Arcanaville

You're not going to find people saying I don't love MA, but there's a difference between liking a set just because you like it, and saying you like a set because of objective criteria that may be either shaky or strange to compare to other sets.

For example:

1. MA does not have knockup. MA has knockdown in Dragon's Tail and Knockback in Crane Kick. That might be a minor error, but its worth noting. And unless you have *really* high recharge, Dragon's Tail is not a great crowd control power. Its recharge is too long to generate enough knockdown.

2. The immobilize in CAK is only guaranteed mag 2. That's only enough to immobilize a minion. It has only a 50% chance to generate mag 3, which is only enough to immobilize an LT. There's two problems with this effect. One: you're only immobilizing things you're capable of killing in only a few attacks. Immobilizing an LT is barely useful. Immobilizing a minion is essentially worthless, and I'm not in the habit of saying things are worthless. Two: its a melee ranged immobilize. If you're using immobilize on a runner, chances are pretty good it might have decided to run before you use it. If you're going to give a scrapper an immobilize to deal with runners, a ranged immobilize makes a lot more sense. Take a look at Impale in the Spines set. Its a *guaranteed* mag 3, so it will at least immobilize LTs in one shot every time. And it has 40 feet of range, which makes it an effective runner tool. Now take a look at Focus in Claws. Its also a 40 foot ranged attack, but instead of immobilize it has knockdown. Guaranteed knockdown. So as long as the critter doesn't have KB protection, Focus will actually stop a runner for about five seconds. All runners including Bosses. More than enough time to catch up with them.

3. Ah, stun. MA used to have one guaranteed mag 1, one guaranteed mag 3, and one chance at a ridiculously short mag 3. Then it was changed to a ridiculous chance at a mag 2, a guaranteed mag 3, and a chance at a ridiculously short mag 3. Now it has a ridiculous chance at a mag 2, a 75% chance at a mag 3, and a guaranteed ridiculously short mag 3.

For those not familiar, Thunder Kick as 10% chance at mag 2 stun. That's a 10% chance to stun a minion. Need I say more? Cobra Strike has a 75% chance of mag 3, enough to stun an Lt. That is a decent, but not stellar stun. It it could stack to affect bosses, that would be better. But stacking with itself is problematic because the stun has base 11.9 second duration and the power itself has base 11.7 second cycle time. It starts off with no overlap, and while slotting (particularly recharge) can help, it starts off not easy to stack and without high recharge usually ends up with only a little bit of overlap. It could stack with thunder kick, but TK has only 10% chance to ever stun. It could stack with Eagle's Claw, which is a guaranteed mag 3, but, well lets talk about Eagle's Claw.

Eagle's Claw is a guaranteed mag 3 now, but its duration is ridiculously short: 4.8 seconds at level 50. I should point out that the stun takes effect 1.3 seconds after activation of the power and the power has a 2.53 second cast time, 2.77 second arcanatime. That means that EC's stun only lasts about 3.5 seconds after you activate EC, and only 3.3 seconds after EC finishes and you can use another attack. To the best of my knowledge, that is one of the worst mez durations that exist anywhere. In fact, back when EC had 3.0 second cast time and I didn't know what the delay was between activation and the effects starting, I used to say that Eagle's Claw was the only mez that could actually expire on a +3 target at the same time the power finished casting.

So EC's stun is so short its control effects are not very good on its own, and its a poor candidate for stacking with Cobra.


As some people have said, at high levels control is not as valued as damage or mitigation. But even if you like control, the control is just not there in MA relative to other scrapper primaries. I think Claws has more mitigating control. I think Dark Melee has more mitigating control with just one power than the whole MA set when it comes to tough targets.

The phrase melee ranged minion immobilizer has been such a long-standing joke for me that I lobbied really hard to get something, *anything* else into CAK, which is why it now has defense debuff. But then Castle decided not to make it slottable for -DEF sets, sigh. I understand why, but I'm still going to try to get that decision reversed eventually.


I still like MA: my MA/SR scrapper is by far still my favorite scrapper. But MA has had issues for every second of its existence from beta to two seconds ago. Its a fun set to play, but it has always had performance problems.

