As soon as there is any limit whatsoever, then any system can be trivially reduced to a "class based MUD" if you allow the possibility of there being so many classes you can't give them a name. But this is a pretty silly and certainly useless definition of "class".

As DH said, this is inevitable. If you think about classes in this manner, then there is no such thing as "classless," and what I would be trying to make is a class-based MUD with about 3-4 million classes, depending on how many skills I end up with.


I haven't made any mention of being able to "change classes" (or skills), yet. I've toyed with the idea, and it may very well be my next forum topic :wink:, but no matter if I do allow this or not, the game will still be "classless" by my definition. If it makes things easier for you to understand, just look at the kind of system I'm proposing as a class-based MUD where you have hundreds of building blocks with which to create your own class, rather than being restricted to choosing from a small handful of system-defined classes.

I'm unclear as to what you mean by "intra-player balance".

I'm not sure that favouring specialisation is necessarily a bad thing, so long as there are sufficiently large number of viable specialisations. If you don't favour specialisation then you are more likely to end up back at the "all characters absolutely identical" end of the classless scale. However hard you try, some skills will inevitably be more useful than others, even if only slightly. If you have a completely free choice of skills then people will always pick the same combination of useful ones. Favouring specialisation allows you to rig the system in favour of diversity. If, for example, using fireballs is always (on average) slightly less effective than using a longsword, regardless of which other skills you know, then, given a completely free choice, people will always choose long blades and never choose fireballs; but if the efficacy of fireballs is substantially increased as a result of studying other forms of fire magic, while the efficacy of long blades is increased by studying other combat techniques, then some people will become fire mages and others will become warriors.

You don't have to go all the way to a fixed "only-certain-classes-can-acquire-certain-skills" system; you can fine-tune the magnitude of the inter-skill dependency factor to get the balance you want.

I imagine that intra-player balance means that you have several interesting options available to you, whereas inter-player balance means that it is possible for players to meaningfully compete with each other. I don't view the distinction as being hugely important, though, because presumably you might find options interesting in principle but uninteresting in practice because they hamper competition. So I'm also curious what exactly was intended.


I disagree with this. If you build things around trade-offs, such that it's reasonable that you might find yourself in situations A or B that each favor opposite ends of a trade-offs, then it's hard to say that one skill is clearly more useful than the other. I'm not sure these are "specializations", unless you consider specialization across every possible dimension of skills – as opposed to, say, fire mages and ice mages which are a basically one-dimensional specialization, or offense vs. defense mages, another one dimension. But if you have two dimensions, fire/ice and offense/defense, you have four possible "specializations" – but at this point I don't think it really makes sense to speak of specializations anymore, especially as you increase the number of skills and trade-offs. You end up with a trivial specialization, much like was said earlier about classes themselves: you end up with a huge number of possible "specializations", with indeed each build being a specialization unto itself!

Note that this form of "specialization" is quite different from the dependency among skills that you introduced with fire mages vs. blade-wielding warriors.

I do note that you said "some" skills, although I don't view it as a huge problem if only some skills have this property but there are enough skills that don't.

Not more powerful overall, more powerful in a given situation. The fact of reality is that someone who squanders their time and energy to try and learn some of everything will NEVER be as good as someone who focuses on just one thing. A well designed game should take that into account and provide puzzles which are easier to solve for specialists in one field, but difficult for specialists in another field.

I also disagree that it's desirable to design your game system around solo content. If you are building a multi-player game, it should be designed around group play. If you wish to provide some solo content for times when there aren't many players around, that's fine… but designing the game around it is pointless. Design a good solo game instead.

A great example is World of Warcraft. When it launched, the game was easily soloable up to a certain point, and then it became extremely difficult and tedius to progress without a group. Some stubborn people still did, but while they slogged around killing a zillion boars in the forest, those who chose to group not only progressed faster, but got to see content that simply wasn't accessible without a group. Over the years, the devs have tried to balance the game more and more towards the solo player, and the result is that finding a group is almost impossible until you hit the end-game content, where it's required.

To me, that isn't progress. That's dividing your content into a nice solo game that you play up until a certain point, and then a multi-player game you play later. You may as well have provided a solo game client to play your character offline and then just upload him when you hit the level cap, the experience is the same.

Personally I don't see a problem with allowing someone to play a jack-of-all-trades, but even that could be considered a type of 'specialist' from a character design perspective - a character specifically designed to be well-rounded and able to hold their own in a variety of different situations. Perhaps they can't beat those animated snowmen as easily as a pyromancer, but they should be able to beat a wider range of less powerful opponents.


