Member Discussions

I'm curious, why isn't anyone interested in implementing a turn-based combat system in their muds? I mean, every single mud i've visited and tried had the same boring type-kill-and-go-away-for-10-minutes. I'd like to see a combatsystem similar to the good old console rpgs' combatsystems.

It exists in plenty of muds.. The ones that I've played recently are Final Fantasy Compendium, or Final Fantasy Tactics: Online.

They aren't the only ones, just the ones I've visited most recently.

It's not true nobody's interested. There are muds that don't have an automated combat system, and some of these systems are quite complex.

See here:
http://www.mudconnect.com/discuss/discuss.cgi?mode=MSG&area=adv_code&message=9973

Turn based in the style of standard RPGs is indeed rare but the "boring type-kill-and-go-for-10-minutes..." style of fighting is stock combat and unless you have visited a mud that has changed some of its stock code thats what you will get. Thats not to say all muds are like that- you are either very unfortunate or very bad at choosing which muds are worth your time. More sophisticated muds have updated their combat with various ways to keep the action intense but I am reticent to name them in this forum... all well I will endure the scorn of the forum-nazis to help out a fellow enthusiast, but next time be conscious of what forum you are in before asking such a question lest you too become the object of their ridicule.

disclaimer- I am not affiliated with any of these games nor do I even play them anymore, they are just good examples that I know for modernized nonstock combat.

One of the better games to modify stock combat in such a way to keep it interesting is the line of games that fall under the Toril/Sojourn spinoffs- these include Basternae, Duris, and Sharune and probably a few others. The concept behind their combat is stock enough to be familiar to any Diku player who logs on but I guarantee if you leave after typing kill you will never win a fight. They use all the dynamics of fighting in a very exciting way and promote group oriented combat such that all combatants are required to be on their toes and to fullfill a role assigned to them be it healer, tank, basher, nuker, etc. One slip up could mean disaster for the entire group. If you are looking for good stock combat kicked up a notch in sophistication I recommend these.

The other option is a concept that emulates turn based very closely and is reminiscent of the final fantasy active battle systems. In this concept the player is given attack points, attack speed, or an equilibrium statistic and performs attacks only when an exact command is sent to the mud for it, and that attack will offset their equilibrium or cost them attack points which then must be replenished before they can attack again. This system I beleive is superb for those who really like RP because the battle is controlled very much by the players while still being kept fair by the code, if you want to stop in the middle of fighting and intimidate your opponent with witty banter- you can. Nice in theory but kind of complicated in practice the muds that have used this system best are those that fall under the Achaea list, including Aetolia and I think Imperian but Im not sure. Also Alien vs Predator (AVP) has a similar system but that game doesnt play like your traditional mud and is more like a mud version of an FPS.

I'm sure there are other muds out there which have enhanced their combat code but those are the ones I have played and enjoyed at one time or another, hope this helps.

> I'm curious, why isn't anyone interested in implementing
> a turn-based combat system in their muds?

Some people are, but personally I find they slow combat down too much. Sitting around and waiting for your opponent to make their next move just doesn't give the sort of fast-paced action I'm trying to achieve.

> I mean, every single mud i've visited and tried had the
> same boring type-kill-and-go-away-for-10-minutes.

There are many more alternatives to turn-based combat than just typing "kill" and getting a coffee.

> I'd like to see a combatsystem similar to the good old
> console rpgs' combatsystems.

Unfortunately those games were designed for single player games. I don't think you'd have found them so fun if each computer opponent took 30 seconds to decide what it was going to do.

Actually, i've played an online multiplayer rpg with a system similar to nearly every console-rpg and to me it seemed like the perfect combatsystem. Even in 5vs5 fights it worked very fast. There was a 15 second timer when you had to choose your moves and if you didn't, it would set you automatically to dodge. The difference was that the battles were sometimes resolved in like 2 or 3 turns and battles taking over 5 turns were very rare unless you were alone fighting NPCs, but then again it was faster cause NPCs didn't have to think.

