Arcanaville

Hypothetically speaking, tankers would still be limited by having most of their offense be melee instead of ranged, and without taunt there's no mechanism to automatically get critters to foolishly surround you.

You could also manipulate the numbers in conceptually palatable ways to account for teaming situations. For example, blasters could have the offensive advantage that their damage doesn't drop as much when they have "attention" than tankers do, to account for their offensive focus (they are just better at scoring good hits even if you focus on them and try not to get hit). That would allow you to still give blasters enough of an advantage that their offense tended to be superior overall in most circumstances.

I've always thought that the way this should be managed is to assume that you can only focus your attention on a limited number of things, and things you aren't paying attention to are by definition a more serious threat (because you can't see what they are doing).

In such a model, the heavily armored thing standing next to you is only doing 74 because you are actively trying to make sure he doesn't squish you, while the guy in the robe far away is doing 134 because you're willing to ignore him for the time being. If you elect to ignore the heavily armored guy and try to shoot at the guy farther away, the heavily armored guy would then squish you for 200.

In CoH terms, tankers would only do less damage than blasters *because* foes paid them proper respect and treated them as the biggest threat, not because they *intrinsicly* did lower damage. They would intrinsicly do similar damage, but that intrinsic damage would never actually land unless they were ignored.

The model in my head for aggro and damage balancing has always been the WW1/WWII statistic that most bullets fired by soldiers were intended to prevent the enemy from shooting back effectively, not to actually kill the target (except by chance). In other words, you shoot at something not just to kill it, but to prevent it from using its offense against you. I think that model would work especially well in CoH. The reason why you attack the tanker is not because of taunt, but in effect to debuff their otherwise substantial damage. They are the biggest threat *logically* and not just because of a game mechanic. And this means (accepting that this is not the only issue in PvP) that even in PvP there's a good reason to attack the tanker rather than the defender on a team. By making it the *intelligent* decision to attack the tanker in PvE, you give them a reason to exist in PvP as well.**


I personally think the combination of reverse-bodyguard and an aggro system like this is so potentially interesting, it might even be worth experimenting with it in a game already well-established like CoH. If it were up to me, I'd introduce it as an optional mechanism: something tankers could opt into, or alternatively a new archetype could be created that used the mechanic, to see if players find it superior to the traditional aggro system. If players actually liked it, I'd kick the current aggro system to the curb.


** Although I'm mentioning tankers, the general principle can work for all archetypes: simply ensure that whenever you want a critter to attack something over something else, you give it a good reason to do so mechanically, so its always the intelligent choice, not just an override. Eventually you build a situational table that says critters should attack the tanker under these situations, the defender under those, the blaster under these others, always because its the best situational choice for that critter.

Perhaps you're thinking of a specific finger?

I don't know why this keeps getting alleged though. Power customization required *zero* change to the graphics engine. You're not seeing anything today that the graphics engine couldn't do long ago. The changes were server-side changes to support the extra customization data and the tools necessary for the devs to support power customization.

What I'd like to see, besides the typical requests for higher rendering rates with higher tessellation/polygon counts is better support for physics-influenced behavior: moving hair, flowing capes/cloaks/skirts, better destructible environment support. Basically better interactive immersion with the overall environment.

I should point out the context of the quote was a question from someone who almost certainly had not played the game through yet. Once someone has played the game through at least once, they certainly wouldn't need my or anyone else's advice on whether they should play through again.

But even if my preference was the same as yours now, I would never recommend someone to skip through to 50 under any circumstances if they had not played the game through once yet. I think almost by definition if someone is asking if they should play through, the correct answer is "yes." The correct answer is "no" only for people who aren't likely to ask the question in the first place.

Super Strength / Willpower will be the hardest thing to bring down overall, SS/Rad(Emission) would probably be the most offensively dangerous thing to someone relying on soft-capped defenses.