There is a big difference between a solo game, and a game that allows you to play solo, and I see no problem with a multiplayer game that focuses on competition over coorperation. I would also point out that while some players enjoying grouping, there are also players who hate relying on others just to be able to play.


Thanks for the warning, I definitely don't want my mud to fail as badly as WoW :P

Yes, that's what I meant. A specialist will be more powerful than a jack-of-all-trades as long as their specialty applies. If you neutralize the fire mage's ability to use fire, you make them incredibly weak. The jack-of-all-trades can't be neutralized as easily, but doesn't have the raw power the specialist has if they can use their abilities fully.




*grin* Well, just realize that WoW had made those design choices specifically to gain a larger audience of paying customers, not to make their game a better game. As you say, many people do hate grouping and prefer to solo, and changes like that draw in that large segment of the target audience.

It all comes down to why you're making your game. If you make it to be an incredible game with complex interaction and puzzles that require teamwork to solve, you are targeting a different set of people from those who want to be able to log in, bash stuff, and log out. It is indeed a smaller population.

I don't think anyone makes a game with the intent to have no players, but it's also possible to go too far in trying to appease the masses with over-balancing and trying to ensure everyone and their dog can play however they like. :)

It's been my observation that strong support for solo play makes it easier to build up an initial playerbase, while strong support for group play helps keep the playerbase (as the players tend to form friendships that keep them coming back even after they've lost interest in the actual game).

I would therefore argue that if your primary goal is a large playerbase, you would be better off supporting both styles of play - solo and grouping.

There's an interesting contradiction in the statement that you'll get a lot of paying customers by making what is not a "better game". I suppose there can be an art vs. mass consumption debate, but, well… you know.

Well, it's a given that a skill will be applicable only in certain situations. A lock-pick skill is only useful if there is a lock to pick. The usefulness of a lock-picking skill is therefore directly dependent on how many pickable locks there are in the game, how many of these can only be circumvented by picking, and the value of the rewards obtainable by a successful pick.

Similarly, the usefulness of "social combat" skills is directly dependent on the positive results that can be obtained by using them, e.g. the number of susceptible mobs.

Even if somehow you could exactly balance the potential rewards that a "bribery" social combat skill makes available with those obtainable by picking locks, all you have to do is add one bribable NPC to the game, or add one new locked door, and they are no longer in perfect balance.

@David - It's not a contradiction, but rather an indication of what kind of players are out there. If you want a game where people spend much of their time soloing, then you build a game designed to make soloing easy. Likewise, if you want a larger player base, attracting the solo players (who typically only play a few hours at a time, and often not every day) will give you a larger pool to draw from. This doesn't yield a better game, only a more popular and accessible one.



and that is why I say spending all this time on hypothetical "balance" is foolish. Why not spend your time making the game fun and interesting, rather than worrying about what happens when potential super specialist fire mage from Corinth duels super-specialist smoke mage from London, and they decide to fight on the frozen wastes north of Reykjavk, under a full moon when LARP specialst magi gain an extra +1?

So you've made an area which has a door that can't be broken down by brute force, and can't be opened by magic, thus giving the poor thief a reason to be needed afterall. Why is that a bad thing? Waaaaaahhhhh, I only solo and I play a fightzor, and I can't get into that zone! Tough.

At some point, there will be some encounter or trap or something that the thief can't handle without a fighter. I'm sure, if your game has enough interesting content, a mage or cleric will be essential to solve other puzzles. In short, why *should* everyone be equivalent? Why *should* I think a fight between a fighter and a mage would be an even match?

I'm not sure anyone is suggesting that's a bad thing.


If the MUD contains significant PvP elements, then the combat potential of different classes really do need to be similar.

In a PvE MUD, encouraging diversity is a good thing in theory; but, as I've said in previous posts, it depends on the rest of the game world being designed in such a way that there are lots of situations in which non-combat skills are useful. Many MUDs are based almost entirely on killing; you gain XP by killing, gain equipment by killing and looting, gain access to areas by killing guard mobs, etc. So long as that remains the case, combat-related skills will always be of more use than anything else.

Poorly balanced character options, particularly in a very competitive mud, can make the game considerably less fun and interesting.

On the other hand I don't think that anybody is actually suggesting that everything be so utterly perfectly balanced that if you take the sum total of all mathematical equations across the entire world and set of future worlds the builders might or might not put in, the sum is a perfect zero. I think that there are fairly reasonable approximations that one should strive for.

And incidentally I was not thinking about binary applicability of skills like picking locks, which either works or it doesn't. It is more interesting, IMO, to discuss skills that work, but work less effectively in some circumstances. Combat-based skills (magic or weapons) lend themselves to this fairly well. It might be appropriate in some fights to be fast and weak, and in others to be slow and strong, but being of average speed and strength will prevent you from ever being at peak efficacy. Somewhat more extreme examples are things like creatures being vulnerable to flame and resistant to cold (or vice-versa) and having magic of one type or the other, but this gets a little closer to binary applicability.

