aethereal

All the stuff about party roles are... true enough, just wildly overstated. Look, let's take a look at a Brute. What's its party role? Well, it can be a tank, of course, but it's not as good at tanking as a Tanker (less durable, worse aggro management). It can be melee DPS, but it's lower DPS than a Scrapper or Stalker. So, that's another way of saying that its party role is kinda bad. If you were talking, "How do we try to make a tip-top party that can perform at the absolute maximum in the hardest content in the game," you wouldn't put a Brute in that team. It turns out that lots of content in CoH is pretty easy, and you don't need tip-top tanking to be able to handle it, and a Brute is just fine as the main tank for 95% of the content in the game, and, with even minimal support, can indeed handle the rest of the content. And you also don't need tip-top melee DPS, so even if the Brute isn't being the tank, it's also fine. But people don't agonize about the Brute's party role. Because approximately 0 people care about making the perfect optimal team in CoH, and the Brute can pull its weight well enough. Teaming in CoH runs a gamut from "we genuinely don't give the tiniest shit about team composition, we'll take literally everyone, it doesn't matter if this team is eight Blasters" to "We'd kinda like to get at least someone able to tank, a couple of people who can DPS, maybe at least one support role, and then we don't really care about what else." People who care more about team composition than that are a tiny, ignorable subset of the game. The Sentinel's team role is just "ranged DPS." And all it needs to fulfill that role is a damage boost. If the Sentinel's damage is more-or-less comparable to other DPS ATs, all this stuff about "omg how do we make sure that we're leveraging the exact balance of its armor sets" stuff will evaporate, and people will treat it like they do all the other characters who , yes, aren't quite perfect for the party composition, but are good enough for the very, very, very loose composition requirements that we actually have.

It seems super unimportant to me whether Sentinels recapitulate Fortunatas or not. The Widow AT/Fortunata sub-AT or whatever we want to call it isn't a versatile way to bring to life a wide variety of superhero concepts, it's a narrow lore-tied thing. Sure, you can ignore the lore part of it and use it for some other psychic concepts besides literally Arachnos Widows, but you can't use it to make Iron Man. If Sentinels have a similar game role to Fortunatas, but also evoke Iron Man or a number of other ranged bricks from the comics with their aesthetics, then cool.

Because their damage is too low.

My recollection from another time someone parsed logs is that there are some powers that only report a log-line when they miss (ie, they don't mention a hit roll when they hit), which leads to an inflated view of the frequency of misses from the logs.

People have to be realistic that there's never been a change as drastic as switching the entire type of every primary for an AT and there never will be. Swap the sentinels to assault sets will never happen. There aren't even assault set equivalents for several sentinel primaries (like archery, dual pistols, or beam rifle).

Performance shifter predates the lockouts they've lately been putting on procs (procs: what they need are more illegible, complicated rules), so I imagine that just each power with the proc checks the proc on its own schedule and when one fires, you get your endurance.

It was created for last year's winter event, as I recall.

It can't create bias on the scale that people are looking for. It only create a large bias because you're working with unrealistically small numbers. If you take an actual 32 bit integer and mod by 10, it will, yes, be biased -- we'll be missing one 7, 8, and 9 out of about 4.3 billion.

It is my understanding that the code simply calls the operating system random number generator, it doesn't do anything screwy like implement its own random number function. It beggars the imagination to suggest that the random number generators in any major OS are so badly flawed that they would do something like "produce a noticeably skewed distribution of percentile values towards higher numbers." Like... that would be a HUGE problem that would affect many, many, very seriously more important pieces of software than an old MMO. Once you've gotten a random value from the operating system, all you do with it is basically multiply it and/or mod it so that it falls into the range of values that you want. Managing to bias the results there would be an impressively bad bug. It seems overwhelmingly likely to me on the basis of this that anyone who is finding bias towards missing in either their experience of the game, or in parsing combat logs, is detecting either: 1. Flaws in human cognition that make them create false patterns from random things. 2. Flaws in exactly what is logged and how, or what how the logs are parsed. rather than 3. Flaws in the generation of random numbers between 0 and 100.

I think you could make it do that or not? Normal wings definitely flap when you jump. Do the tech-style wings snap open when you jump? The ones that look like veritech pieces? Rocket boots don't turn on when you jump, do they?

It's definitely codeable. All (or most) wings do different things when flying vs not flying. There's clearly a "costume variant while not flying" and a "costume variant while flying" thing. You just make the one where you aren't flying have no actual geometry.

If Sents do have their AT threat modifer set to 1, like squishies, then they shouldn't -- it should be at least as high as Stalkers (2). All armored classes maybe except Sents have at least 2, and Sents should too. (EDIT: Thanks Luminara!) But these calculations are not, I think, particularly valuable -- they abstract away almost everything of note. The main reason that Tankers keep aggro isn't their AT threat modifier, it's Gauntlet + Taunt Aura + Taunt. Defenders and Controllers generate high threat because debuff and mez generate lots of threat, out of proportion to damage. And using the Sentinel's relatively high damage scalar as an estimate of their damage isn't very valuable -- they absolutely do not do 95% of the damage of Stalkers, for example. Threat is generated from actual damage, of which damage scalar is only one part. If a Blaster and a Sent duo and they both initiate against a large spawn by using an AoE, the threat modifier of the Sentinel isn't going to be relevant because the Sent will only damage somewhere in the 1/3-2/3rds of the mob due to their target caps -- even if they do more damage than the blaster, the blaster will damage everyone in the mob and will get substantial aggro as a result of that.