I would specifically say that the change was designed to implement the minimum change that brought the set into balance. Its not my fault the American education system fails to deliver the sort of skill set necessary to determine that the implemented minimum change was zero.




The official City of Heroes spokesperson for jiggle physics. Coincidentally also the official spokesperson for the Perfectly Balanced Cannons of Justice.

People used to say they flickered off and on, or sometimes worked for ten seconds and then turned off. I clearly remember one person saying Build Up buffed lucks, because they were never hit while Build Up was up and they had lucks popped.

It was a strange, strange time. Believe me: I attempted to see if I could see what other people were reporting, without any success. The magic words that finally presented a case I could reproduce: it seems to consistently take more lucks than necessary to cap defense.

I think there is some reasonable debate over the costs of higher incarnate powers, particularly rare and very rare. I don't think the costs of unlocking the slot and crafting a common power are all that high no matter how you attempt to do it. Even if you avoid the trials altogether it only takes about 30 threads to buy the iXP to unlock it, and that can be had for thirty threads. It takes three common components to make the common power, and even with zero drops that's 60 threads. 90 threads total. If you're even moderately patient, as in you can wait a week and two days, you can convert that from 90 shards.

I would expect 40 shards plus 5 NotW breakdowns plus 3 ITF runs to get you close to 80 shards. I can believe your friend ended up with 84 - 6 six short of crafting the common power - after the effort you mention. That leaves basically one WST to go to finish that power. Seven total task forces plus your original inventory to get the common judgment, going nowhere near the Incarnate content. Seven hours of his time maybe?

That's not a level of time that I believe is unreasonable, specifically for someone that doesn't want to team in large teams, doesn't want to raid, and in fact doesn't want to run the Incarnate content at all. But that's not my decision to make.


Then just run Lambda. Just running Lambda, even if you fail is going to allow your friend to slot up *all four slots* even faster than he's doing now. Suppose you run one single Lambda and you fail it, and suppose we ignore iXP altogether. Just earn, say, three threads from it and two astral merits in the attempt and you have the equivalent of 11 threads from that run. Three *failed* runs like that will unlock Judgment or Interface by converting threads to iXP.

Suppose you succeed *once*. You're likely to have at least four threads, at least three astral merits (probably more), and one empyrean merit. That equates to 36 threads from breakdown: enough to unlock either Judgment or Interface in one run, and get 80% of the way to unlocking Lore and Destiny. And that doesn't count the successful drop which could be as bad as just ten threads, but more likely will be a common or uncommon drop at least. Lets say you average getting 10 threads sometimes, and a common drop the other times: the two worst drop tables.

Suppose you run just two Lambdas a week with your friend, and you fail half the time. How long would it take to unlock all four slots and slot a common power in each? Well, every week you're going to get 57 threads from breakdowns and one common drop. Even at this low activity level, and even failing as often as half the time, and even getting nothing but the two worst case drops when you succeed, and even ignoring all the iXP from these runs and looking only at the drops, and even assuming your friend was starting from scratch with a different character, in week one he would have unlocked Judgment (30 threads) and got one common, converted one common (20 threads) and have seven threads left over. In week two, he would be slotting Judgment with his next common drop, unlocking Interface (30 threads) converting another common (20 threads) and have 14 threads left over. Week three: he'd be unlocking Destiny (44 threads), converting another common (20 threads) and with the drop and the left over from the previous week he'd be crafting and slotting Interface. Week four: he'd be unlocking Lore (44 threads), getting one common drop, converting one common drop, and be one common drop short of slotting Destiny. In week five, he would be slotting Destiny *and* slotting Lore.

That's ten Lambda tries in five weeks, with five successes and five failures, and one of the worst run of drops on the successful runs ever. This also doesn't count any earned iXP towards unlocking any of the slots, which would happen half the time when you run only Lambda. So I'm actually painting a very pessimistic picture: reality is likely to be a little better than this, at least.