That's how lockpicking works in my game, although its based on player skill rather than character skill. The experienced players pick locks very fast, the less experienced players require more time, and the newbies usually end up triggering traps on the locks and dying as a result.

Likewise with combat, the experienced players can usually defeat their opponents fairly fast, the less experienced players require more time, and the newbies end up dying.

They're both boolean in so much as you either open the lock or you don't, and you either kill your opponent or you don't. But there are definitely different degrees of success. As such, I don't disagree with shasarak's use of lockpicking in his example.

I wasn't saying that lockpicking was some kind of inherently bad example: I was saying that the true space of examples to consider is far wider. However, I disagree with you that your examples are binary: they are only "binary" in a very uninteresting way (either you win or you don't). If that is how you look at the world, then there is really nothing other than binary – you succeed or you fail at your task – especially if you allow the weaseling out by saying that you can do each to varying degrees. It's like saying the only options are 0 or 1, but you can be a better 1 or a worse 0 than somebody else. Or, that you can more effectively get to 1 than somebody else. In that case, can you truly say the space is binary?

That said, I don't really think you spoke to the point I was making. You only discussed things based on player skill, not other circumstances the world (or even other players with their tactics) might present to you.

You referred to it as a "binary applicability", "which either works or it doesn't". My point was that lockpicking isn't inherently any more or less binary than combat, unless you're talking about specific implementations (such as Diku).


Yes, that's the point I was making.


Because it was a real example. I could easily envision a lockpicking system that utilised a range of character skills, and the underlying point (lockpicking being more than a simple "binary applicability … which either works or it doesn't") would be the same. But I don't know of any real examples of such a system.

All right, so we agree that the balance can't ever be exact.

What I was suggesting was that not only can the balance not be mathematically perfect but, in practical terms, it would rarely be close enough to foil MinMaxers, because MinMaxers thrive on figuring out and exploiting even the tiniest of margins. If this is true, and if every skill is available to any character, then all MinMaxers will inevitably end up with identical characters; and there are enough MinMaxers around that this will lead to a high degree of character-homogeneity. Hence, some sort of "soft class" system - i.e. interaction between the efficacy of skills, such that one becomes more effective or easier to acquire as a result of the others you possess - may be necessary in order to prevent this.

I'm not sure why this is relevant. The point had nothing to do with lock-picking per se; this all seems like a red herring. The interesting part is what came after the bit about lock-picking. I don't care about lock-picking itself being binary or not. The point I was trying to make was the following: "It is more interesting, IMO, to discuss skills that work, but work less effectively in some circumstances."
Ironically enough it seems that this is also what you're interested in discussing, although I think you're overly focusing on lock-picking.

I still don't think you addressed my point, though, about applicability based on the world rather than player skill. You said your example was "real", ok, fine, it was, but I still don't see how that makes it relevant to world circumstances. :smile:


I think you're somewhat overestimating the ability of minmaxers to necessarily find the optimal solution, rather than merely think that they found the solution.

I was looking around online for strategy guides to a mini-game of chance in Fable 2, which I've been poking around with recently. I very rapidly found that the vast majority of people talking about the game had absolutely no idea what they were doing, despite there being many claims of finding the "right" strategy, sometimes even with consensus forming. I sat down and did the math myself (having a relatively significant background in probability) and it was easy to show that their "good" strategies were in fact rather poor.

I will of course grant that the larger the audience, the higher the probability that you'll get people who actually know what they're doing with their analyses.

You might say that this doesn't change the fact that if people think they've found the right answer, they'll all look the same. Well, really, this is true no matter what happens. Inevitably, a system will have some options better than others. If everybody thinks that a given option is the best, even if it really isn't, you'll get this homogeneity. So, the problem of consensus leading to homogeneity is really a different one from some abstract notion of "balance". In fact, it almost suggests that you don't necessarily want an abstract notion of balance, if it's too hard for players to actually tell that the game is balanced.

As David pointed out, there's quite a difference between "popular" and "imbalanced." While homogeneity is never overtly desirable, it has to be tolerated to an extent, because it's human nature to do what works. There just aren't that many trail-blazers out there. I'm not necessarily trying to control what's popular, though (as long as that popularity has nothing to do with imbalance). As long as I can take any two sets of skills in my game and honestly say that they both have the same relative usefulness/power, then I'm satisfied. If one happens to become far more popular than the other for non-balance-related reasons, then I'm not going to sweat it.