Arcanaville

Because game diversity is a topic I'm personally interested in, because the conversation has been heated but relatively polite, and because I'm still thinking of the best way to actually calculate the number of possible builds that exist in City of Heroes in a reasonable amount of time and effort. I don't want to start a whole new thread if I actually come up with a number. The calculation is mostly irrelevant to the discussion, but its an itch I now have to scratch.

Incidentally, wouldn't this be equally true if I respeced into fighting *before* inherent stamina? You haven't stated why inherent stamina would put more people into fighting than before. And "because there's no other good choices" is not a valid reason on its own, because lots of players seem to think there are other alternatives.

This leads more directly to what I said above. Fitness can preclude people from taking other powers for two reasons: they ran out of power choices, or they ran out of power pools.

If you ran out of power pool choices itself, then that means a potential option you might want - say fighting, fitness, medicine, flight, leaping - simply didn't exist before inherent fitness. That combination is literally impossible. Inherent fitness makes it possible. Its an option that was *created* by inherent fitness.

The fact that such options exist mean the logical argument that reducing the number of power choices by four must mathematically reduce the total number of build options is not automatically true. QED.

No blaster before inherent fitness had fitness, flight, speed, leaping, fighting, and force mastery. No blaster had then what I have now, because inherent fitness created that option where it did not exist before. And very specifically, no blaster prior to inherent fitness had hasten, combat jumping, hover, tough, weave, temp invuln, and stamina to power it all.

If you only want two pools, and one is fitness, inherent fitness doesn't add many valuable options. But if you want five pools, inherent fitness created options. And again, if most people had fitness, then the amount of diversity giving it away dropped by only a small amount. It would only take a relatively small fraction of players running out of power pools to mitigate that loss.

Actually, I can. Lets consider the alternative. When the notion of making fitness inherent in the first place came up, I suggested that was not a good idea: it eliminated the option of taking it or not taking it. The better alternative, I said, was to simply increase base recovery to reduce the impact of low endurance. Endurance problems would be partially aleviated, but players could *still* take fitness if they wanted to.

Now, suppose the devs actually did that, much as they did for VEATs. Instead of making fitness inherent, they simply boosted base run, jump, regeneration, and recovery. In effect, that would be no different than adding powers that increased those base stats, if such a thing were possible (and it sort of is). Whether we have a power that increases a base stat or it happens invisibly makes no difference: its a distinction without difference.

So if the devs add four passive powers that increase base stats, or increase the base stats directly, everyone is changed in the same way. Builds are impacted identically. Diversity is the same in both cases, because these hypothetical passive powers don't affect the number of choices players actually have: they still have the same choices they had before, we just added something to *everyone* that does not in any way affect the power choices they have.

Similarly, the existence of inherent fitness by itself doesn't reduce diversity. In fact, because those powers can be slotted, it actually has more diversity options than my previous example.

The difference relative to what actually happened is that the actual choice to choose or not choose fitness itself disappeared with inherent fitness. It isn't the fact that everyone *has* inherent fitness that changes anything, any more than adding a passive power to everyone all by itself changes anything. Its a power, but not a power *choice*. We still have the same 24 power choices we had before.

It *is* true that we have less actual powers to choose from: in effect, four less than before. But that is four less than the approximately 82 powers we had to choose from: 18 primary and secondary powers (given a specific powerset combination) plus 6 normal power pools plus 4 travel pools (with five instead of four powers) plus at least 4 epic choices. In practical terms, with prereqs and level unlocking and the pool maximum all 82 options were not equally free choices, but the number of choices is still very high. Thousands of combinations, in fact, even following all the rules (actually, I had the bright idea to write a recursive program to count them all: it might finish running sometimes in January, so I need to optimize it a bit).

One of the big logical leaps you are consistently making is arbitrarily declaring almost all of the options that exist as being irrelevant or immaterial. Even in the face of players explaining why those options actually make sense, such as having multiple travel pools. *If* you restrict yourself to the very severe restrictions you seem to place on what constitutes a valid build, then current diversity could be very low. But that is not the fault of the game: the vast majority of the options it is presenting to players and lots of players are taking advantage of are options you simply aren't counting as valid.