And thanks for everyone posting some examples of muds with a different system. My mud experiences have pretty much been with what I had found here in the new mud listing and every mud in there was pretty much the same...

Some friendly advice for all mud-hunters:
The New Mud listing is great and when I was mud-hunting I used to check that all the time looking for the "next great startup mud" what I found instead, however, was that they were all pretty much: "I just downloaded the codebase and decided to add myself to TMC while I work on the game." So yes if all you did was check the new listings thats probably what you got. Instead, what I found more useful was the Updated Mud listings- I found that recently updated muds whose owners cared enough to notify TMC were muds that the owner had put a lot of time into and recently got most of the customazation completed, enough to re-introduce themselves to the TMC community. There were exceptions in both cases but thats generally how it goes.

My staff and I have been talking about this vary topic for quite sometime. After much debate and design over how it should operate (commands, timers, etc), we've decided to go through with it. The problem is, I'm somewhat at a loss as where to start. Any suggestions would be welcome.

Done that before, one of the best things to keep players interessed, at least when it comes to combat. There a few things you must decide before attempting to code it. First should be how combat will be handled, I've implemented 3 types already and I don't really know if people came up with something diferent than those:

1. Turn based
Each character gets a number of actions based on his combat speed, faster characters go first and damage/messages/whatever are handled as soon as the character types the command. Diferent commands/skills take a diferent number of action, but the character can only use a command once on his phase, then it's everyone else. If the character still has actions, he gets another phase inside the turn and so on until the turn is over.

2. Continuous Flow
No turns, no phases, each skill/command incurs in a time delay minus the character's combat speed. Commands and skills happen as soon as they're sent. For an example; Combat starts with time 0, faster character goes first and does a slash movement with his sword, which has a delay of 5, the enemy is in a defensive stance and got to block with his shield automaticaly, which increased his delay by 2, he then counters with his mace and tries to smash the player, increasing his delay to 2+5=7 ... Since everyone acted you go right to the next delay which is 5, player goes again and then on 7 it's the enemy.
This is a nice type of combat system, I prefer it over the turn based one but of course, it's not as simple as turn based combat. You have a lot of things to balance.

3. Reactive combat
Turn based, but the faster character goes last. Commands are -not- processed as they come, they're left to be dealed with on the end of the turn. This is similar to Ars Magica. The slowest character claims his action first, the other characters may or may not get to see what he's attempting to do, those who saw it may actualy do something about it like defending themselves against that attack or countering it or whatever. After everyone had their actions in the turn then the mud can handle all the commands to see what actualy happened, who was happy with their actions and who realized they just screwed up. This is the most fun system to play with in my opnion. It's pretty much based on Ars Magica which can be downloaded for free (legaly too :p) on their site.


As for delay, 10-20 seconds are enough on mosts cases, depending on how complex your combat is. If it has movement, such as a map where the player can walk in while in their turn, you should prolly use 30-60 seconds.

> 1. Turn based
> Each character gets a number of actions based on his combat
> speed, faster characters go first and damage/messages/
> whatever are handled as soon as the character types the
> command.

Hold on...if the actions are handled as soon as the character types a command, then how do the faster characters go first? Or do you mean the faster players go first (i.e., because they can type faster)? Or do you mean that everyone has to wait until the fastest player has typed their action, then the next fastest, and so on?

> Diferent commands/skills take a diferent number of action,
> but the character can only use a command once on his phase,
> then it's everyone else. If the character still has actions,
> he gets another phase inside the turn and so on until the
> turn is over.

So each character only gets one action per phase? What if I'm using a rapier and my opponent is using a sledge hammer? Do I only get one attack off before he can hit me?

> 2. Continuous Flow
> No turns, no phases, each skill/command incurs in a time
> delay minus the character's combat speed. Commands and
> skills happen as soon as they're sent.