Now, the question to pose to your friend is: given that pace of effort, which you could spread out over as much time as you want, is that too much effort to unlock four Incarnate slots and getting at least one common Incarnate power in each? If it is, I'm not sure I can be of any help beyond that. If that makes the situation less bleak, perhaps there's hope your friend can still play the game on his terms and still get some participation in the system. It may encourage him to participate more over time. If nothing else it buys you a month where he can at least earn that much.

Maybe one time in a thousand. The devs rarely "insist" anything is working correctly, so that specific instance is incredibly rare. But there have been times where persistent problems were brought to the attention of the devs and the players had to persist over a significant period of time to convince them to examine the problem. I can think of only a few:

1. The type-stacking defense bug
2. The labeling error of lucks and insights
3. The bugged RNG in the mission drop system
4. The Pool B drop mechanism
5. The strength of invincibility

There might have been a few others I'm omitting off the top of my head, but of those five only one of them has happened less then three years ago (#3).


My favorite example is the luck (and insight) problem. For years players reported that lucks sometimes were weaker than they were supposed to be, sometimes flickered off and on, sometimes did nothing, and sometimes acted in even more bizarre ways. There was no way to fix them because they weren't actually broken, and strengthening them just because players said they were too weak seemed like a bad idea given that other people saw them act very strongly.

What could explain all of that? Turns out the label were wrong. Lucks used to be labeled 25%, 33%, and 50% defense. They should have been labeled what they are today: 12.5%, 25%, and 33%. That's what they have always been. That mislabeling caused a lot of people to expect one thing, get something else, and infer radically weird behavior.

In one sense, the players were right: there was something intuitively wrong with them. But in another more important sense they were wrong, and changing the lucks themselves in reaction to those perceptions would have been equally wrong. Especially since many of those observations turned out to be basically hallucinatory.

The lesson here is probably the most important one: you shouldn't discount player complaints, but neither should you automatically believe them no matter how many players believe the same thing.

Since almost the day this was resolved, there hasn't been any complaints about lucks malfunctioning. Every strange effect players used to see in quite substantial numbers vanished almost overnight. Imagine how dumb everyone including the devs would have felt if they had actually done something to the lucks, like buffed them.

Totally off topic, but every time I hear someone referring to their "huntsman" all I think about is this. Product of the eighties, I guess.

Well, obviously "the people" would not be okay with it, because the example states "the people" think the devs are wrong. But if the people are in fact wrong, my own priorities would remain what I stated.

This isn't hypothetical either. When the devs first said they were going to buff blasters, the majority opinion at the time was that blasters were not remotely close to being the archetype in greatest need of help. Dominators had problems with domination itself, stalkers had issues with team utility, tankers had overlap issues with scrappers and brutes and defenders had trouble soloing, plus peacebringers were broken and cosmic balance was considered anemic. Blasters, the AoE damage kings and the most popular archetype, couldn't possibly have any problems.

Except for the whole "being dead a lot" thing, the "worst soloing performance of all archetypes" thing, the "worst soloing performance of all powerset combinations thing, and even "worst leveling performance in or out of teams" thing. And the "getting killed while mezzed an order of magnitude more often" thing. None of which was seriously considered a likely possibility before the devs stated that datamining proved basically all of those assertions.

So if the playerbase is demanding attention for kheldians, dominators, stalkers, tankers, defenders - do you turn around and say nope: we're working on blasters because we have proof they are the most in need of work?

The devs say yes, and I agree. Eventually most people came around, especially when those other things also got attention downstream, but many players never (at least publicly) accepted the devs assertions, claiming the datamining had to be wrong because it contradicted their own beliefs.

We should be responsive to the players, even if they are wrong. But if they are wrong, I'm not going to break the game to satisfy them, nor would I take up valuable designer time appeasing them. I would, in fact, do exactly what I said: confuse them. Its not hard. Although I didn't suggest them, the scaling SR passives are an example. What are they worth? I spent a significant amount of time and used some pretty sophisticated math to conclude having all three at level 50 is survivability-equivalent to having about 24% resistance to all but toxic and psi. That's an incredibly complex assertion that most players would not likely be able to derive. In the absence of my estimation, I've never seen a systematic approach to determining their overall value ever posted, beyond some very rudimentary calculations. Most everyone just guesses, and guesses more or less intuitively and randomly.