Another logical leap is in assuming that most people do not take very many power pools in the first place, so the power pool cap of 4 is not important. But lots of players run into that cap, including myself, and as a result granting me inherent fitness doesn't actually reduce the number of power pools I now take. That number is still four, and therefore the number of options available to me didn't drop with inherent fitness. Possibilities that used to be impossible because they required five power pools now can be constructed with only four, because fitness no longer counts.

You're *only* counting the ways options have been removed, but not conceding that options could have been added. Given the numerically small amount of diversity lost through losing four powers out of dozens, any options that are added are important, because we're talking about counterbalancing a relatively small loss.

If you are willing to now state that the loss in diversity due to inherent fitness small, and not extreme, then that small loss is at least partially counterbalanced by the players that ran out of power pools, and now will have one more power pool option available. Whether you understand why players take more than one power pool, it does happen and there are legitimate reasons for doing so, and that small addition to the number of possible options further decreases what you now claim to be a small decrease in diversity by some difficult to estimate, but certainly non-zero value. Without knowing how large this is, its impossible to say if inherent fitness had a net increase or decrease in diversity, because both changes positive and negative are relatively low.

You did specifically say:

However, I provided specific examples that demonstrated the presumptions listed above aren't true, and the net diversity change is debatable: it isn't even obviously lower, because options became available that were not available prior to inherent fitness.

That sort of dramatic "one third" reduction doesn't occur. And you haven't been claiming the reduction is small, but severe even in the general case:

Unless by "decimated" you mean in the original sense of the word: to remove one out of ten. If by "decimated diversity" you meant that diversity has been reduced by perhaps 10%, that's theoretically possible, although I think not likely. Not many people would have interpreted that statement in exactly that way, however.

I wish there was a way to dump all of your powers in [bracket] form into a single tell, just for this occasion.

As in, literally all of them.

No blaster worth the name would have it any other way.

There are almost no bad MMO concepts, only bad executions.

I could tank with that as well. I can also tank with my SR scrapper at 25, with some difficulty. But for me, the question is not whether Regen would be a "good" or "bad" tanker. Those are subjective labels. The question is whether Regen would exist within the nominal range of performance considered reasonable for a new tanker powerset if a new tanker powerset was being constructed from scratch. If it is, it is. If it is not, then just because its a pre-existing set, doesn't exempt it from the same reasonable boundary conditions a new powerset would be expected to follow.

Massive Multiplayer-Optional Really Persistent Game.

I'm afraid at this point I need to explicitly ask what your definition of "more similar" is. Because you're saying that trading fitness which we both had for fighting which we both now have, assuming I stipulate to that trade in the first place, is becoming more similar.

I have to ask you to directly answer the question: how does trading something we both originally had, for something else we both now have, make us more similar. To put it more pointedly, what *difference* do you think we had that we now lost, assuming your assertion about exactly how our builds changed was true. I believe its unambiguously true that if two characters trade a power or set of powers they both have, for another set of powers they both acquire identically, their builds are just as similar and just as different as before. Diversity has not budged. You seem to think that's not true, but you haven't explained how that could possibly be true.

Yes they do, but there's a specific reason why I set both MoG and IH aside for this comparison. If we are just coming up with an average performance, as a simple metric to kick around for discussion purposes, I can and have in the past averaged in the peformance of IH and MoG. But we aren't talking here about average performance when we talk about whether Regen is ready to go for Tankers. What we are talking about is whether Regen's performance is consistently high enough for the most part to sustain continuous or near-continuous tanking of the kind you would see in the average team. So we don't need a Regen tanker to be a powerhouse steamroller tanker, but we do need it to perform at the pace of average players. We also cannot rely on averaging out low uptime powers. We also have to consider actual leveling, which isn't typified by having lots and lots of IOs. We normally consider SOs, even if only temporarily, as a critical segment of a tanker's leveling experience.

So the first big question is: can Dull Pain be perma with SOs. And the answer is no. We have two possibilities: its up 2/3rds of the time, and down about 60 seconds a cycle, or we toss in Hasten, which has a complex behavior given that *it* won't be perma either, but will reduce the average downtime of Dull Pain to about 30 seconds. The worst case downtime will be significantly higher.

