Lots of players love Midievia's combat system and lots of players love combat in Aardwolf, and many other large MUDs. Part of the reason is the size of the playerbase itself and not necessarily the PvP system. Rarely in PvP/PK forums or conventions or development threads is IRE's combat system brought up as an influence for creating a good PvP MUD.

Obviously it works for the purposes of you making a lot of money, but that does not make it the best product out there, nor does it mean that MUD-PvP should encourage scripting (or a PvP system that encourages scripting) merely to appease the players as a whole, as opposed to those dedicated soley towards PvP.

I'm quite experienced in player-killing myself, but that's really beside the point. The fact is, scripting is simply faster than a human. If something is parsable, it is scriptable. Text that gets communicated in a known format is exceptionally parsable and thus exceptionally scriptable. If reaction times are an issue (and you stated that you believe that PvP comes down to reaction times), humans lose even worse than in chess.

--matt

There are many possible ways to show your talent and reflexes is merely one of those. And scripting limits a lot of things, most importantly, it reduces the amount of responsibility you take for PvP, therefor limiting the rush and the notion of danger. This is a topic that gets brought up in PK threads and conventions again and again and again, and the same thing is concluded.

Do you have data on this? Or is this completely anecdotal?

So a system which makes the players happy.. isn't a good one?

Oh? Darn, I guess I didn't realize people didn't like our PvP combat system.

--matt

It most certainly has no effect on your responsibility when it comes to PvP. The actions of your script are the actions of your self.

Well! That sure puts me in my place doesn't it! Some people on some random PvP forum somewhere aren't talking about our PvP system but the creator of MUDs is in his published book on MUDs. Consider me suitably chastised. Next time I'm speaking at a games conference, I'll be sure to look for you in the audience.

--matt

A client-script will always react and act in the same way, whereas a human will not. I react and PK indistinguishably as fast as a script, so it is possible to improve your reaction time to the point where you should not need scripts. However, the predictability of a script allows for me to decipher how an opponent's script works, walk them into traps, plan against them, and do some pretty funny things to them in general.

After now 14 years of PvP (I started at age of eight, mercilessly killing other people instead of the monster in Hunt the Wumpus until nobody would enter "The Cave", then moved to the then-free Gemstone games), I cannot even begin to count the number of times and ways that I have seen scripts get people killed. There's a reason for that.

Beyond just the feasability of a heavy-scripter PKer against a flexible, strategic, creative, fast-acting PvPer; this goes back to the center of the current discussion which is that it is possible, at the designing table, to focus your PvP system on the things that give the players the most rush, and to discourage elements that take away from that (including reliance on scripting).

Some elements are certainly heavily influenced, much like Avalon borrowed heavily from previous games.

You haven't seen a complex script then. No, it most certainly isn't as predictable as you think it is and it most certainly won't react in the very same way every time, even when it's not practical for it to do so.

And I've seen scripts do incredible things to keep people alive. My own script occasionally makes me go "Wow, I don't even remember putting that in!"

Despite what you might think, even with scripting, IRE fights are a rush. Scripting doesn't numb the senses. It just saves you some of the work so you can concentrate on aspects of fighting you deem more important. A fight can still end within 5 seconds of you thinking you were doing quite well.

Your idea of what a script is is pretty limited I think. If you believe a script's rationality of behavior is a liability, you may want to look at why Kasparov lost.

This is a pointless argument with a foregone conclusion that nobody has been able to intelligently disagree with for at least a decade (though I know I tried, futile though it was). If you wish to conceive of scripts as these repetitive, easily-fooled things, that is your right, but I'm not going to argue about it.

--matt

It wasn't my point to put you in your place. My point was that the monetary success of your game does not mean that it is looked up to as a desirable system for PvP - because it isn't, typically, for whatever reason. I do not think your system is bad - it is certainly better than most MUDs that are PvP clueless in their design - however, that is not the point of this conversation. The point is MUD Combat systems, and in particular at the moment, how a MUD should approach reliance on scripting.

