Arcanaville

They just added a producer, a lead developer, and a concept artist. I'm guessing we still have a few issues to go before the new game comes along and kills us in our sleep.

I thought that was pretty good; although build wizards did point out very interesting ways to improve it, I still liked it enough not to burn a respec on changing it. But I had a hunch I could do better with what's available now, and here's what I came up with:

What happened:

1. I swapped Spiritual Core for Agility Core. I lose healing, keep recharge, gain defense and recovery. As a result I lose some regeneration, but with some juggling and liberal use of boosters, I'm still pretty close: 31.45 h/s in the new build vs 32.25 h/s in the original.

2. Using agility and some other juggling, I managed to get rid of Weave. That made room for Spring attack, yay!

3. Getting rid of weave significantly improves my net recovery: 3.22eps in the new build vs 2.8 in the old build. Even factoring in the loss of one shifter proc, that's a net win. With conserve power's uptime, that would have been enough to run a Hybrid toggle. Oh well.

4. Global recharge is now higher: +107.5% vs +87.5%, and still without Hasten. Also, with boosters the build is actually now capable of running the SK-CS-SK-CAK chain in theory, albeit just barely since SK's recharge is down to 1.83s. That's probably not enough safety buffer to prevent a slight studder there, but see below.

5. I left room for Critical Strikes. If it launches as the leaked data showed, I'm going to want to six-slot that. To do that if I build this before that set releases I will unslot Hecatomb and slot a Superior Critical Strikes set there, and then shift Hecatomb to Cobra Strike, losing Hecatomb Dmg/Rech in the process. I can afford that and still end up with higher damage and recharge in Cobra. That will fill in a slight weakness in AoE: the build without Critical Strikes has only 45.5% AoE. Still softcapped, but lower than the 47%+ I have to melee and ranged. Critical Strikes will boost AoE to even higher than what melee and ranged defense currently are. It will also add more recharge, which should eliminate that studder on Storm Kick.

6. For the first time, my MA/SR will cross the 2000 health mark.


So: more recovery, almost the same regen, higher resistances, more recharge, ready for future ATIO set, but also playable now. I *think* its a pretty solid build, but I thought that about the previous one also. But aside from just plain boosting everything, what have I overlooked in trying to improve this build? Its actually fairly close to the original build: the most radical change is swapping weave and spring, although there are other tweaks. Its structurally similar. So I may have overlooked options that involve more radical changes.

Any thoughts or ideas?

Lets say every one of those attacks was slotted with a proc as strong as Hecatomb. That would generate about 24.34 dps by itself. And lets say that I have Reactive Radial and get 75% chance at the DoT. That would generate an additional 34.57 dps. And lets say that the target is continuously affected by the maximum stacked Reactive resistance debuff of -10% anyway. That would amplify all my damage by 10%. Doing all of that simultaneously, I would still need enough recharge to run that chain gapless *and* have a total damage buff of 2.70 to reach 300 dps from slotting and global damage buffs. Assuming all three attacks crit at their higher rate.

Tier 4 musculature core grants +32.25% damage after ED factoring in the ED exemption. Assault grants +10.5%.

Under this scenario, which I don't think is really even possible, you still would fall a bit short. Realistically, even if you go all out here, I think you're looking at closer to 260.

Staff single target under these extreme circumstances is probably only about 15% behind or so, with dramatically better AoE options. That's based on PS-SR-PS-SS, which seems to have about the same recharge requirements as SK-CK-SK-CAK (the critical path seems to be SK-CK-SK for MA which requires at least 3.25 recharge on SK, about the same recharge as required on SS for PS-SR-PS-SS). Also I'm pretty sure there isn't a better MA chain than SK-CK-SK-CAK, but I'm not certain what the optimal chain is for Staff yet.

Hybrid Assault could have gotten MA there, but it also would have gotten Staff there as well.

True, and no matter what the devs do to the game, I can always hack the servers and insert my own code. Somehow, however, I suspect NCSoft will not be giving me a commendation for original thinking if I were to do that.

I'm currently in the middle of updating my MA/SR build and would very much like to see that 300dps MA build.