There are a surprising number of relatively simple ways to make life difficult for people trying to find an easy answer to what is more powerful than what. Most players don't know how to properly assess cancel on miss DoT. People get the average damage contribution of build up wrong all the time. There still isn't unanimous consensus over average brute Fury. And most recently a *lot* of people thought Kinetic Melee would be a dead powerset that no one would play because it was "too slow" and "Power Siphon sucked." I remember players telling me they would be waiting to tell me I told you so. Still waiting.

The truth usually wins out eventually. You don't want to be on the wrong side of it when it does.

Just to elaborate, the other sets aren't one trick ponies compared to MA either. The swords not only have Parry/DA, they also have two AoEs and their tier 9 is a pencil cone. *And* they both have knockup and knockdown. Dark Melee has SL as you mention, plus ToF as you mention, plus Dark Consumption (endurance), plus Soul Drain (a situationally stronger buff than Build Up), plus a power that is one of the few free AoEs in the game (Shadow Maul). Claws and Spines also have the extra trick of having actual range in Impale and Focus.

Fire is probably the only real one trick pony, but that trick is damage: lots and lots of DoT damage.

I did one with eight on Triumph last night. It was a pick up in Pocket D with a mastermind, a defender, two scrappers, two corruptors, a brute and a VEAT. We completed the sabotage phase fine, took out the portals (with the customary "who has acid and isn't using it" part where we have to farm a containment vessel) and then took out Marauder. It actually only took us about seven minutes to take down Marauder with eight from the moment we exited the building. I don't know how many people were double or triple shifted, but I did not see everyone with Lore pets out, and I was only single-shifted myself.

Honestly, it was an almost eerily problem-free run for the most part. I'd do Lambda with that team any time. And some admitted not having done it very often at that point: we were explaining as we went along.

I don't know why, but the mental picture of a "little sub-group" that is "coordinating with themselves on teamspeak" that fails to listen to the league leaders in a trial and jeopardizes it for the entire league actually makes me even less likely to feel sympathetic than the idea this was deliberate griefing.

Actually, there is a right answer. You fix Shepherd's Crook immediately, period. Including the misspelling. You fix Meekling eventually, when you get around to it. You make a highly complex change to Cannon of Justice that no one can really understand, and does virtually nothing. And you let me know you're going to do that, so I don't compare your powers team's math skills to a deranged woodchuck for the umpteenth time.

Enemy AI seems to have a threshold for taking damage without being able to return fire: although its like an "afraid" I call it "frustration." If they get hit and they do not hit back, when the imbalance between the two exceeds a certain point some critters are designed to flee in response. You can see this in the default behavior of the AE critter's standard AI "brain." Make a critter with one attack, and spawn it. Attack it, and it attacks back. Now remove that attack and spawn it. Attack it, and after a couple of attacks it runs away.

I believe its possible the DoT from fire attacks acts like individual "attacks" as far as the AI is concerned: the AI sees "I'm damaged, I'm damaged, I'm damaged, I'm still getting damaged: I have to run away because I can't shoot back." Which may be why Fire is more effective at generating this response than most other attacks which just hit once.

I can only reply to what people say, not what they think they are saying in their own heads. Or I can conclude that the disparity between what they are saying and what they think they are saying is intractably equivocal.

1. My theory is that participation is being measured somehow and leading you to a "supertable" of random rewards. If you have low participation, perhaps you have a high probability of getting the thread or common reward table, and if you have high participation you have a higher probability of getting the uncommon or rare tables. But you always have some finite chance of getting higher value tables at any level of participation.

2. iXP is almost certainly known at this point to be tracked like regular XP: shared among the entire team, but not the associated league. So even if one person is totally idle, it would not reduce their iXP level to zero. In fact, if their team was doing the disproportionate amount of the killing, they would get the most iXP even if they personally did nothing.

3. My data, posted in the thread linked above, makes the odds that there isn't some systematic effect happening in the reward tables, extremely low. Random chance is not likely to produce the results I've seen, even for the relatively small data sets I'm looking at. Its enough for me to draw an objective conclusion. I just don't know *what* is happening to leave the non-random signature I'm seeing.