That's similar to my system, except that I give every attack a fixed time delay, and instead give players a variable number of action points based on their combat speed. The advantage of this is that it seriously reduces any spamming advantages - one player might be able to enter attacks faster than another, but he's not going to earn any more action points, and therefore won't be able to attack any more frequently.

Hold on...if the actions are handled as soon as the character types a command, then how do the faster characters go first? Or do you mean the faster players go first (i.e., because they can type faster)? Or do you mean that everyone has to wait until the fastest player has typed their action, then the next fastest, and so on?

Sorry, I should have mentioned that the mud will hold input back, at least those related to combat, from anyone else who is in that especific combat. So the mud would tell which character goes first and wait, lets say, from 10 to 20 seconds for that character to type his action and as soon as he does, the hit success, damage and whatever is relevant takes place.

So each character only gets one action per phase? What if I'm using a rapier and my opponent is using a sledge hammer? Do I only get one attack off before he can hit me?

I was being generic but the coder can obviously specify the rules for that. But you can make it so characters wielding faster weapons get more phases or actions per phase than a character wielding a heavy weapon. It's not realistic and brings back memories from D&D which has the most basic turn based ruleset. It's simplier to implement tho, rom/diku is essencialy turn based but it's handled automaticaly.

That's similar to my system, except that I give every attack a fixed time delay, and instead give players a variable number of action points based on their combat speed. The advantage of this is that it seriously reduces any spamming advantages - one player might be able to enter attacks faster than another, but he's not going to earn any more action points, and therefore won't be able to attack any more frequently.

Yes, I should have mentioned that. On my original idea for that system, the character would still have to type his action when the mud told him too and not leave it open so the laggers get screwed over.

Actually, at a Moment in Tyme, we are trying to do just that. We've been working on it for quite some time now, just haven't had the coding staff to implement much of it yet. If you're interested in our implementation, feel free to drop by and talk with either me (Duren) or Nicodemus.

tyme.envy.com 6969

-Duren

I'm curious...

On Turn-Based MUDs, is the PK feasible? Is it more strategic? In which ways?

Does killing a small rabbit - a task that should be simple, become menial and boring when having to go through an entire turn-based system?

I'm asking because i'm interested in a variant from the run of the mill stock "kill" system.

Anyone with this sort of system in have any idea what has worked well for them? And what pitfalls did they encounter?

> Sorry, I should have mentioned that the mud will hold input
> back, at least those related to combat, from anyone else
> who is in that especific combat. So the mud would tell
> which character goes first and wait, lets say, from 10 to
> 20 seconds for that character to type his action and as
> soon as he does, the hit success, damage and whatever is
> relevant takes place.

So if, say, 10 characters were involved in a large-scale melee, the slowest character would have to wait between 1.5 and 3 minutes (for 10 or 20 second turns respectively) before he could type anything - for each and every combat round?

Or am I misreading what you're suggesting?


[about "Continuous Flow" combat]:
> Yes, I should have mentioned that. On my original idea for
> that system, the character would still have to type his
> action when the mud told him too and not leave it open so
> the laggers get screwed over.

If the character has to type his action when the mud tells him, then surely that's not "Continuous Flow"? In fact, isn't that exactly the same as the turn-based model discussed in the first paragraph?

> On Turn-Based MUDs, is the PK feasible? Is it more
> strategic? In which ways?

I imagine you could create a very strategic turn-based PK mud - after all, there are many turn-based strategy board games, not to mention various turn-based RTS games.

> Does killing a small rabbit - a task that should be simple,
> become menial and boring when having to go through an
> entire turn-based system?

IMO the problem, as I mentioned in another post, is not with mobs but with human players. Think of chess, a solid turn-based strategy game between two players...

When playing against an AI opponent, it tends to make its moves extremely quickly, and it doesn't mind how long you take to decide your actions.

