping by making any areas with decent loot require a full group or more of characters at or above the average level required to kill a monster in there. A good alternative to getting groups going would be dungeons with a lesser degree of battle difficulty and more of a degree of solving how to get about said dungeon. Maybe a rogue to pick a lock, then the warrior to bash open a door, chanter to mezz the 999 headed fireant, wizard to twiddle his fingers and set torches on fire, etc.

Then make many dungeons places that hold stuff for every class required to enter them. That way clerics don''t have to pay wizards to light torches and the the wizard be in a dungeon with nothing good for him. That incraeses to a certain extent boredom. That way, as opposed to everquest where anyone any level can get into all but a few zones which require keys etc, you can limit who goes where by something other than raw power. Intellect plays a larger part of the game. It actually pays off to train in your skills and you get a character that plays more to your style. Not every warrior in the world is there becasue he''s the best offensive class. Many it would seem train in the arts of defense. So by that you then need a new method of agro management unlike EQ where agro is usually given to the person who does the most damage, or who casts teh most agro gaining spell.

There should be a way for tanks who specialize in being primarily defensive to get and maintain monster focus even though it isn''t doing the most damage. Maybe give these guys better skills at taunting. r something to that extent. Then you just don''t copmpletely alter the classes between offense and defense. So A pure offense warrior will deal much more damage in a shorter time and a pure defensive warrior will deal less damage in the same amount of time, the differnce beign offensives go for faster kills while defensives go for the longer battles. Then you just create enough monsters to keep everyone happy. It gets tough though since you''d think monsters will tend to attack the guy hitting the hardest in order to save themselves. I''ve thought about this for a while and have yet to come up with a good idea that I can''t shoot holes in within an hour of rethinking it. The bad part is if monsters go after the person doing the most damage it means they will focus on the offensve tanks who will presumably have fewer defensive skills becasue they learned in the ways of offense. So this would leave the defensive tanks doing nothing more than mediocre damage per unit time but can hold out longer. They wouldn''t be doing their job of being defensive. I guess you can simply make it so that defensive classes have better taunts to help ensure that they can get and control agro while offensive tanks and other classes can work.

There should be some monsters who just refuse to attack anyone but certain people if they are present. Maybe have monsters (smart ones anyway) realize who in the party is the most threat and attack that person. Kinda like the undead would immediately go for the cleric. Or Ice giants going after fire wizards. Again, this sort of thing can quickly throw a good system off balance.

Any ideas on this?


Webby



Defeatist Attitude: It simply cannot be done.
Realist Attitude: It can be done just not today.

These are the two realms of programming.

For balancing combat systems, get a decent working knowledge of basic probability and basic statistics. Then think about how you want the system to work, and then you can just plug the numbers in. It''s only a black art if you don''t spend the time to make it a science. This goes doubly for turn-based games where you have nice and discrete units of action.

I would advise that you don''t think about what stats apply to what. That''s the bottom-up approach which usually fails in game design. Instead, try top-down. First, think about how you want your game to play, then about how the different systems should accommodate that. From those systems, you should see where the variables are, and from those variables you can suggest stats for them.

[ MSVC Fixes | STL | SDL | Game AI | Sockets | C++ Faq Lite | Boost | Asking Questions | Organising code files ]

Thanks Kylotan

Yeah it was dorta botton-upish. I see what you mean though and my thoughts had kinda flowed both ways. I had moments when I have though "What skills need agility" And other moments where my thoughts were "What stat does this skill need"

As far as gameplay we''ve got a real good idea of how that is gonna run. we''re just into the pre system balancing, shooting all poossible hole in the theory to make it a better one etc. Thanks for the advice.

BTW Got cable modem today. IT ROCKS!!!
Download from 2.33 KB/sec to around 130KB/sec!!

Webby

Defeatist Attitude: It simply cannot be done.
Realist Attitude: It can be done just not today.

These are the two realms of programming.

quote:Original post by WebsiteWill
Or those massivly armored guards that you kill result in at best a wooden shield? If the NPC is wearing it, the NPC should DROP it. Anyway.

Sure the NPC should drop it. But there is one slight problem (with dropping armour anyway), you''ve just run your sword right through this guy and also through his armour. It''s going to be pretty well useless with a hole in the middle of it .

I seem to remember reading somewhere that, that was the reason very few (if any) enemies dropped armour in Fallout 1/2. The armour wasn''t really worth anything to the player after he''d just killed the previous guy wearing it with a rocket launcher.

Maybe, if you have a damage rating for armour (that reduces it''s effectiveness when it''s damaged a lot) then you could have every enemy drop exactly what they are carrying. Then while you are fighting them you can increase the amount of damage on the armour when you hit it. Of course doing this can get complicated very fast, especially if you factor in things like hits to different parts of the body (which can cause different damage to the enemy as well - which each have different chances to be hit). Then different bits of armour will get damaged. So if you really wanted to get that nice breastplate that guard is wearing try and hack his legs off so you won''t damage it too much.

Stats and genuine combat is a bit like the relationship between linear and exponential calculations, because on the one hand you have a direct conflict/relationship/outcome and on the other you have player choice guided by the uncertainty/expectation/avoidance principles. Is a solution like that of logarithms possible?

For all your talk of correctly systematizing combat interactions, you haven''t been able to deal with critical assessments that rely on psychological interaction between combatants. Blow for blow stats is a paradigm you need to break out of. How about calculating fear die-rolls based on pre-fight shows of strength? Or calculating aggression die-rolls based on successive misses? Of course, that merely drags psychology into the stats paradigm. What about the reverse? Gybrook''s insults in Monkey Island represented a move in the opposite direction (better insults lead to winning the fight, not better swordplay). The real answer, naturally, lies somewhere between: blow for blow stats to player attack choice to extrapolations from match statistics to emotional responses.

Perhaps, following from my comment about logarithms, you could give every stat a dual function of action/emotion such that successive strength attacks give way to higher adrenaline and so on, or where evidently aggressive attacks can be countered by attacks from a specific stat set designed to ridicule, etc.