You're never supposed to try to balance individual experience. If you're doing the same job as someone else, and I pay you both the exact same amount of money, that's fair. If I pay you half the amount for the exact same work, I can try to appeal to you that you're actually enjoying the money more than the other guy so its still fair, but I don't think you'd buy that argument. Particularly because I've just ensured you can't afford it.

Balance is not about all the different ways players can produce variable experiences. Its about everyone getting the same opportunity to do so within certain limits.

I think what you think is impossible I don't even think is desirable. What I believe to be possible is to create a framework where a wide range of player options provide a similar opportunity for success, and the rest is up to the player. The framework can also ensure that the best player in the world beats the worst one 95% of the time, not 100% of the time.

And it can ensure that not even infinite skill and intelligence can generate an insurmountable advantage. You do that by making it mathematically impossible to generate any high benefit without a compensating disadvantage. If you give the players spackle and duct tape, a smart enough player can make a spackle and duct tape cannon. But if you give the player one of these:



They can only move the hole around: they can't make it go away no matter how clever they are.


Design is not easy, made doubly so by the fact that design as a science is rarely taught. How to design X is taught, but the science of design itself isn't. Most people probably don't even think design is an intrinsic skill outside of specific context.

Hypothetically speaking, an interesting experiment would be for me to design something and then see if you could find a way for players to break it, to see whether it was possible to make mechanics that were at least extremely difficult to exploit. It would be interesting to me to see what the results of such an experiment would be.

Acknowledging the reversal of fortune in Fire Blast and Flares, I don't think I would call taking Fire Blast to be a foolish decision. In fact the Flares/FireBlast combination is one of the highest DPS options for ShootWhileMezzed under Defiance 2.0 because of their very low recharge times, unless you have an extremely high recharge build.

Notice:

Staff staffs should not be confused with staffed staff. Fighting staff staff require staff verification to prevent staff fighting staff fighting non-staff. Staff fighting staff with fighting staff staff should staff staff fighting fighting with staff fighting for staff to staff fighting.

Cc: Staff, Staff Fighting Staff, Fighting Staff Staff.

Technically, it is a set.

The more "casual-friendly" it becomes, the less effective its going to be, because the only reason its allowed to have the ridiculous strength it has is because its difficult to fully leverage. If everyone could, that would obviously show up in datamined statistics, and the devs wouldn't even let it get there: they would anticipate that change and reduce the strength of the powers to compensate.

You take the good with the bad with DA, or you lose the good with the bad. It would be different if its average performance was bad, but its not; its actually pretty good.

But you have to remember that it was partially that conflict of interest that prompted NCSoft to decide it would be better to buy Cryptic's CoH interests and form Paragon Studios, and the first thing they did was add development resources to the game.

NCSoft doesn't have that same conflict, because both City of Heroes and the Other Game are just line items on a budget spreadsheet: spending more money on the other game doesn't "divert" money from City of Heroes, because that was never City of Heroes' money to begin with. NCSoft is not taking from CoH and giving to the other game: NCSoft takes *everything* from CoH and gives back whatever it wants. I don't see why they would give back less just because a new game is spinning up: the calculus at work is return on investment, not spending quality time with all their children.

In terms of resources, the new game competes with City of Heroes about as much as Guild Wars 2 does.


Another conflict of interest perspective. David implies that the chinese wall that existed with MUO doesn't exist to the same degree within Paragon Studios now. I'm sure there are limits to what they can openly discuss, but when Cryptic was working on MUO they were working on a non-NCSoft property. There were ethical reasons to ensure the dev teams did not overlap, and did not share time or resources and didn't leak ideas from one property to the other (although I'm sure it happened a little: otherwise I wouldn't know about MUOs skill trees).

There's still the basic professional requirements of devoting all of your work time on your project, and maintaining some need to know between the two teams, but there is likely not the same conflict of interest component: both teams are working on NCSoft games. Everything they do is fungible in a way CoH/MUO was not.

And that means all other things being equal, its better we lose developers to the other side of the building than another company entirely. Paragon Studios hasn't lost their expertise or their knowledge or their perspective, and they can still make the odd contribution to CoH, as David is free to work at the pummit.

So not very good, then.

Actually, that's a half-truth. 1/5 slotting was standard, because everyone thought 1/5 slotting was standard. It was very common slotting on blasters, but on things that soloed at higher difficulty more often 2/4 slotting was very common, because that actually produced better results against higher level critters.**

Except most players do not slot a lot of IOs prior to 50 (or at least 47) so that's usually not a major factor until the level cap.