However, there are lots of wild cards that would be tricky for soft-capped characters. Earth(control)/Plant(control) would be an example. As a pet, carrion creepers would have base 75% chance to hit, not 50% (in effect, +25% tohit) and I believe quicksand autohits. The combination can be nasty if the player is unaware of these facts and simply charges in. My recollection from I14 beta was that these were too nasty to put in significant numbers in my original scrapper challenge mission (which had both SS/Will Bosses and Fire/Fire Bosses stacking autohitting auras and -perception, if that puts things into perspective).

At the time I posted that, I wasn't actually thinking about whether there actually exist any regen/recovery Strength debuffs: I was thinking more about the general principle that things tend to obey strength modifiers by default, unless there is a strong overriding reason not to, because in general not obeying Strength modifiers means you simultaneously lose both effects: the ability to buff with strength modifiers and the ability to debuff with strength modifiers.

Now that you mention it, though, the only regeneration *or* recovery Strength debuff I can think of are the recovery strength debuffs that blue mitos possess (at least, I think they still possess those). That might be enough of a precedent, however, to ensure that +Regen and +Rec buffs obey Strength modifiers by default, just to prevent downstream problems with new content.

Pthhth thhhh thhhh.

That's why my suggestion doesn't quite work right with CoX's current aggro environment. Its possible to make it work using the taunt-effect override mechanism, but since that mechanism has significant hysteresis (you need a very large taunt to override a previous taunt) it doesn't have the fine control necessary to make this work very well. Otherwise, the multiplier for taunt is too high to allow for any real leveraging of the hate mechanism itself in this context.

The display could also be wrong: Real Numbers displays don't always fully account for target resistances: some player effects show a similar discrepancy between the net resisted effect and the displayed strength. At least, this was true the last time I checked which was a while ago, but I know it has been an issue in the past.

The question is, under the test conditions, if the damage numbers in combat chat (which are, as far as I know, authoritative) agree with the resistance debuff being resisted, or with the resistance debuff being unresisted.

Unfortunately, I'm not in a position to directly test that at the moment: I'm still waiting on my new PC (and I had to blow the client off my laptop temporarily for a project).

It would be better if some things required higher levels of taunt/hate to draw their attention rather than simply ignoring taunt/hate altogether. In effect, they could be resistant to taunt, and have intrinsic preferences that are mechanically like taunting themselves to attack something. So something with an "attack the healer" preference wouldn't simply ignore the tank and kamikaze the defender, it would act as if healers were taunting them with mag X, and the tanker would notice they would need to generate at least mag X+1 to get their attention (this is an oversimplification: the way taunt works doesn't quite align with this idea as I understand taunt and hate).

For me, that's somewhat of a denormalized perspective. Which is to say, within the context of what we're describing, if I were to replace all the CoX shared zone spawns with ones that had twice as many critters with half the power each, that would not change spawn density as I'm using the term here. What matters in context is the area that a player/team will "sweep out" when they engage the average threat limit. In that sense, making the critters weaker (or making all of the player stronger, which is the same thing) is in effect reducing spawn density when it comes to the issue of aggro distribution.

But as to the rest, I think we're in general agreement regardless.

The problem is not balancing: that's a relatively straight-forward problem, if a large one in terms of total work. The more fundamental problem is whether most players actually want to track that level of detail. Chess would not be improved by making the Knights better simulations of horses.

Because everyone is different, there is no "right" answer to the question of how much detail is the right amount of detail. The only thing you can do is pick a level of detail and design a game that is attractive to the people who *would* appreciate that level of detail. CoX's complexity target is significantly lower than, say, Champions Online is. That isn't a good or bad thing: its just a thing.

(And actually, my personal complexity threshold is far higher than CoX - or even CO - so I would enjoy more complex game play and game mechanics. But even so, I think CoX currently does a better job of being consistent with its target than CO does. CoX targets lower complexity gameplay and tends to do relatively consistently across its design. CO tries to be more complex in many areas, but doesn't do so in a way that always makes complete sense. In particular, the advantage system seems to be just short of vestigial, relative to what it was originally intended to be, and the stats system is just short of being opaque, which is nonsensical as a character building option).