But playing against another human can quickly get annoying if that person spends ages on each move - and equally they will get annoyed at you if you take ages to decide your own moves. For this reason you typically need some sort of timer, and even then it can get irritating having to keep waiting between moves.

The same logic can be extended to the mud, and is then expanded because of the multi-player aspect. Fighting a rabbit shouldn't be a problem, because the AI can work out its moves with lightning speed, and really doesn't mind how long you take. But fighting against a player adds a waiting factor - and battles between multiple PCs is going to be even worse.

But it really depends on what you're trying to achieve. Are you trying to create chess-like strategy, or Tribes/Counterstrike-like strategy? Do you want strategy to be purely about out-thinking your opponent, or should it also require fast-thinking?

Personally I'm trying to balance clever strategy with fast-paced action. A good strategy prepared in advance gives you an edge, but so do quick reflexes and the ability to adapt to changes at a moments notice.


In response to irwinr12: Rather than just slapping an advert into the middle of the discussion, how about joining in and telling us about your design, and its pros/cons?

Heh. You ever get rid of that cheating Narsil coder/Rand player chap?

-D

So if, say, 10 characters were involved in a large-scale melee, the slowest character would have to wait between 1.5 and 3 minutes (for 10 or 20 second turns respectively) before he could type anything - for each and every combat round?

Or am I misreading what you're suggesting?

You're not misreading. 1.5 to 3 minutes is actualy fast for 10 characters. If you play in a system where combat is quite deadly (such as shadowrun) it only takes from 3 to 5 turns for it to be over. Ars magica takes around the same ammount of turns. I'm used with roleplayed combat which takes 10 minutes a turn with only 3 characters. -That- is slow.
The idea is that the player has 10 to 20 seconds to type, not that he will have to wait 10 or 20 seconds for his phase to be complete. From my previous experience such system working with shadowrun rules and 5 characters would take only 1 minute per turn and roughtly 2 turns per character to be over.

It is indeed slower than automated combat or combat which has time delay, but It's more fulfilling in my opinion.

If the character has to type his action when the mud tells him, then surely that's not "Continuous Flow"? In fact, isn't that exactly the same as the turn-based model discussed in the first paragraph?
Like I said, it was like that on my original idea, which was still turn based except that it had no fixed turns along a time-line. I actualy coded it later similar to your own mini-mud code, gladiator I think, with some time delay. Since I always have around 2 seconds of lag, i never did well on it. If you considers your players to be lag-less, you don't have to pause and wait for their actions, you can simply block the combat commands until they can act again. Or do as you said and use action points which are given every x seconds.

But playing against another human can quickly get annoying if that person spends ages on each move
Thus I give them 20 seconds or so. Actualy, on the mud I'm actualy coding, the player can set his wait preference from 10 to 60 seconds, the mud calculates the average of all players involved in combat and then uses that wait timer. Not only that, players who have been idle for too long will act automaticaly as best as the mud dumb-ai can handle them until they type something to warn they're around. Therefore you don't have to wait a century just to find out the other player is idle.

Do you want strategy to be purely about out-thinking your opponent, or should it also require fast-thinking?
Both, 10-20 seconds aint exactly enough time for you think where you're gonna move (yes, i use maps normaly, but not on what I'm coding right now) and what you're gonna use from a tons of manouvers and options.

I like combat to be meaningful and not something D&D-like where you do hundreds of damage and then a simple punch on the end knocks the enemy over. My current project is a Shadowrun mud solely based on the Matrix and the Decker aspect of the game.

I've been discussing this with the other co-owner, and this is a (very basic) idea of the sort of thing we want to do.

Each player has a certain amount of action points per-round. (Player with the most action points goes first). The player can chose to attack, change stances, use items, etc. Each of these actions requires a certain amount of action points. For instance, if the character has 8 action points, and wishes to take a defensive stance (+1 to def. lets say for this example), he can defend 8 times for a total of +8 to his defense. Otherwise, he can spend all those 8 points on one of his attacks that costs 8 points, or two attacks that cost 4 apiece. I like this method especially because it incorporates all the skills a character has into attack, not just the most powerful one.