*And* no matter how much recharge a player has, they are often still compelled to get more: otherwise LotGs wouldn't be as popular as they are. If one more +7.5% global recharge buff is worth slotting LotG, the recharge buffs in Staff will also be considered valuable.


Brutes get a lot of mileage from +rech and +enddisc, and less from +dmg. They make out very well with staff. And I would imagine tier 3 soul is looking really good for Dark Armor anything right now.

Actually, I rarely use Body right now. What I'm finding is that the AoE output of the set is already plenty, and the goal is to power it. And the forms give me an excellent way to manage inspirations. When I have a ton of blues, I use Mind and my AoEs just keep coming back faster, but I burn endurance a little faster also. When I run out of blues, I switch to Soul and my endurance problems basically disappear, while I'm still pumping out AoEs at a high rate. That strategy is working so well, I don't think I'm likely to be using Body except in teams until I get to a much higher level, possibly not until I approach the level cap.


** Addendum: my Blaster's pre-ED slotting was 1acc4dmg1rech in the SO build before I built an HO based build. Because that was provably better than 1/5 even with perma-hasten

All hail Loctar, patron god of MMO forums.

Having two attacks you can use while mezzed helps twice as much as having one. That's a good reason for taking them. But it is a choice.

Specifically what powersets did this blaster have, that you decided taking both tier 1 and tier 2 attacks was actually a foolish decision? Archery perhaps?

I can see three other ways.


The problem with that perspective is that it applies equally well to PvE balance. By what justification do the devs balance archetypes by measured reward earning rates? What if players decide to judge an archetype based on the breadth of the content it can surpass instead? And suppose hypothetically people teamed 80% of the time. Because of the way XP works in teams, it would be almost impossible for anyone to lag behind, no matter how poorly they performed. And why isn't the popularity of the archetypes a major factor? Whatever the performance limit is for PvE performance to be considered within the margin for error, it isn't the factor of two or more difference in the popularity difference between the most and least popular archetypes.

The devs *pick* the metric for balance. Its not the one every player would choose. The question is does it matter?


So getting back to your conjecture that the only way to make PvP fair is if all the circumstances and abilities are exactly identical. Proof by counter-example:

Suppose there are two characters, identical in all respects except for the following:

1. Character A has twice the health of Character B
2. Character B has twice the damage modifier of A

That's fair, because the amount of time it takes A to deliver enough damage to defeat B is identical to the amount of time it takes B to deliver enough damage to defeat A. QED.

Its a trivial example, but its upon seemingly trivial relationships like this that complex balanced PvP systems can be balanced. I said there were three other ways besides making everything exactly identical. This is one: normalization. The tactic is to make the numbers look different but the critical relationships between them identical for opponents.

The other two methods which are not mutually exclusive and can be used jointly are factorization and non-transitivity. With factorization, you break down the abilities of the combatants into smaller subparts and then remix them, so that different characters have the same net strength but broken up into different powers in different ways. Exploiting non-transitivity attempts to create chains of superiority that are cyclical. For example consider three characters A, B, and C. A has high damage and low defense. B has moderate damage and high defense. C has moderate damage and high mitigation debuffing. These can be constructed so that C beats B, B beats A, and A beats C. If this is done at a complex enough level, you can end up with a hundred options, any one of which is beaten by about half the others and beats about half the others. It would then be impossible to select a "best" option.

If you mix these three options up in a complex enough way, you can make the system impenetrable to most computation analysis. Players would have to judge by experience and instinct more than calculation. And that's *very* possible. Its easy to overestimate the ability for players to quantify game systems, but six years after they were introduced there's still no generally accepted model for the value of the SR scaling passives, and even the quantitative estimation for the strength of Scourge is incomplete.


The biggest question when it comes to PvP balance isn't PvP balancing mechanics. The biggest question is what is the definition of fair? And its not proper to say its impossible to say. If you're the game designer, its your responsibility to say, no different than when you decide the definition of fair in PvE is equal opportunity to earn rewards at comparable rates. You shirk your responsibilities when you decide you can't define fairness.