No, its really the simple issue that if you let critters notice you with a more realistic line of sight algorithm, they will notice you at longer ranges. This means they have to be farther apart, or radically weaker (or else an individual player or team will aggro too many critters).

If you make them farther apart, there's less targets overall, and that means the zone will support less players fighting anything: the combat density has to decrease. If you make them weaker, you get the same result because while there are the same number of critters, each player will now "sweep out" a larger number of them when they engage in combat.

This matters if you care about shared-zone combat, and you are trying to balance your shared zone size and structure relative to the number of players you want to support within it. The relationship between aggro radius and player maximum density is not a simple proportion, but the former does significantly affect the latter when it comes to designing the shared zones.

That depends on your point of view. City of Heroes has become increasingly more focused on the instanced parts of the game, or rather its more precise to say that the playerbase has become increasingly focused on the instanced parts of the game. But there are still shared zones with combat in them, and you still have to deal with the density problem there.

Furthermore, I wouldn't characterize City of Heroes as a "low density" MMO, since CoX shared zones are density-balanced around low aggro radii. The fact that there aren't many street-sweepers these days is not relevant to how the zones are designed. Saying CoX *could* be reformulated around a higher aggro-radius lower-density spawning system is only a conjecture that would require some evidence to support the notion.

Personally, I think that its a short bus ride from lowering the spawn density in the shared zones to support high-radius aggro to simply tossing the zones and replacing them with an instance list.


In any case, this is going around in circles just a bit. I said that the aggro problem the author describes is really not a problem related to aggro: its a problem related to the issue of density control. You're saying you don't care about the density control problem, which is fine: that just means for you, the aggro problem is resolvable by nullifying its underlying foundation. It doesn't contradict my assertion that the aggro problem is really just a symptom of the density problem.

But do the crates initiate combat? And if so: with what?

Instancing doesn't solve the problem, unless you have an MMO where there are no shared combat areas at all - which is in effect not fundamentally different than having a giant world where everyone is out of sight of each other (while in combat). Instancing is just the part of MMO gaming that deliberately separates players by design. But that doesn't solve the problem elsewhere.

This isn't a problem that can be solved by cleverness, because its simple arithmetic: if a single player (or group) can aggro a larger radius, then it will take less people to sweep out a given area. That reduces the maximum (combat-active) player density supportable in a given shared combat area as a logical consequence.

Most player self-buffs are tagged to be unresistable, which is why players do not tend to resist their own buffs. But I believe Achilles is tagged to be resistable, so even though its a "self buff" it can be resisted by the target. And yes if this is set incorrectly on a player self buff power by mistake a player can resist their own self buffs.

(An example of player buffs that are deliberately resistable are most heals. That is how Hamidon can make them ineffective: Hamidon is actually *buffing* player heal resistance.)

He combines two separate complaints into one: "aggro" in the sense of attention-splitting by foes:

And the issue of "drawing aggro:"

I considered the phrase "in plain sight" to mean he wasn't referencing line of sight. You can't be "hidden in plain sight" out of line of sight.

And this latter issue is not an issue of aggro: the devs could set the aggro radius to anything they want in theory, and they could add line of sight. But both are limited by the density problem: if you allow critters to notice you at higher, more realistic ranges, you have to lower the density of critters. This lowers the density of players, and you have an MMO where most players are out of sight of each other, which is undesirable.

Well, its not necessary to know how to solve a problem to point out that a problem exists. But I think the author of the article suffers from a more fundamental problem: he's willing to state there's a problem without even any notion of whether there even *exists* an option.

A separate problem is that he didn't even try; specifically with regards to #4 (Aggro). Lots of people have come up with ways to replace the aggro mechanism in teaming. I did myself when I came up with my "reverse bodyguard" mechanism for tankers - which actually predates the (mastermind) bodyguard mechanism by a significant amount of time and worked somewhat differently, but I renamed to "reverse-bodyguard" after the mastermind mechanism was introduced. Basically, rather than creating an aggro mechanism where the tanker has to take aggro, allow players to block, deflect, and absorb damage directed at other players, to protect them. The critters can still attack whichever target they want: its up to the protectors to protect those targets with the mechanisms they have to do that.