A few kinks i've been thinking about are: what happens to any trailing action points you don't have anything to do with, if a player has an insane amount of action points, they can be doing 2-4 times the number of attacks as the other player, so how much *points* and costs of the action should.

Can anyone think of any problems a combat system like this would entail? If you have experience with this sort of stuff, does this sound possible (and, in your experience, is it enjoyable?)

Does anyone think this would be a good system?

Feedback greatly apreciated :D

> Each player has a certain amount of action points per-round.
> (Player with the most action points goes first). The player
> can chose to attack, change stances, use items, etc. Each
> of these actions requires a certain amount of action points.
> For instance, if the character has 8 action points, and
> wishes to take a defensive stance (+1 to def. lets say for
> this example), he can defend 8 times for a total of +8 to
> his defense. Otherwise, he can spend all those 8 points on
> one of his attacks that costs 8 points, or two attacks that
> cost 4 apiece.

That sounds more like a dice-pool system, like that used some RPGs (the White Wolf ones, The Riddle of Steel, and various others).

In my old White Wolf-based Diku I had something similar - each player had a dice pool based on their stats and fighting skills, and could choose how to distribute their dice among attack, parry and dodge, by specifying a minimum, ideal, amount used and maximum number of dice for each category. The actual number of dice in each would then be calculated depending on the priorities you'd set, and as the dice trickled back over time they would refill the most important things first.

You could also configure your actions in other ways, such as defining which hand should be used to attack and which to defend, or deciding to go full attack (using the defence dice pool for an offhand attack) or full defence.

Although the combat system was still automated, it allowed a great deal of fighting style customisation. For example one tactic I found particular good was to put about half of my dice into dodge, one die into my primary hand and the rest into my secondary, then specify both hands to attack. The first (really rubbish one-die) attack then worked as a feign, forcing the opponent to waste their defence, and opening the way for the secondary attack. People then started developing counters, then counter-counters, and so on. Strategically it worked quite well.

However it was a completely different style to the action point system I was referring to previously - with the approach I use now, a character earns 1-26 action points per second, based on various factors such as stats, encumbrance and fighting style, and stops earning them once they reach 999. Each combat command then costs around 50-125 action points, and results in a fighting technique (or sometimes a combination of fighting techniques) being executed. Defences are automatic, but are adjusted by each particular technique - so a shield block will give a better chance of stopping an incoming attack than a shield bash, while many of the more aggressive techniques will weaken or even cancel your defence with that location (for example it's more difficult to dodge if you've just kicked someone). It then becomes a case of balancing attack against defence, speed against power, armour against maneuverability, etc.

That seems like an interesting idea.

Another thing we were tossing around was the ability to attack specific areas. For instance, you can choose to take a swipe for the head (harder to hit, but deals more damage), as opposed to going for the body (bigger target, but usually you've got a big plate of armor there). Size could also come into play here too.

Does anyone think this is viable?

> Another thing we were tossing around was the ability to
> attack specific areas. For instance, you can choose to
> take a swipe for the head (harder to hit, but deals more
> damage), as opposed to going for the body (bigger target,
> but usually you've got a big plate of armor there).

That's something I'd been considering for my older combat system (the dice pool one), but never got around to adding. I think it would work pretty well, as long as it's well balanced.

I decided not to bother implementing it for my current system though, because many of the more complex combat techniques already target specific locations - eg "Kiss of the Viper" is aimed at the throat, "Goring Horns of the Bull" is aimed at the body, "Embrace of the Serpent" is aimed at the legs, and so on. These techniques often wouldn't make so much sense if they could be aimed at other locations, and allowing that sort of customisation for the more generic moves such as "slash" and "thrust" would take away some of the value of the more advanced moves.