How do you define fairness? You define fairness by asking what choices should be neutral, and what choices are allowed to be consequential to the outcome of a PvP match. Working from the fight backwards, most people would say every tactical decision in a fight should be consequential: what powers the player uses and in what order and in which situations (as well as the counter moves by the opponent) should greatly determine whether they win. That's how most people would define "skill."

Your statement that everyone must have the same skill in order for PvP to be fair really contradicts most people's definition of fair. By that definition there exist no fair games in the universe except for ones based solely on random chance. That's really conflating the notion of a fair game with an equal outcome. An equal outcome is where both sides win equally often. A fair game only offers them the same opportunity to win, given certain limiting conditions. In basketball you can put a taller team on the court than your opponent. That's fair. You can't put eight guys on the court. That's unfair.

As you progress backwards, should the build decisions you made before you entered the fight matter, and in all cases to the same degree? That lies at the heart of what PvP balancing means for a game with PvE. One of the first choices we make is what archetype to play. This is considered a defining choice for the character: its defining enough that its immutable: you cannot change it by respec. So the question is: should some archetypes be better at PvP than others in general? This is a dev decision: the devs can choose to make a game in which all archetypes have the same opportunity for success in PvP, or the devs can decide some are better at it than others, and warn players about that fact when they make their archetype decision. But they have to decide.

The same goes for powerset decisions: these are also immutable, so the devs have to make a decision here as well. But when you start getting to individual powers and slotting, that's entering grey areas for which the decision is different. Since you can have multiple builds, the meta question is whether people should be required to make a PvP specific build to get "fair" performance. If so, then all you need to do is make sure every powerset *can* be built equally valuable for PvP: you don't have to make sure every possible build can be equally strong in PvP. That's not a reasonable definition of fair in that case. Conversely, if you believe every possible randomly generated build should have the same chance for success in PvP, you're essentially either committing yourself to make the most complex advantage/disadvantage system in the history of gaming, or you are intending to decouple the actual character builds from PvP performance altogether (like by forcing players to PvP with gladiator builds that are pre-balanced or something). But either way this is a grey area not because the devs can't decide, but because the way the powersets and powers themselves are constructed depends on this decision and vice versa. This decision constrains how you make things work, and how you make things work constraints this decision. And this is probably the critical moment when the wheels fall off the wagon for PvP balance in most MMOs. These two things are decided independently. Its right here that what I've been saying for years - that mechanics and design have to meet in the middle for PvP balance to be possible - either occurs or fails.


In either case, however, this only shows what the challenges are in doing it right. It is not a justification to claim its impossible to do.

I see your schwartz is as big as mine.

If I ever saw 228 on MA I would assume something was bugged. And MA has less than half the AoE potential as Staff fighting has.

Staff Fighting seems fine to me, and I see nothing in the numbers that would suggest otherwise. In fact the leveling curve and performance on my Staff/SR brute has so far been very high in actual practice.

Staff seems to be the converse of Kinetic Melee. While neither are perfect sets, a lot of people hated the way Kinetic Melee looked or sounded, and claimed that on top of that its performance sucked due to totally invalid estimations for the strength of Power Siphon. With Staff, its the opposite: a lot of people like the way it looks, and that's being blamed as the reason why its low performance is being overlooked when in fact I think once again its performance is being judged too harshly in too narrow a set of metrics.

High AoE plus your choice of endurance reduction or recharge by itself suggests many of the ways its been analyzed have been extremely faulty. I think eventually it'll end up like Kinetic Melee: not extremely highly loved, not hated, but no one will be complaining about its performance either once enough people have played it and can judge for themselves.

Not really, but Willpower really requires a strong invention build to fully unlock its power, and it often takes a lot of slots to get there.

Electric Armor is a possibility, but it also has a hole: toxic. But it does have some interesting advantages: it has a heal and practically unlimited endurance, and its not very slot heavy.

Another interesting possibility is Ninjitsu: it has positional defenses *and* toxic and psionic resistances, plus a heal. Its potentially a very strong set.

Its also possible the people moving have some say in the matter: given a choice I'm sure most MMO developers would give one of most things they have two or more of to be one of the people who gets to shape a game in its formative stages. So this could be a form of upward advancement opportunity for Paragon employees.