Unless we're presuming Hasten will be a mandatory power for Regen, we have to consider the 60 second downtime scenario. The second question is can that downtime be covered by IH and MoG collectively. And the answer to that is also no. IH has a 650s recharge time. Its going to be available once every 325 seconds, more or less. Dull Pain's duration is 120 seconds. With 60s of downtime, we would need IH to be available every 180 seconds. Basically, IH can cover every other Dull Pain downtime window, if we save it to use only during those windows.

MoG's recharge is only 240 seconds, so it can also be available every 180 seconds. But its duration is only 15 seconds: it doesn't last long enough to cover a Dull Pain downtime window, which will be 60 seconds. And its only available at 32 (slotted for recharge by 33). Instant Healing is available at 18, so it can cover a tanker's leveling experience from a relatively early level. But the levels from 18 to 32 are not insignificant.

So we have to regimes to consider: pre-32 without MoG, and post-32 with MoG. Pre-32, we have to consider the case of a 60 second downtime of Dull Pain where IH is recharging and MoG is not available. We can then calculate the survivability limit of that 60 second window, assuming extremely well slotted Regen powers (which is the best case scenario). Best case, we have about 570% total regen, which is 100/240*5.7 = 2.38%/sec health recovery, and about 50%/30s = 1.67%/s from reconstruction. That's about 4.05%/sec health recovery. At level 50, just for comparison purposes, that would be about 76 h/s recovery. If we assume well-slotted resilience as well, that would be about 19.5% resistance. If we did these calculations at level 50, that would imply the maximum incoming damage rate that was still survivable in 60 seconds was about 115 dps. At level 50, that would be the rough equivalent of 10-12 minions or equivalent.

That's quite a lot for a scrapper, not so much for a tanker. A small mix of minions, LTs, and a Boss could exceed that level of damage. Now, scaled down into the level range we're interested in from 18 to 32, the situation could change. So lets see: at level 50, minion melee damage scale is 120.48 compared to tanker base health of 1874. At level 25, minion melee damage scale is 66.03 while tanker health is 1027. At level 50, it takes 1874/120.48 = 15.6 scale minion melee damage to deal base tanker health in damage, and at level 25 it takes 1027/66.03 = 15.6 scale damage also. So actually, the situation scales down fine, at least to around level 25.

However, the average level 25 minion has slightly less damage than the average level 50 minion. How much less is pretty complex to calculate. Lets say its approximately 10-15% less: that's not likely to be far from the correct value.

Overall, then, we're looking at something like 11-13 minion equivalents as being survivable on average by a level 25 regen tanker with very strongly slotted mitigation, before power pools are tossed into the mix (which would help, but in this level range they would also steal slots from the primary). On average, LTs deal about 50% more damage than minions, while bosses are all over the place but tend to deal at least 4-5 times the damage of a minion (the average is closer to 6). So a spawn with four minions, two Lts, and one Boss, deals the equivalent of about that 11-13 minion equivalent, and that's a spawn strength that can occur in four player teams.

The rough calculations show that spawn is *barely* survivable on average inside that downtime. Random chance will make it occasionally not survivable.

Now, the point of the calculation is not to actually calculate the precise survivability of a Regen tanker, so quibbling around the edges isn't particularly useful. Its intended to provide a rough estimate of how regen tanker performance is going to oscillate around, to ensure the downtime isn't severe enough that averaging it out masks a problem. And it seems that averaging it out is dangerous, because the downtime is relatively long (60s in this case) and the level of vulnerability during the window is high enough to encroach upon the minimum level of performance tankers should actually have, at least as the devs have stated that minimum in the past.

The point is its close enough to be too close. Even if I'm off and in actual play that situation ends up being barely survivable instead of randomly not survivable, its still too close for comfort. If Dull Pain was perma with SOs, I'd retract my concern. If there were better ways to cover the downtime, than currently exist, I'd probably also retract my concern. I personally don't need Regen tankers to be uber. I just require them to be functional, as tankers have been defined to be functional. Right now, its too close not to consider reexamining the set before porting to Tankers. You're going to have to do it anyway when you start deciding where to put the taunt aura, which the devs have all but stated must be there.

In the contest of this thread, I believe practically everyone except you understands that what I meant by "improves more" is "has their survivability increase more, given the conventional definition of survivability that has been standard since 2005."

Yes, I remember this line of thought. The conventional notion of being "more survivable" is "being able to survive more damage." Its not "taking longer to die in non-survivable situations."