As far as my reference to conferences, I was referring to MUD-specific ones that are usually held online via a server. There are several of them annually, and they've been around for years. If you hunt around, you can probably find the logs involving PK-Design and development yourself if you are so skeptical.

For someone who is so successful financially from your capitalistic MUD ventures, you are so overly sensitive, Matt. :-p

-----

Har;


A PvP system that is designed for the average MUDer is not going to be as good/innovative/etc, typically, as a PvP system that is designed specifically for people who enjoy PvP. I think that is a simple enough concept.

The script acts the same way every time. Using a script, therefor, means that in the individual moment, you do not have to take any decision-making responsibility as your script is already pre-written, which therefor reduces "rush". Again, that idea is a pretty simple enough concept.

On that note, can we get back on topic?

What are the different and innovative combat systems out there?

EDIT: Apparently not. Here we go again.

You make a baseless claim that IRE's combat system is bad and you back it up with nothing but phantom claims of hidden RP design conventions. What you write is your personal opinion and quite honestly it doesn't hold quite as much weight as you're implying. It would be especially good if you could justify your statement in some way.

A huge selling point of IRE games is the combat system. Many many people play precisely because of it. If you happen to think that all those people are wrong, it would be kind of you to outline why.

You're not very experienced with scripts then. A script behaves the way you write it to.

It's probably been said already in this thread, but the only effective way to reduce peoples reliance on scripting is to introduce random elements to to make their scripts less effective, such as critical failures, or spoofable combat text.

Despite all the needless obfuscation, IRE's combat system is rather mechanical and pedestrian and lends itself excellently to botting. I am guessing this is largely due to how they make their money. Nobody wants to spend thousands of dollars to max their skills and get all the leet gear just to get pwnd by bad dice rolls or emote spam.

We're not talking about most combat systems though - we're talking about complex combat systems. And more specifically, in this case, we're talking about complex combat systems which don't have an element of chance, in which case such correlations are particularly likely to be taken into account in order to increase the number of permutations so as to avoid being overly predictable.

Not necessarily. I show you an ASCII pictures of a flower - could you write a script to figure out that it's a flower? I show you a sequence of words (eg "lion, tiger, wolf and panther") - could you write a script to tell me that 'wolf' is the odd one out? There are numerous things that humans can instinctively do which computers have a great deal of trouble with. Technically they could be scripted, but I highly doubt any mud player would be able to (they'd need to create a custom client specially for it as well).

You certainly can reduce it - indeed you even go on to say that you can "design the combat system to not benefit from it particularly", so I'm not sure why you're trying to suggest with the above.

Scripts can require a great deal of creativity and tactical thinking - just not necessarily that of the player using them. I enjoy competitive games (including chess, although I dislike playing the same person repeatedly because of the predictable outcome). But I wouldn't enjoy playing chess if my opponent was using a chess game on his computer to calculate each of the moves - and the same is true for a PK situation in a mud. I might as well just play against the computer.

The Magic the Gathering card game includes a number of elements which I consider a good model to follow, and which I took into account in my own design. Your setup (deck choice) is critical, but copying someone else's deck isn't going to do you much good if you lack the skill to use it properly. Then there's also the element of chance, which you can manipulate in your favour by hedging your bets (eg, don't rely too heavily on any one tactic, have the flexibility to adapt even if you can't get the right cards, etc).

I only speak from many years of experience of PvP on more MUDs than I could ever hope to count, including Aetolia (and to a lesser extent, Achaea), to the smallest Smaugs, to Midievia and many MUDs.

One can use another's scripts against them - it is not an arguable fact, it happens all the time. And if a script is advanced(I have, for fun, written plenty complexed combat bots and am quite good at scripting, from my days of designing leveling/questing/mini-game bots), you can still learn how it works. No matter how complicated, it will never have the immediate innovation of a real-life opponent. PvP is not chess - it is not as immediately mathematical/logical, and thinking of it as a chess game is a testament to why your thought-process is flawed to begin with.

I agree. Its system works well for it, because it is a commercial system meant to make money.