My only concern is that the reward system should, like it does everywhere else, make some effort to reward merit. Suppose two teams split on a Lambda and one team is far better than the other, not just getting all their assign weapons but also clearing more critters in the process, and *then* having lots of spare time left over to go help the other team. The current system would, I would imagine, reward that team with somewhat more iXP. Shouldn't it?

The problem is really that while that situation presents roughly the same situation to both teams and then gives them both equal opportunity to excel over the other (and its rare one team really does excel so far over the other that it impacts iXP too badly), the adds in BAF create an opportunity for huge asymmetry if the leader allows it. But while I think something needs to be done, I'm just a little uncomfortable saying that since everyone on the team is working to a common goal all rewards should be split perfectly evenly among all participants because that's only fair option. If there was a better way, I would suggest it. I don't have a better way in mind at the moment, though.

That might be an "afraid" temporarily overriding their preprogrammed path. Score one for fire. But I've also seen knockback throw a minion off the road, have them get up, turn in the opposite direction, and almost run out of sight, then stop, turn around, and then head back the way they were originally going. That's kinda odd.

I know they are affected by knockback: in the early part of the escape phase if I'm stationed anywhere near a door on my energy blaster I stand right there and tab-target them before they even exit the doors, and start spamming torrent and explosive blast on them. The minions usually can't make it past the doors: the Lts sail right on by.

Knockback also tends to confuse the minions a bit if you hit them on the paths as they run by: they often get knocked off the road, run around randomly for a bit, then figure out where they are and head back to the road and back into their escape path. I've sometimes seen them run the wrong way for a couple of seconds before figuring it out and reversing back to their escape path.

Entirely possible. I'm trying to figure out if I can write my own damage calculator to calculate damage in the trials. Simultaneously, because its a necessary first step anyway, I will also be counting power activations. Perhaps that will generate a stronger signal.

It may not be until the weekend when I do that, though, because I owe someone else something else, and of course I'd like to get some time in to continue "collecting data."

It wouldn't be in general: it would be a proxy for my team's overall contribution relative to all other teams. I'd be in there somewhere, so I thought there would be a weak statistical signal from influence in general at best.

But there's a catch: on *some* BAFs I was exclusively on an AV team, essentially shooting at nothing but the AVs. On other BAFs I was *still* on a designated AV team but elected to shoot at the adds when I had the opportunity to do so and it didn't interfere with the AV assault (i.e. when my AV was pulling ahead). Doing that noticably improved my iXP rate per trial on BAFs. Since iXP is split among teams as well as far as I know, both influence and iXP would be a rough gauge, I thought, in factoring that change in behavior into the statistics: when I did that, my relative influence earning would go up. So would the rest of the team, but my overall damage contribution would definitely be higher.

I could just *count* damage, but that's a bit more difficult to do in the logs than counting influence. I can't count iXP, because once I unlocked both slots in the physical tree I stopped earning it, and that became worthless as a metric. Influence was the easiest thing to count that might have any relationship to actual activity.

I'm actually finding the trials better in this regard than something like the ITF. The ITF throws numbers at you: you're often in a tunnel with seven other densely packed team mates surrounded by what looks to be fifty targets. That doesn't really happen on the trials, unless they spiral out of control. Each individual thing in the trials is a genuine threat of at least some amount, and it doesn't take many to pose a serious threat. So while clearing the outside of Lambda is often an AoE fest, even there its usually generally coming down to a few one on one or a few on one fights with the bosses. I don't find myself overloaded with targets most of the time. In fact, I'm finding my MA/SR scrapper, a primarily single target melee ranged attacker, to still find lots of one on one fights to get into that are meaningful.

No, what you said, and I always quote what I reply to including this time, was this:

As I said the first time, if you are in such a team where you'd be lucky to get a single attack off before the team killed everything, even if I were to hand you the entire Incarnate tree plus all the purple inventions you can slot, you'd still be lucky to get an attack off.