In your first example, the survivable level of damage of A is 33.3 dps, and for B its 111.1 dps. B is enormously more survivable. But adding 20 percentage points of additional mitigation significantly reduces the gap between them: A increases to 50 dps, while B only increases to 142.9 dps. B has gone from being 3.33x higher in survivability to only 2.86x higher in survivability. The difference isn't huge, but its there.


Now, your argument seems to be that this is situational: that there exist theoretical situations where a resistance power stacked on regeneration can improve survivability by a larger factor than when stacked in a resistance power. That's a given, given the mechanics of regeneration. Everyone (who understands regeneration) knows that unlike defense and resistance which mitigate proportional to incoming damage, regeneration mitigates a fixed amount of damage per time regardless of incoming damage. And therefore for basically *any* two combinations of resistance and regeneration, there is a break-even point for them; even ridiculous pairings have a break even point.

If I say 80% resistance is better than +5% regeneration, a nit-picker could say that is a situational statement, and produce numbers to prove it. Specifically, if incoming damage was extremely small, then 80% of it can be made to be less than 5% of net regeneration. Those would be cases where incoming damage was small enough to be negligable, but they would exist. Similar situations exists with lower resistances and very high regen: you can always calculate the break even between the two situations, and then find cases where the regeneration would be better.

But that then makes comparisons meaningless. You just have to keep putting in disclaimers everywhere saying "assuming we're comparing survivability based on the highest survivable case."

In any case, I'm perfectly fine with you continuing to promote your theory that the normal way mitigation is compared is wrong, because it doesn't account for all the cases where the player will die. I'll continue to use the comparison point of the best survivable situation as the starting point, and closed survivability points as modifiers.

Most players have to be able to play their tankers from levels 1 through 50. At what level do you expect them to be slotting enough IOs to have perma Dull Pain.

What I said was that I thought your numbers, which required essentially taking a boat load of power pool powers and burn every single slot and then some on defense was unrealistic. When I say taking a gigaton of power pools is unrealistic, that does not mean I am saying the opposite, that what people do is take *no* power pool powers. When I say something is unrealistic, what I mean is that thing is unrealistic. I do not mean the exact opposite equally extreme situation is more likely.


My baseline for how well a tanker set works is how likely it is to work when actually played by actual players. Which was my point with blasters: however players play blasters, blasters are judged by that average. If all the players playing blasters played with their eyes closed, then blasters would be judged on that basis. If they were all range-capped billion-inf builds, they would be judged on that basis instead. The baseline for how well something works is how well it actually works.


In fact, numerically speaking a Regen tanker would be about 12% stronger under SO slotting conditions, excluding instant healing. However, that performance would oscillate up and down as Dull Pain went up and down, and ditto Instant Healing: its uptime means while you can calculate an average, in actual play a regen tanker would sometimes be far higher than the average, and most of the time be far lower than that average.

In the worst case reasonable scenario, a Regen tanker without perma Dull Pain would experience significant stretches of having lower survivability than a Fire tanker. And that factors into my judgment that Regen would need to be adjusted to be ported to tankers. Fiery Aura's current strength is higher than it used to be because Healing Flames was buffed. The implication is that Fiery Aura prior to the buff was canonically underperforming.

Fiery Aura ought to be outperformed by all tanker sets that do not make the same trade it does trading mitigation for offense. A hypothetical tanker regen would not really qualify for making that trade. Therefore, Regen should be clearly and unequivocally stronger than Fiery Aura. But its barely stronger on average outside of Instant Healing (which has low uptime) and during Dull Pain downtimes its performance can actually drop lower than Fiery Aura. Its borderline, but I believe that dip makes it impossible to state that Regen would *unequivocally* outperform Fiery Aura in the general case, within the context of sustained tanking.

Which is why I say that while it would not take an astronomical buff to make regeneration appropriate for tankers, in its current state it is simply too far over the line in the wrong direction.

Pool powers are irrelevant to this point, because Fiery Aura and Regeneration benefit roughly equally from power pool mitigation powers. If anything, Fiery Aura benefits more due to having higher smash/lethal resistances which would stack with tough. Neither have any defense to stack meaningfully with pool defenses.