Hard; re-read my posts. I stated that IRE's system is better than most. However, and KaVir has said this, we are speaking specifically of complex combat systems. IRE's system, while complex, is not immediately considered a role-model for PvP MUDs because it is based around commercial marketability, giving it a different goal than the majority of most other combat focused MUDs.

---

I have to bow out of this conversation for a few hours and get some food, but KaVir's back now (and holds a similar philosophy and stance on these subjects as myself) so I am sure that it will continue. Beyond a few instances of us bickering, though, this is a very good thread, so I hope it keeps rolling.

I agree. Even a combat system completely devoid of chance still is very unlikely to have correlations by all of those variables though - quite a few simply wouldn't make sense. To script it, the player still doesn't need to calculate every possible permutation, but rather to pick out the permutations that are significan enough to respond to. Even if for instance health is tied to walking speed, a player could perhaps safely ignore that if combat is typically conducted in a single location.

The same can said about the burden on the MUD coders though. Are you going to make a new ASCII picture every day for every different event? Are you going to have new word puzzles out every new week? Until you can make the burden on the scripter exponentially higher than that of the MUD coder there will be no reason not to script. You can make scripting harder, but you can't prevent it.

I meant that you can't actually stop a player from scripting. You can try and make it more difficult, but they'll always have their client and the ability to try and script. That's why it was followed by the reducing of benefit suggestion.

There are usually too many non-uniform variables to deal with to get predictable outcomes. And a great deal of what a player does is usually left unscripted due to it not being too practical. Often times scripting is left to just handle the defence while the offence is manual.

I disagree. The level of complexity makes it very open to innovation and creativity. The system isn't nearly as mechanical or easy to script as you might think - some 400k of python code and I still haven't covered nearly as much of it as I'd like to.

And what is the goal of the majority of other combat focused MUDs if not player enjoyment?

ASCII pictures can be generated - there are plenty of programs which do that already. You could even generate ASCII images of sequences of characters and/or letters, add a fuzzyness factor so they're never the same, and display them out to the screen - it can be automated without much difficulty.

Trying to write a script to read that image, however, is far from easy. Unfortunately, for the same reason, blind players would also have serious difficulties.

They only take a few seconds to write - why not? You could put together a database of them, perhaps even find an external source where you can grab the data from.

True - except in a few rare cases, scripting will always be possible and provide at least some advantage. However you can certainly reduce the value of scripting so that it merely helps the human, rather than replaces them. I believe you can also reach the point where a skilled player can do just as well without scripts.

In the end, if it's significantly important somebody will do it and spread it around - the algorythm to read the ASCII images, that is. You also have to wonder if having such constantly changing and odd syntax won't do more damage to your game than scripting ever will.

Granted. But if you only change one line, that won't be significant enough unless that line is very, very crucial. And then you have to consider that just changing the trigger is actually faster than thinking up the riddle in the first place. Unless you can make the difference in burden be very, very much in favor of the MUD coders, it'll only be a race to change triggers versus changing trigger-lines.

Yes, I completely agree here. Balancing things so the maximum number of players can take part in the activity while keeping it exciting at the same time.

IRE's systems are designed with the same immediate goal in mind as others, I'd imagine: Entertain people. We're just more successful at it than most.

--matt

For sure. There's just no getting around that.

I've been thinking about the assertion in that sentence and I'm not sure that's true unless the system is solvable, in which case the optimum decision at any point in the decision tree is already known, so human and computer can follow that branch with equal facility. I would tend to say solved games make for bad game experiences, but that's probably only true for the people who are aware of the solution and capable of applying it to the game. Lots of people still like checkers after all.

On the other hand, you have the example of Go, in which humans still kill computers. It's inevitable that computers are going to surpass human Go players, but the equipment isn't there yet. So, do you think that it's possible for a human to match a script of unlimited processing power (for the sake of argument) in an unsolved formatted-text PvP system? I would tend to think it isn't.

--matt

If you force combat decisions to be highly circumstancial you may meet a point where the speed of the computer script doesn't give an advantage over a player's own flexible thought.