Yes, you are, because you're not talking about objective team contributions. You're talking about relative perceptions:

That's a perception. You're actually saying there's a certain rate of speed that tankers needed to hit, and they needed blasters to reach that mark and now they don't. And more than that doesn't matter. That's a purely subjective observation, itself about a personal feeling.

To the extent that the Incarnate system makes players redundant in general, they were already mostly redundant. The only sense in which it makes the players further redundant on teams is that it makes players more powerful. *All* systems that make us more powerful will make other players redundant on teams that were not redundant before. That is mathematically impossible to avoid. Once a team goes from "not enough" to "more than enough" they stop needing anything. Everything else becomes redundant. And you can't buff everyone's performance and not create that circumstance. Inventions make people redundant. SOs make people redundant.


The difference between an SO'ed character and a high end invention build is still higher than that between a high end invention build and an Incarnate powered one. If post-Incarnate play was going to create a schism, inventions would have done so long ago.


My point was that long before I20, one force field defender could already make you redundant in the Ghost Widow fight. A bunch of lucks could also do that. You specifically mentioned your specific character properties in dealing with Ghost Widow as being a specific case where the Incarnate system directly challenges your ability to contribute meaningfully to a team. However, I'm pointing out that the Incarnate system has to get in line for that.

Incidentally, what radiation debuffs do you believe are making the largest contribution to Ghost Widow? And what Incarnate powers make that contribution redundant?

A failure (or a disruption to the best of my knowledge) hasn't happened on one I've been on yet because I'm pretty discrete about it, and in some cases I was the team leader and just plain told the team to stay on the AV while I switched to the adds. But I know it is causing problems elsewhere which is why I only do it if the raid is 100% under control and my AV is getting too far ahead anyway, so it would be a matter of shooting at the adds, switching to the other AV, or not shooting at all.

Still, anything that causes people to elect to not follow directions to the actual detriment of the raid should be carefully reviewed.

Hmm, it kinda sorta looks like rares are a little more prevalent above, and commons more prevalent below. But its hard to say. Now look at the same list sorted in descending order by amount of time spent:

Statistically, this is what is known as a "smoking gun." There is no way that is random. In fact, consider this: the only common not at the bottom just happens to have the second worst influence value. Other than that, that list is *sorted*.

Does the same pattern appear in BAF? By influence:

and by time:

Not as strong a signal. But its important to note that BAF is a lot squirrelly than Lambda, because BAFs are more complex in terms of activity. There are different roles you could potentially be on, you could be dominated by shooting at an AV or at more easily defeatable critters, single target or AoE, and the activity is a bit more paced. In Lambda there is a much longer clearing phase and more chances to deal more or less damage across all of them than peers.

Also important to note that all of the above statistics come from damage dealers: an En/En blaster and an MA/SR scrapper. So there is no cross-comparison of damage contribution vs any other kind. Both characters primary contribution is damage.

Although this is a relatively small sample, and the BAF signal is very weak, I believe the Lambda signal is strong enough to state that there is strong enough evidence to conclude that *some* performance monitoring system is in place which weights reward tables based on some measure of participation. What that is specifically I don't know, and its clear that while the Lambda signal is strong, its not perfect, and the BAF signal is weaker, so I don't think it is a simple "do this, get that table" system. It might be a "do this, get this table" and that table itself says "10% common, 60% uncommon, 20% rare, 10% very rare" or something like that. A double-indirect reward table if you like: a table that determines which table you get.


Why the signal is so much stronger for time rather than influence, particularly for Lambda, is something I have some theories about, but none that really make complete sense yet. I doubt it actually *is* time that is being monitored: rather I believe something about longer trials causes *me* to somehow behave differently in a way the game is seeing. Maybe longer trials favors players that deal more damage and have marathon builds. Maybe people die more often in longer trials and on average I die less often and have to go to hospital. Maybe longer trials are longer just because there is less damage on them, and as a damage dealer I'm now a higher percentage of the damage total. Its hard to say without a lot more (and more detailed) data.


Edit: I should also mention this is preliminary data. I did this late last night and its possible there are errors of some kind in it. Normally I would spend an extra week or two verifying the numbers but the signal in Lambda was so strong I decided to post sooner rather than later.