The one area you might think would benefit Regen is +recharge, and specifically Hasten. But that's not true: Fiery Aura and Regen are remarkably similar in terms of non-health recovery mitigation. The difference comes down mostly to their differences in health recovery. Regen's health recovery comes part from regeneration and part from heals. Fiery Aura comes almost all from heals. FA's health recovery thus benefits from recharge more than Regeneration's does. The added mitigation Regen gets from closing the Dull Pain gap is not enough to compensate. You add +100% recharge to both (i.e. Hasten and global recharge) and the 12% edge Regen has on Fiery Aura on average vanishes: the two essentially equalize.

But that equality comes with the caveat that Fiery Aura traded down to that level to get better offense, while Regen starts there and gets nothing in return.

I swear, that hand grenade was just sitting there. Along with that nuclear warhead and that Warrior's clothes and the Longbow Mech.

They didn't: at least, no one person knew and understood both exactly how it worked in the engine, and exactly how it was being used in the data (the powers).

When I wrote those, defense and tohit was a big unknown and also in flux, and there was no paragonwiki. Paragonwiki does a much better job of illustrating the critical information, so I've left it to them to keep that info up to date.

Never say never, but honestly the number of things that need guides that don't have people willing and able to write guides is pretty small these days. I've sometimes thought of updating my scrapper comparison guide, which is really more of a calculation methodology guide: its less powerset guide and more Principia Mathematica Mitigation (sorry, my Latin is crummy). If I was going to do that, I would probably study mitigation secondary effects. What's the real impact of endurance drain. How much damage does knockback really mitigate. Stuff like that.

Its just that stuff like that takes time to test, to analyze, to write up.

That's sort of an ambiguous statement, because its unclear what you mean by "overcome." Other Dark Armor players overcome the problems you mention as well, so we have the tools to overcome those problems as well. But if you are trying to imply that the problems I listed for SR only occur late in the game whereas tohit debuffs happen much earlier, I will point out that non-positional psi first appears with the Lost around level 11. The first time you encounter heightened tohit is probably Frostfire's pets, except on the red side where its probably RIP Lts in the teens. SR has no protection against AoE attacks which arrive long before Lucky does at level 26 for scrappers and Evasions does for Brutes. Autohitting damage such as from caltrops first appears with the Tsoo in the teens.

If you're aware of some major tohit debuff problem Dark Armor has at level 2, I'd like to hear about it.

The original designers thought resistance was more valuable that defense because they were wrong. The second gen devs thought defense was not as problematic as resistance because they could just keep adding defensive counters, and they were right but annoyingly so. The current generation of devs wish all attacks were autohitting and defense was some bad nightmare they could wake up from and I was famous for writing a guide to hat options.

By my calculations I think you're going to run out of slots attempting that at level 33 with SOs. At level 33, you'll have 34 discretionary slots. The list above nominally requires three extra slots (assuming endurance reduction slotting, which is mandatory with this many toggles even with quick recovery) for maneuvers, tough, integration, and weave: that's 12. You need two additional slots for fast healing, combat jumping (assuming you skimp endred here), resilience, and MoG (assuming you don't slot for resistance). That's 8. You'll need five slots each for Dull Pain and Reconstruction. That's 10. That's 30 of your 34 discretionary slots. You need some for Instant Healing. If you're slotting it for *both* healing and recharge, you will spend your last four there, still not fully slot recharge or healing (one or the other), and have no slots left to slot quick recovery, stamina, or any offensive powers.

So actually, that's not possible at 33. The earliest its possible is actually 34, and even then its only mathematically possible, and unlikely to be possible in any playable build.


3354/240 * 5.69 = 79.5.

That's not true. First off, CJ+Maneuvers+Weave for tankers is 15.2% defense slotted. Its not obviously true that a build that adds recharge is going to add more defense simultaneously. In fact, Perma-Dull Pain builds don't even have to have CJ+maneuvers+weave in the first place.


If you're saying its *possible* to build a hypothetical regen tanker that is pretty strong, there's no argument there. But that's immaterial: when we ask if blasters are too strong or too weak, we don't paper-calculate what the absolute best strongest blaster is. We don't look at range-capped blasters with acrobatics and say they are fine. We look at blaster performance in the actual game and when they drop dead too often we say there's something wrong. Are players going to play regen tankers by loading them up with power pools and have totally unslotted attacks for most of their existence to get to performance my SR scrapper can get with just a few inventions and aid self?