For instance, you're battling on the edge of a cliff - clearly if you manage to push your opponent to the edge you'll gain an advantage, but scripting this would be unnecessary unless speed of commands was a deciding factor and if you battled over vastly different landscapes with vastly different circumstances, it might be possible to reach a point where scripting is more of a waste of time than a good investment of it.

Scripting usually happens to automate things which are too fast and usually repetative while things like tactics don't usually get scripted - hence to reduce the impact of scripting the battle system would have to be geared more towards tactical rather than reflexive play. In such a combat system, often the player won't gain much of an advantage over other players via scripts, providing that the results of the tactical decisions aren't completely obvious before they're made.

Sure, as long as combat involves guessing which type of flower is represented in the ascii picture that pops up on your client before your opponent does.

I for one can't wait for this forthcoming new revolution in combat.

-H

Everwar, Utopia, and other MUDs I've worked on and alongside(and I assume MUDs such as Godwars II and many others that I have less immediate experience with) have their combat system's immediate goal be to challenge PKers in an innovative way, using a balanced and player-skill oriented system.

Whereas skill is certainly a factor in any online PvP game, including IRE games, there are many things about IRE's combat system that does, indeed, show its goal to be to entertain the masses and turn a profit, as opposed to creating the most balanced, player-skill oriented, innovative and challenging system that it can.

You cannot assume all games have the same goal.



I believe KaVir was trying to using ascii picture generation as an example of an idea that could be worked with and implemented to reduce player combat scripting.

This is going half a dozen pages back at this point, but I wanted to respond to it.

The interesting thing to do here as a game designer is to put an opportunity cost on eating a whatnot. To take the simplest example that comes immediately to my mind, put in something like 3E+ D&D's attack of opportunity concept in. You eat a whatnot, I get a free chance to crack you for it as your guard drops.

If it's balanced well, eating the whatnot to cure your poison is sometimes the right answer and sometimes the wrong one. These kinds of situations are harder to script.

Free-flowing chaotic PK is generally harder to script than arena-style duel PK or PvE. A dynamic environment is harder to script than a static one, and so on.

There are lots of things a game designer can do to promote human thought and ingenunity over scripts. There's clearly some advantage to be gained by scripting in almost any game, but by the same token, I think most good games minimalize it, either intentionally or as a simple side-effect of their style of gameplay.

What about being forced to fight multiple targets?

Sure you can script chess, its a 1 on 1 and you can see every piece on the board. But what about say, poker against 7 people? Then you have a whole bunch of unknowns, and in addition, there's the strategy of bluffing, going all-in, etc.

I'd think its years away before anyone's making a zmud plugin to clean out a full team of skilled poker players, if ever. Surely it can't be that hard to base a mud combat system on it.

-----
As for the combat systems, I guess I'll mention the warrior at Abandoned Realms. It is a diku derivative with the usual auto-combat, hitroll, thac0, bash/kick etc, with more complex manual elements & skills. e.g. .. the three combat styles (2H, dual wield, shield) are good vs one style, poor vers another. And weapon types (blades-sword/dagger, shafts-axe/mace/spear, segments-flail/whip) are good vs one type, poor vers another.

Who would be behind those multiple targets though? Players? In team combat in most combat systems that I know it's best to take out the enemy one by one instead of splitting your targetting, so that usually doesn't make fighting much more complex so far as targetting goes.

And if they aren't players, they'd have to be MUD-run, which in the end will be predictable enough to script around.

We aren't talking about IRE players, though, we are talking about complex MUD combat systems - if IRE has fostered a combat environment that encourages or causes a large amount of reliance on scripting, then that is very different than what most PvP MUDs which I have encountered(which is to say, a very, very, very large amount), as most PvP focused MUDs try to discourage scripting for the previously posted reasons.

Yes, it might be fun for the MUDer to write scripts and compare scripts - however, again, the target there is for the average MUDer and not necessarily the MUDer seeking a game whose point itself is the PvP combat. As most PvPers will say(and they do say in forums and online conventions), heavy use of scripting certainly takes away from the adrenaline rush and immediate excitement.