I said I could replace the aggro mechanism for teams. I don't what the heck the author means when he suggests that the actual act of aggroing a critter at all should be replaced. He suggests that non-MMOs don't have this problem: oh really? There's a game that has a mechanism for NPCs initiating combat with the player that don't involve the NPC noticing the player? Heck: I'd like to see that game. I think its here that the author demonstrates their true colors in trying to be more clever than thought-provoking.

I don't recall off the top of my head, but I think the Recharge debuff one works the same way, so its not just Achilles.

I think the recharge debuff one can stack unlike Achilles, but I'm not 100% certain about that. But I think the logic behind making the debuffs a grant power in general is that normal effects can either stack or not stack. One or many. But powers can be limited to a specific limit. For example, the "5-rule" for set bonuses is enforced by placing a limit on the passive powers that implement those set bonuses: the game only allows you up to five copies of the power. Proc debuffs that are implemented as grant powers can theoretically have the same kind of limit: you could make a debuff that could not stack (limit: 1), or stack up to three times, or five times, or any limit you want.

It sounds like he has no peripheral vision and can't add.


Whether you keep track of numbers and statistics during actual gameplay is, for the most part, purely a matter of personal preference. If you hate doing it, but do it anyway, you have no one to blame but yourself.

Let me be direct: prior to about I4, essentially all numerical min/maxers in CoH were doing it wrong. I can say that with almost absolute certainty because a) the public state of the art sucked prior to that, b) no one outside the dev team knew enough of the relevant numbers and mechanics, and c) even the people inside the dev team consistently got it wrong in public posts. Consider that most of the damage mitigation calculations prior to I3 didn't even factor in regeneration, the tohit algorithm wasn't well known prior to I4, and all attack chain calculations have been wrong prior to the discovery of the server-clock alias issue and the actual source of combat rooting. And that's just scratching the surface.

Its one thing to say that people who ignore the numbers can succeed: I'm saying people who pay attention to the wrong ones can still succeed. I think MMO developers don't do enough work to make their mechanics transparent, but that's not the same thing as saying people have to be statisticians to play MMOs.

I'm not saying there's no advantage to knowing the numbers, I'm just saying the advantage in conventional PvE play is usually small except in corner cases, relative to experience. It doesn't define whether you can succeed in an MMO, at least not this one or any other I've played.

Achilles Heel "ignores" the purple patch because the Achilles Heel -Res debuff is *not* an effect that the player imposes on the target. The way the debuff works if you trigger the debuff your attack actually Grants a passive power to the target that applies the debuff.

Since the target is affecting itself with a power you gave it, the debuff always affects the target for full strength: never more, and never less. It is always an even-con effect by definition.


Edit: that's probably also why it uses the Melee_Ones table: if it didn't every target you hit with it would have a different strength effect because the effect would use the target's table, not yours

I'm not talking about another tohit check. I'm saying that the game engine already supports separating evasion and deflection/absorption as two separate mechanics already. And in fact, we already have minor hints of separation: a non-positional psi attack is an attack that can be psionicly deflected, but it cannot be "evaded." Conversely, entangling arrow doesn't care about your ability to deflect damage but can be evaded.

The tohit and debuff side of the equation is a bit more tricky, but could be retrofitted to the existing engine with very minimal changes to the current system.

The main problem with holds is the problem you suggest: its not that "holds" are treated poorly in CoX, its that "holds" cover too much conceptual ground to be specific. I personally wouldn't have made a game mechanic called "hold" if it were me: I would have broken up "hold" into different smaller mechanics and used them to construct the specific mez powers.

We don't really need a new system per se: my guess is that positional defense and damage-type-oriented defense were conceptually intended to separate evasion and deflection/absorption. However, they just aren't used that way in the game.