I don't think that the accuracy has any meaning at all for the pet summon. No to-hit roll is involved.

Guys. Sentinels have blast set, not a support set. Their role can't be support because they don't have support powers. Increasing the scale of their debuffs (for those sets which have them, which is not all of them) might be nice icing on the cake, but it's not the cake.

I have a 50 DB/Ninj scrapper. As people say, it's a solid set. Low DDR and not much offensive boost (there's the crit from "hide" thing, but it's once per fight and with low stealth sometimes hard to set up) are the negatives. It's fun to be totally unconcerned with psi damage. I usually run barrier to paper over defense debuff situations.

DDR one would be problematic. The other two would probably be fine?

I wouldn't pair DM with SR on a mechanical level, since the -to-hit from DM rapidly becomes overkill with SR. I understand the appeal for getting another flurry-style attack on an aesthetic level, though. What about Electric Melee/SR for a Flash-style speedster? Bad side: a lot of the electric melee attacks (esp. Thunderstrike) animate slowly. Upside: Maybe you could imagine that it was a slow-mo camera effect? And Lightning Rod is one of few "charge attacks" in the game. More attacks than DM that look like punching and not dark magic. Arguably better synergy with SR. You can of course still use flurry, spin, and sands of mu. Savage Melee? Its cone attack is similar to flurry, it has one of the other charge attacks, and its single-target attacks animate pretty fast.

This is a fantasy. My first 50 on HC was a sentinel. You take damage, in team settings. It's not like nobody ever shoots at you. Also, the concern that you'll become an offensive powerhouse by slotting your attacks is misplaced -- lots of people in normal builds slot their attacks for lots of damage. Even if a sentinel did have more slots to play with their offensive potential will be more determined by the basic stats of their powers than anything else. There will always be scrappers/stalkers/blasters out there who also put 4+ damage procs in their attacks. If their attacks/inherents do more damage than yours does, base, then they'll also do more damage when both of you are devoting all six slots to optimal damage. We don't know what the AT modifier is for threat for a Sentinel. It's possible it should be tuned up! It's also possible that it's just fine, and the fact that sentinels don't attract a ton of aggro is because their damage is too low. But if threat does need to be tuned up for Sentinels, it doesn't need to be tuned up a lot. But also, in at least most content in the game, teaming up with a good tank or debuff defender or controller means that there's very little threat to go around. This problem that you perceive of OMG this character doesn't generate enough threat for their armor values, I assure you that Stalkers and Scrappers experience it as well. The main reason I used to go haring off from the group as my old Ice/Bio stalker was that I didn't need to use my defensive clickies. On my DB/Ninj scrap, when I just duoed, not 8 person group, I would forget to watch my endurance bar, detoggle myself, retoggle myself, and keep going. The game limps on. I'm getting repetitive (or maybe I'm long past "getting"), so this'll be my last post on this topic, but, look: At least 80% of the problem with sentinels is that their damage is about 25%-35% too low. If people want to come up with some clever, complicated way to increase their damage by about 25%, okay, cool. But it doesn't need to be clever and complicated. You could just increase their damage scalar, maybe give build-up instead of aim, maybe do some fairly straightforward damage-increasing inherent. The important thing is that you increase their damage by 25-35%, and figure out whether it's 25% or 35% or what. If you then want to also do something where you create some unique team role, sure, cool! But that won't work without increasing their damage, because the main problem with sentinels is just that their damage is too low. And they don't really need an elaborate team role, they'll be fine without it.

Just increase their damage to be competitive with (just a sliver lower than) a Scrapper or Stalker. Don't give them Blaster-scale nukes. That's really it, from a balance perspective. It won't make any AT obsolete, and there is not a highly delicate situation where one false move dooms an AT. People will play Sentinels, Stalkers, and Scrappers because their superhero concept does or does not demand ranged damage, stealth, or whatever. None of those ATs will usually be the absolute best choice you could make to fill a role on most teams, but they'll all be fine. (I think there's another another dimension of a sentinel revamps that's about injecting a few more decisions into their play experience. Just "a blast set that does a lot of damage" plus "an armor set" might be a touch boring to play. Their current inherent, despite its many flaws, does provide a certain amount of choice that, I think, is welcome. But probably there are plenty of people who are there for a straightforward play experience, and in any case, that's a different problem than balance or team role.)

Every damage class besides Blaster does fine with just one of their power-sets being a damage set. Scrappers and Stalkers are in great shape. It would be trivially easy to tune Sentinels to do plenty of damage with their primary -- it would be easy, if someone wanted to, to make them do more damage than Blasters with just their primary (obviously, that wouldn't be good design. I'm just saying that the idea that there's a problem with Sentinels "only" getting one powerset focused on damage is a non-problem). People spend a lot of time trying to make the issue with sentinels way more complex than it actually is.