Supporting an MMO is something many gaming professionals do, but launching one is something a lot less ever get the opportunity to do. Its not an indictment on the future of City of Heroes if some of our developers and artists decide to work on the new shiny. If they offered me an opportunity to work with the design team of the new thing, I love City of Heroes and all but unless it was a pokemon facebook game or something I would sign up for that fast enough to incinerate the pen I was using.

PvP balance isn't supposed to neutralize player skill. But it is supposed to level the playing field when it comes to tools. The problem is that when it comes down to actual power levels, City of Heroes is so wildly unbalanced** that of course PvP is going to be as well. The only reason why those wild imbalances don't translate directly into wild measurable performance imbalances in PvE is because the game's design doesn't allow the reward system to always proportionately reward performance.

When it does, like oh, say, an AE PL farm, it becomes a lot more obvious.

But as to "perfect" balance being impossible, perfect anything is impossible. But "balanced enough to not matter anymore" is very possible. It just also happens to require a particular approach to the problem I think game designers are unable or unwilling to execute.

The enormous advantage MMO game designers have is that they not only make all the toys, they also make all the rules under which those toys must function. But I have *never*, *ever* seen an MMO in which the mechanics and offensive and defensive systems have been constructed in an integrated manner. If you do that, you can make a balanced PvP system because you can have mechanics and powers meet in the middle. If you try to bolt them together after the fact, you've just made the problem a hundred times harder, but still not impossible. Just highly improbable.

I suspect, however, the only way to prove this is proof by counter-example, and that's unlikely to happen anytime soon.


I'll make an even stronger assertion. I assert that PvP balance is *easier* than true PvE balance. The only reason this appears to not be the case is that its even harder to start with a system designed for PvE and then extend it to balanced PvP, and because most PvE systems aren't very well balanced either: they are not so much balanced as the reward systems are designed to dilute imbalances. But you can't do that in PvP, where the primary reward is "the enemy dies."



** There is also the argument that its wildly unbalanced PvE is part of the game's overall attractiveness, but that severely hampers PvP when the thing on the other side of the screen is another human being, and not a critter.

Was this just the cone radian conversion bug, or did it also fix any other PPM bug? Because I don't think the radian conversion bug alone was enough to fix cone PPM mechanics.

Which doesn't answer my question at all, which is what are you doing while you're standing around waiting for Zapp to recharge, if you are using it from the range you are harping as a benefit. The point is that the *extra* range of snipes has limited functionality, because no matter how much range you build for, the *extra* range of snipes is practically worthless. How much different would your character play if Zapp had only 80 feet of base range like all the other attacks you claim are your "main blasts."

Sniper blasts *pay* for that extra range, but blasters *get* almost nothing from it. And your example does nothing to prove otherwise: if anything it only goes to show that if you want to hit something at 150 away, you don't need a sniper attack to do it.


If they weren't broken, you'd have said you have four main blasts, not three.

The devs have said in the past that unresistable damage in PvE is not really on the table because it breaks a lot of things. The problem is that it has a small benefit against normal critters, but against critters specifically designed to be very highly resistant the benefit increases dramatically. There are special ways around that, but they themselves have complications.

To give you a hint of the problem, suppose you made 10% of blaster damage unresistable. Against a critter with no resistances that would do nothing. Against a critter with 20% resistances you'd do more damage: 90% of your damage would be resisted by 20%, and you'd end up doing 72% of your damage, and then on top of that 10% would be unresisted, giving you a total of 82% damage vs 80% if your damage was all resistable. A tiny increase.

But against a critter with 90% resistances, you'd now be dealing 90% of your damage against that resistance, dealing a total of 9% damage, and another 10% unresisted for a total of 19% of your original damage. You'd now be doing more than twice the damage you would have been doing if all your damage was resistable. That's a huge buff against a target that was *intended* to be hard to bring down. And when you start getting to things like Marauder in the Lambda itrial, things get even messier.

Another quirk to unresistable damage: if your damage ignores resistances, that also means that part of your damage will not *benefit* from resistance debuffs either. There's no real way to make damage unresistable, and yet also benefit from resistance debuffs. So in some situations, unresistable damage can actually sometimes be a penalty on blaster damage. Sonic blasters would be particularly problematic.