Probably not. And when I consider what a likely regen tanker is going to look like at level 15, at level 25, and level 35, I find its performance to be not so good.

I think its amusing that when I say Regen is actually one of the better scrapper secondaries when leveling up, a bunch of people are there to say "oh yeah" and when I say I don't think it scales strongly enough to translate into a good tanker set I get a completely different group of people to say "oh yeah." Because the thing is Regeneration doesn't scale like defense and resistance do. It linearly scales with health, and tankers don't have better regeneration modifiers. So its mathematically impossible for Regeneration to be equally as good as a tanker set and a scrapper set. Its either an overpowered scrapper set or will be an underperforming tanker set, or its somewhere in the middle of being a really good scrapper set that will translate into a relatively not so good tanker set. It cannot be just right in both cases, because defense and resistance increase with tanker modifiers while regeneration and healing doesn't, and +health as a scale-invariant change benefits everyone equally.

Everyone who thinks it would be a great tanker set implicitly thinks it is an astronomically overpowered scrapper set. Everyone who thinks its a perfectly fine scrapper set implicitly thinks it would be a terrible tanker set. Everyone who explicitly thinks its fine for both doesn't understand how regeneration works.

Not as you've described diversity. You've stated that diversity of builds has decreased because the actual number of valid options has decreased. If the number of options has decreased, less options have to exist. And if an option exists, it exists, no matter how many people take it.

If you're saying the same number of options exist, but less people take them, that's a completely different thing, and so far you haven't implied that is happening.


You keep saying that, but you have yet to provide a single example of a case where there are only a few options. I've provided several cases of having many different distinct options. Since I've posted lots of builds as examples that the energy/energy combination has lots of different potential options, and has different avenues for specialization that still exist today, perhaps you can post a specific example, mids or otherwise, of a build situation where, at some point during the build process, the number of reasonable options drops to a precipitously low level.

I think that's important, because:

Except that if you are referring to the case where you described the difference between my original build and Angelius' build, I already demonstrated that was logically false. You stated that if we both started off without fighting, and we both speced into fighting when we received inherent fitness, that represents a loss of diversity. But by definition replacing something we both had in common (fitness) with something else we both have in common (fighting) would not change build diversity at all. There's no reasonable way to interpret that as a loss of diversity.

Because of that, I think its important for you to provide concrete examples of your method of judging diversity, because in at least one case you're judging diversity by a wildly different way than I think most people do.

Half the playerbase minus Arcanaville is still about half the playerbase. When half the playerbase minus Arcanaville is not still about half the playerbase, Paragon Studios won't need a long-term game plan.

Compared to, say, Resist Elements?


As to Regen tanks, the old school requirement on tankers is that starting around the early twenties, they have to be capable of tanking for a team of four on standard difficulty without significant ally assistance when slotted with SOs. If they can't for the vast majority of standard content, there's a potential balance issue.

As a very rough estimate, take a regen scrapper, strip it down to SOs, pop two small sturdies, and go run 0x4. That's the *minimum* difficulty I believe a regen tanker is supposed to be able to handle (there's some fudged numbers there: an actual team would kill faster than you will solo, but a regen tanker would be somewhat more resilient than that with higher health and slightly higher resilience numbers: that's my best guess on estimating the relative difficulty).

Personally, I think its very borderline for the average player. If Dull Pain's +health was fully enhanceable, I'd say go for it and see what happens in beta. But with the set in its current state, I don't think it scales well enough for tankers.

Its an exploit with a lower e, meaning the player is taking advantage of an unintended game implementation flaw to create a benefit the design didn't intent. As a lower-e exploit, it could go away at any time without warning and without notification. The devs have no obligation to ask the players permission to change this, nor do they even have any obligation to forewarn players. In fact, if a "bug" in the game suddenly fixed this, the devs would be under no perceived obligation to change it back.

Its not an Exploit with a capital-E: the devs are not going to punish players for doing it. Except possibly by taking it away one day. You can't get banned for doing it or anything, because the devs have acknowledged the bug and said it was unintended but not a priority to address. Such bugs are never the cause of disciplinary action by the game operators.