Naturally, opinions on this will differ - especially if you are used to PvP on a MUD that fosters the use of heavy scripting.

It is obvious that you and I will differ in our opinions of whether or not heavy scripting is a positive force for a MUD combat system, but you do not even need to look further than this very thread to see that others agree that heavy scripting can (and should) be discouraged for advanced combat systems.

IRE players are just like any other MUD players. I don't see how you can try to somehow discriminate against them and say the ones who enjoy PvP in IRE are different from other PvPers on other games.

Most PvP MUDs try to limit scripting because:
A) That's the intuitive thing to do.
B) Because not all players will want to or will be capable of scripting and hence that will lead to a drop in population.

While B is indeed factual, A is simply assumed. I don't know about another MUD that benefits as much from scripting as IRE MUDs do, however the players who come to them for PvP (and they are a large part of the demographic) do enjoy them as such and will actually defend the way combat works, scripts and all. The pleasure a player gets from scripting and then putting their script and their skill to the test is different from what most people think of when talking about PvP MUDs, but it isn't irrelevant.

That assumes there is an optimum decision.  What about a system in which there are numerous viable decisions at each point, each with their own pros and cons?  With the random element, no decision is perfect, and some may well be more risky than others.  The script would be able to avoid the bad choices, but there would be no optimal good one.

Actually this rather reminds me of the .  Building on Greggen's RPS example, imagine a very simple combat system based on the following five options:

1. Thrust (advantage against slash and feint)
2. Slash (advantage against pummel and feint)
3. Pummel (advantage against thrust, equal against feint)
4. Riposte (advantage against slash, thrust and pummel)
5. Feint (advantage against riposte, equal against pummel)

There's no way to know what move your opponent is performing.  Each move takes 5 seconds to execute, although the defence part (for the purposes of advantage vs disadvantage) is in place the moment the command is entered.  Your base chance to hit is 15%, +5% for every second you wait before entering the command, up to a maximum of 40% after five seconds.  Your chance to hit is doubled if you have the advantage and halved if you have the disadvantage.

Which is the optimal choice?  What is the optimal time to wait before launching your attack?  How would you set up your script to fight in such a situation?

Given unlimited processing power you could script anything - but obviously nobody has unlimited processing power.  I think it's possible to create a reasonable combat system that cannot be feasible scripted to be better than a human player.  People who use a mixture of scripting and regular play will most likely still have an advantage, but you can certainly force the scripting to take a secondary role.

The point I was making is that not everything can be feasibly handled through scripts.

Interestingly enough my 'war' combat system is based on poker, and does often result in battles between large numbers of players - and, indeed, I've yet to see anyone script it.  Admittedly it's turn-based (3 actions every 60 second turn), but even so, people will sometimes miss actions - so obviously thinking speed is still a factor.

That's assuming the only way to fight is to inflict pure damage, though.  What if you want to stun one of your opponents so that he can't heal his friends?  What if you've got an attack which allows you to hurl one of your opponent's at another, allowing you to injure two people at once?  What if you're fighting on the move, and only have to inflict enough damage to knock your opponents off their horses?  What if you're able to inflict damage-over-time (poison, bleeding wounds, etc), and can therefore do a larger amount of overall damage by splitting your blows between your opponents (once one opponent is poisoned you move on to the next)?

Then there's also the question of defence.  Maybe you want to focus on one opponent, but you're still going to have multiple people focusing on you.  Will you go pure offense in the hope of evening the odds as quickly as possible (or at least taking someone down with you)?  Go pure defence in the hope that an ally can reach you before you die?  Assume some sort of balance between the two, so that you can survive long enough to inflict reasonable damage?

If decisions have totally unpredictable outcomes then the only thing that would really count as skill is intuition. Could you say that you're very good at RPS even if you play it a lot? A combat system, in my opinion, has to give you a lot of room to grow and perfect your skills - if the system is reduced to each move having a no more predictable effect than the other then you're operating on pure intuition, not skill or experience. And the moment you reduce the ambiguity of the decision making, a script system can be made to make the best decisions based on the present knowledge.

