complex combat mechanics

[CygnusX]

Hi.  Still new here, but I'm about to get my feet wet.

I have a working theory on game mechanics.  When I first started programming, I thought I had to have a complicated combat script.  One that took in many variables and computed a paper-rock-scissors type algorithm to find the winner.  The more complicated (I thought), the better.

These days, I'm starting to experiment with a system that is just the opposite.  Instead of making a complicated system that makes balancing even more complicated, why not just fake it?

My thought is, after a users levels up his character, invests his points, buys weapons, armor, etc, either he can beat the next level bad guy or he can't.  So, why not dumb everything down and simply provide the illusion of a valid combat system?  I think I have a solution for this, but was curious to know everyone else's approach.

[SpaceDoG]

Not sure I understand. You want thoughts on a combat system you've come up with but haven't actually detailed out for us to get an idea of what you mean by "complex" and "illusion?"

[CygnusX]

I'm not asking for a critique on my system.  I think it would be more appropriate to code it and let people try it out as opposed to trying to explain the math (it should be simple enough to do after all).  At that point, I'll ask for a critique.  Rather, I'm asking if anyone else has tried to dumb down classic combat mechanics for a dungeon crawler, or does everyone still run some sort of modified D&D algorithm, a standard Damage = (Power - Armor) * AttackSpeed + Random function, etc, etc.

[SpaceDoG]

That's what I'm planning on running though my users won't know. They'll never have access to the formula and it'll be really complicated once the type of weapon, shielding, and other special equipment gets into play.

[133794m3r]

All formulas are most likely based off of the D&D like formula. How i've done mine to put it simply without all of the various variables and function calls is quite simply. If they manage to not miss, and the other person manages to not doge.  Damage = Attacker_Damage - Defender_Resistance. And of course there is a chance that they get a critical hit. All of us still use a system similar to the D&D one but i'm sure other games have done what you are going to do. Where you basically decide if they can win or not before actually calculating the battle. I personally like the feel of an intense battle so that's why i'm using all of the 27 skill trees, 3 armor classes, 10 types of weapons, 3 types of base damage, 6 types of magical damage, and various other temporary boosts for the users. Since if i was going to be playing it, i'd want to be able to do well almost anything and have loads of choices for my(the character).

If your game does combat via a click then i'd say do yours match them up see if player a has more hp, armor, and damage than b and if so let a win if not and the opposite is true make b win. If still nto then do a random choiec like thing to determine the winner or do a draw. It's about how integrated the combat is within your game. For mine thus far, it's a very important aspect of the game so thus it's going to be as complex as it is for a very good reason so that i can try to make all of the simple mathematic formulas look like different things when in reality they're all very similar and thus provide a higher level of smokes and mirrors.

[Sagefire135]

so, if you have "stats" of 10 and you are fighting someone with "stats" of 9 then the system gets rigged so that you win? that would be tough to accomplish with  more than 1 or 2 kinds of stats, if im even understanding the idea right .

if you just have str and agi then saying that 10 str+10 agi beats 10 str+9 agi works but is it even possible to fairly decide who wins when its 5 str+7 agi vs 8 str+4 agi?

its a decent idea, but coming up with a way to fake the system is probably more work than simple making a combat system.

[CygnusX]

I'll have to make a demo to show how it plays.  I'll post back to the site when its ready.

[raestlyn]

The Challenge system will the the next one I'll use. It will make complicated formula very simple. I actually have the script somewhere already.