Stalkers generate very little aggro. See here: https://hcwiki.cityofheroes.dev/wiki/Threat They get about 2/3rds as much per constant damage that scrappers do (while typically doing around the same amount of damage that scrappers do), and my experience playing various Stalkers is that on a well-functioning team, their armor is indeed serious overkill. Honestly, same with scrappers unless you're playing one of the sets with a taunt aura and there is no brute or tanker on the team. But that's not a big deal, because, as it turns out, as long as you're doing your primary job (doing damage) well, having more survivability than you need isn't a problem. Maybe it's not perfectly ideal, but this isn't a game where people generally embrace highly deterministic team role and tip-top performance.

Do you feel the same about Stalkers? Scrappers?

Stalkers are contributory as an armored class with no taunt (and inherently low threat!). It's just that they do more damage than sentinels. I think it's instructive to look at how much more. Stalkers have a 1.0 damage scalar. On an 8 person team, they have a base 31% chance to critically hit, so that's essentially 1.31 scalar (but also you have your ATO and Assassin's focus, so let's say it's a total of 40% chance to crit). They have build-up, which is recharged by their ATO: it has a 5% chance to recharge for every attack (I think that's how it works). With the Gaussian's proc, each build-up does +160% damage for 5 seconds, +80% damage for another 5 seconds. Build up's natural recharge is around 30 seconds (90 seconds base, 200% recharge through some combo of local + global): using it 3 times per minute is not unreasonable, that means that in any given minute we have about 5 seconds used casting build up, about 15 seconds of +160% damage, about 15 seconds of +80% damage, and about 25 seconds of +0% damage. If you have about +100% damage enhancement normally, then that all comes out to about 3.4 scalar. Sents have a .95 damage scalar. They get 5% damage resistance debuff. They get Aim, which let's say they get it every 30 seconds, and have the Gaussian's proc in it, so using it twice per minute means they get 10 seconds of +100% damage and 10 seconds of +20% damage. And they get opportunity, which gives a large damage resist debuff, but only against one target, plus a damage proc (if offensive) for the Sent. The resist debuffs scale poorly against higher-level opponents, let's say that you get an effective 4% debuff. They get about 50% uptime on opportunity. If you discount the 20% opportunity resist debuff, then you have the whole thing comes out to about 2.15 scalar personally. Now, sents magnify the rest of the team, too -- but if it's only 4%, and it's not on alpha, and the rest of the team isn't necessarily high-end damage performers, that doesn't come out to that much. If the rest of the team averages 2.5 scalar (ie, the average for the rest of the team is above the sent's personal contribution), then even if they get the entire debuff all the time, that's only another 0.7 scalar for the sent. Those are rough numbers, they don't capture the full complexity of the situation, but honestly I think it's more generous for the sent than it should be. All the stalker really needs to do is hit build-up when it becomes available. Sentinels have to try to keep their damage debuff on enemies, be on high-damage teams, use offensive opportunity when it becomes available, etc. This is also not counting the fact that Stalkers get simply higher-damage basic attacks. It's easy to look at the sent .95 scalar and the stalker 1.0 scalar and the scrapper/blaster 1.125 scalar and be like, "Oh, okay, the sent does like 10% less damage than those classes." But opportunity is WAY less good than crits. Blasters getting aim + build up, and the stalkers getting build up plus free recharges of it, and scrappers getting build up are all significantly better than sentinels getting just aim -- uptime for these powers is large! Getting high base damage attacks like snipes, blaster nukes, assassin's strike, etc is a big deal. These magnify the difference and leave sentinels a major step down in damage compared to the other DPS classes. And that, not team role, is why sents aren't that good. Fix their damage to be comparable to Scrappers/Stalkers/Blasters (doesn't need to be quite AS good as theirs, but close), and you'll generally see the concerns about what are they doing on teams melt away. The answer to "what are they doing on teams" will become "they're doing damage." They no more need a protective role than stalkers do. Like, the idea of having a single-target taunt function that lets them bodyguard certain other players is fun and all, but most people aren't going to actually understand that and use it. Giving some kind of auto self + damage or enemy debuff thing is, like... a fine answer to the damage deficit, I guess, but you could also just give sents more damage.

I'm not sure. I didn't know until now that rebirth had a different version of stalker ice melee. But my guess is that live didn't have ice melee for stalkers. Score probably proliferated it, and when they did so, chose to drop greater ice sword for assassin's ice sword. Then rebirth separately proliferated it (starting with the i24 code base that had no ice melee for stalkers) and when they did so, they chose to drop ice patch for assassin's icicle. Just a guess though. Edit: thought of a simple way to check. Paragon wiki archive lists ice melee as a set only available to tankers. So yeah, rebirth proliferated it separately.

That's their default animation: their power is called assassin's icicle, whereas ours is assassin's ice sword and is a sword by default. I agree that the icicle look is a good option, and one somewhat low impact way we might diversify our set would be to import the icicle look for our sword attacks. That preserves the connection between those attacks' lethal damage and the animation, while still making the power seem a little more less of a "I can just make anything it of ice."