Oh certainly, if you change the mechanics around you can cause situations that are more and more difficult (but not impossible) to script. That's very much in line with what I was suggesting about making the circumstances vary too wildly and putting the focus on the player's flexible thinking rather than reaction time.

What I actually didn't agree with is that simply introducing multiple targets would make a difference. You can have multiple targets or a single target and still be equally limited in your 'best' decisions. However, you can have multiple targets or a single target and still have a wide array of 'good' decisions all with their long-term tactial pros and cons.

I'm not discriminating against IRE players, as I have been, myself an IRE player. I am however making the point that the reason why scripting is so easily accepted as a possitive force in IRE is because that the system and administration itself fosters a pro combat-scripting environment.


KaVir - you should check out the link below;



It is a matrix based dueling system set up to work in a similar way to what you are describing in very basic theory. It has, however, been running for over 15 years now, and supports a very large playerbase. It won't be the thing for most PvPers, but its system can certainly be taken into consideration when developing one's own system, and it could also potentially make a solid system for a friendly-dueling mini-game.

Our own system, with Utopia, gives each attack command a "focus" on the opponent's body. By using an attack command that has an opposing focus, players can set-up combos and high-combos. That is a very very reader's digest version of the Utopia focus/counter system, but it certainly makes use of a form of matrix, or "rock-paper-scissor" using system.

. However what I described is far less predictable than RPS.

But I'm not suggesting no predictable effect. Even the highly simplified example system I posted provides you with a number of ways to try and outwit your opponent. There's certainly an element of chance, but you have ways to turn it to your advantage, through a mixture of second-guessing your opponent and deliberately biding your time.

Only if there is a 'best decision'. Thus my example.

Yes, but I believe the point Davairus was making is that introducing multiple targets dramatically increases the complexity required for a script.

As I've said before, from a theoretical point of view anything could eventually be scripted (even the creation of a mud). But from a practical point of view, many things simply cannot be done, or are not feasible, and the advantages gained from what is possible can potentially be minimal.

Rock-paper-scissors is the extension we made to AR's diku pvp combat system. I guess I'll call it that now I know it has a name.

That's actually not true - there have been several changes set up precisely to make scripting more difficult. The best example of this was changing a trigger line for a particular affliction so it was particularly difficult to catch with regular expressions. I'm fairly sure Matt has spoken out against scripting in his MUDs. It isn't that the administration is actively supporting it, it's that the combat system in it's current form provides many benefits to scripting players. I'm fairly sure that wasn't intended in the first place. However, it seems to have worked out in it's own favorable way.


But is second-guessing an expression of skill or just intuition? If you don't have good options and bad options then the player can't make mistakes.

Eventually there should be a 'best decision', but one that is not linearly deduced - rather one that is part of a flexible combat strategy. The problem is with making that decision more difficult to script than is practical and removing the desire to script in the first place by keeping the pace to something that is humanly manageable.

What I'm saying is that this isn't necessarily the case. You can raise the complexity by introducint multiple targets, but only if you implement the already discussed flexible decision making as a part of choosing the target. If you have a 'best' target then adding more targets changes little.

Most certainly, that is true. However, I feel it isn't so much that effort has to be made in restricting the player's ability to script something, but rather in making the combat system exciting without making it so the benefits of scripting are particularly large.

It's the same sort of skill used for games like poker - you learn your opponent's patterns.

There can still be good options and bad options - the trick is not to have a best option.

I strongly disagree - a 'best' option results in linear gameplay.  You can have some options being better at certain things, or providing advantages in certain situations, but there should never be an option which is the all-round 'best'.

The most accurate attack might also be slow.  The fastest attack might also inflict the least damage.  The most damaging attack might also leave your defences wide open.  A knockdown attack might inflict no damage, but leave the enemy open to followup attacks.  A bash attack might do very little damage, but disrupt your opponent's moves.  A disarm attack might have a low chance of success and inflict no damage, but leave your opponent without a weapon.  Yet none of these moves are innately better or worse than the others.

To use the chess analogy again: What is the 'best' opening?

Once again you're using the word 'best'.  Who's the best target, the warrior who's hitting you or the cleric who's healing the warrior?  What if the cleric has spells which deflect your strongest attacks?  What if your weaker attacks result in damage which the cleric can't heal?

Multiple targets results in more tactical options, which in turn increase the difficulty of scripting.

Perhaps this is personal preference speaking now, but I don't particularly like the idea of a combat system functioning under rules similar to poker rules. While there is surely skill in poker, a large part of it is trying to stack probability in your favor and then depending on chance. If one player is significantly more skilled than another, I feel that that player should win consistently until the other works up to the his/her level. Losing because you got a proverbial 'bad hand' would cheapen the experience, for me at least. It's possible that this is down to personal preference, of course.

I'll use the chess analogy to demonstrate what I mean. The 'best' opening in chess is the one that is most in line with your strategy. The 'best' move in chess is the one that moves you further to your goal in a long and complex, dynamic sequence that changes along with your opponent's moves. I don't think chess would be a good game if every time you managed to maneuvre your pieces in a strategical way, you had to roll to see if you actually get to take the opponent's piece - that lowers the value of strategic thinking and instead takes too much control out of the player's hands.

I'm saying that so long as you have a 'best' target, the number of targets doesn't matter. Multiple targets don't innately cause more complexity, they only do so if the choice of target flexible instead of linear, which isn't implied automatically.

I said multiple players increase the amount of unknowns (hands, and playing styles). Uh, when a system has more interwoven unknowns in it, I call that a more complex system.

Are you using a different definition of complex here?

BTW, the target in poker is the pot, not another player.

It will usually make things more complex to some degree, yes. We're not talking about a specific system though. In a system where damage is the main fighting component multiple targets wouldn't really mean all that much since usually all attacks will be directed at a single target. In a different system picking a target can be a very strategical and dynamic choice, of course, I'm not disputing that. I'm just saying that multiple targets versus single targets don't make the difference between having a flexible choice and having a linear one - it's not implied.

And the point about poker was unrelated to the multiple targets point, I'm sorry if I wasn't too clear on that.

Well as I mentioned previously, I've created a minigame "combat system" roughly based on the poker rules, and it has resulted in some very skillful and tactical play.  People will occasionally draw bad hands, but probably no more often than rolling a fumble in your standard tabletop roleplaying game (and even if you do draw a bad hand, there are ways to get around it).

However if you've played chess, you'll know that there are often a number of different strategies at each decision point.  Do you take a pawn, knowing there's nothing they can do about it?  Do you sacrifice your own pawn, strengthening your position on the board?  Do you take a risk which, if they don't notice what you're doing, will give you a quick checkmate?  None of these moves are innately 'better' than the others.

I disagree, although obviously it's a matter of personal preference.  Most wargames are little more than an advanced version of "chess with dice", after all, and yet they can allow for a great deal of strategic thinking.

Indeed, if you wish to better simulate a combat situation, the element of chance is pretty much the only way to take into consideration the countless tiny factors which your game can't realistically simulate.  The wind blows a fly into your eye at a critical moment, the blood and sweat weakens your grip on your sword, your opponent slips on a patch of mud, and so on.

And I'm saying that there isn't necessarily going to be a 'best' target.

Unless combat is designed to only support one-on-one fights, there will always be an element of choice involved.  Even if there's an obvious target for you to focus your attacks on (which is by no means guaranteed), you'll still have to handle your defences against the multiple attackers.

I've played and enjoyed some games that were pure scripting exercises. Having a decently strong computer science / AI background, I generally do well in them. It's just not what I personally find enjoyable in a MUD.

There's a lot you can do as a game designer to limit the strategy of ganging up on one guy in a group vs. group fight. For example, maybe the more people are trying to focus on one opponent, the more they get in each other's way and prevent attacks, making it a matter of diminishing returns.

If it's possible enough to overcome bad luck with good decision making, then that's good enough. But if a bad hand dooms you from the start, I feel it's going to reduce the role tactics play in deciding the victor.

Of course, I completely agree and that's my point. There is no innately 'best' move in chess. There's a best move when considered in the context of a strategy - i.e. a sequence of moves and predicted opposing moves. You don't need an element of chance there, because your opponent is unpredictable.

If the system ends up one that you can lose even when you do everything right and the opponent doesn't, then I'd consider this a bad thing. Adding in an element of chance can bring about that sort of outcome. If the element of chance is insignificant enough to be overcome then the system won't lose much and the small difference chance can make would 'spice it up' as it were. If however the element of chance isn't possible to overcome or is hugely detrimental, you'll end up with a toss-up rather than a real fight in most circumstances.

I never disagreed with that statement. I was simply trying to add to it and say that the presence of more than one target doesn't necessarily mean that there isn't going to be a 'best' target.

That's still heavily dependant on the system, of course. I'm guessing some systems don't provide you with any real way to defend yourself other than running. In most cases team fights will be more complex than 1vs1 fights, but there's no guarantee that every system is going to become more tactical the moment there is more than two players involved in a fight.

A run of bad luck is certainly possible, but unlikely. You can do a full discard/redraw as your first action of the first turn if you wish, which still leaves you enough actions to get a full defence in place by the time your opponent is able to attack. Equally, you can use bluffs and weak attacks to clear out unwanted cards, distracting your opponent long enough for you to build up a reasonable defence.

If you're only playing one-on-one, and your opponent gets a really good hand while you get a really bad hand (even after your redraw), and they know what they're doing, your chances of survival are very slim. But that's not something that'll happen often.

Equally, if the system ends up with one that you always win if you do everything right, I'd consider that a bad thing. It's too predictable, and promotes complacency. No longer do you have to worry about the unexpected, or prepare backup plans. Where is the thrill and excitment in that?

Well yes, it all comes down to how much emphasis is placed on the element of chance - there are many shades of grey between "chess" and "snakes and ladders".

A fight is likely to require many attacks, so the law of averages will probably make it very unlikely to win through luck (unless the fight is very short). What's more likely is that an otherwise very close fight may shift sufficiently in one direction to give one player the edge. In this respect, small errors of judgement can sometimes be 'forgiven' by the combat system, rather than ensuring defeat.

Well no, I don't think we can really make any guarantees about "every" combat system - but the point is that multi-way fights can be used as a way to add tactical options to a combat system.

So what happens if your redraw is as bad or worse than your initial draw? I can see chance 'setting the stage' as it were, but I think when it 'plays the parts' it's going too far. If chance is such that it can often punish sound tactical decision then that would make the system counter-intuitive.

Oh, surely. However, just the fact that you're fighting another human being means that you won't be facing the same number of choices over and over. Your opponent can be unpredictable enough to make any element of chance unnecessary. In fact, that's the way many games function, chess included.

Well, the law of averages will give you a general idea of what the most likely scenario is. However, if you get some multiple cheap wins/losses, even as an exception, the law of averages won't matter. The way I see it, the larger the element of chance, the more control you take away from the player and the less responsible the player is for the outcome of the battle.

I'm just disagreeing with the correlation multiple targets => tactical combat. Team fights usually have much more potential for tactical systems, of course - in that case team fighting acts as an amplifier to an already existing tactical element instead of just creating one by itself.

Then I'd be at a disadvantage. What I'd probably do is move some junk into my offense hand until I had enough to build up a reasonable defence, then clear the offense hand with a weak jihad (particularly if someone had attacked me, as it'd force them to shift back to rebuilding their defence) and start building up a proper offense. If I was still struggling after three or four turns I'd burn a resource I didn't need to bluff an attack, hopefully giving the impression that I was in fighting shape and encouraging my opponent/s to waste resources pumping up their defences.

Of course that wouldn't help one-on-one against a player who goes pure offense...but if you think they're going to do that, and you have a bad hand, you could call their bluff and go for a pure offense as well. You'd both die, resulting in a draw.

It's not really a case of chance punishing tactical decisions, but rather the tactical decisions having to take into account (and make the most of) the resources that chance has given you.