aethereal

I think a lot of players don't successfully build to maximize the (potentially very useful) scaling resists on SR because they're so "invisible." So a few points: 1. You need to successfully survive for a bit at lower-HP totals in order for your resists to kick in. Max hitpoint increases are deceptively useful for most sets, but it can seem like SR doesn't particularly need them because it doesn't have a heal. It also doesn't have a place in-set to slot the +7.5% HP unique from Unbreakable Guard. But you should be seeking +hp (slot the unique in Tough, look at +hp set bonuses, get the accolades) to give you more breathing room between when your resists kick in and death. One reason why Tanker SR is so incredibly tough is that Tankers have enough hitpoints to survive at like 20% health when they have very strong resists. 2. Resists, just like defenses, are increasing-returns. If you have 50% base resist, +10% resist is twice as useful to you than if your base resist is 10%. So scaling resists will be more impactful if they're building on top of a decent resist basis (which you'll have to build out of tough + set bonuses). 3. I mean, this is still your secondary level of mitigation. Scaling resists, especially on non-tanks, are not going to keep you alive in situations where your opponents bypass your primary defenses (such as scenarios where enemies are auto-hitting or have a massive to-hit buff (such as Quartzes for Devouring Earth)). You'll still want beyond-softcapped defenses -- even SR's DDR lets some debuff through, you need a buffer. 4. An orange insp or two can take your somewhat-enhanced resists from mediocre to excellent, consider popping them in difficult situations. They'll also slow your rate of HP loss and let the activation of the scaling resist power catch up.

Brute and Tanker are the same as Scrapper I believe -- only Stalker is different.

ALL BRUTE ARMOR SETS HAVE TAUNT AURAS. Yes, including Energy Aura. Yes, including Super Reflexes. Yes, including whatever armor set you think doesn't have a taunt aura. On Energy Aura, the taunt aura is Entropic Aura.

Yeah, you have to have visible armors in PvP, presumably so that players will realize if you become vulnerable.

This doesn't sound true to me. Maybe you were just very briefly mezzed?

The way this is implemented is though damage types (I think). Fires take cold damage. If you wanted them to take damage from earth powers, they'd need to take smashing damage, and then you could punch the fire out.

People seem to have largely ignored my proposal back on page 2 or so of making procs have a percentage chance to fire and then giving powers a materialized multiplier to that chance, but I really think it does a good job increasing legibility of procs, removing some of the senseless quirks of PPM, and allowing us to tame a few powers that may be genuinely overpowered with procs without nerfing all procs into the ground (which, I agree, is not warranted).

This would both be incredibly disruptive to existing builds (and what would we even do? Force respecs?) and also actually really easy to make overpowered. We don't need Neutrino Bolt more-than-doubling its DPA because it's guaranteed to fire a 100 damage purple proc with every invocation (25 times per minute).

No fundamental changes. There were beta changes to the "internal timer" of epic powers (that basically made them proc as though their recharges were half as long). People have claimed that in the private beta they briefly fixed and rolled back burn. But they've never made even beta changes that has sweeping effects to all procs.

And to be clear, we've now moved on to, "Sai realizes that he's wrong, but doesn't want to admit it, and now is going to try to claim to have been right all along and simultaneously claim the moral high ground." The data is not, in fact, misleading, and you seem to have largely accepted that. And I can tell you things you don't know about the power data, because you started off trying to claim that it was misleading and are now abandoning that claim. See how it works?

So just to be clear, this was something that you felt strongly enough to post about four times, with dire warnings about how "three of four sets underperform," but now all of a sudden you're too shy to say the name of a power. Huh. Could it be that you kinda know that in fact none of these objections you've been raising actually are important, and that if you do take half the time you've spent replying to this to get specific, you'll be proved wrong? (The specific ways that Rad Blast and Dual Pistols underperform are not particularly important to the comparison that we're making here today. We can use ice blast or whatever, we'll still find that for ST powers for blasters, a damage proc enhances damage by somewhere around 35% plus or minus a few percent, and for a non-nuke AoE power, a damage proc enhances the damage of the power by 50% plus or minus a few percent. This is specifically because damage of a power mostly follows a formula in CoH, and what makes a set under or overperform are mostly animation times, which don't have a big effect on the comparison between the damage of the power and the proc-rate-times-proc-damage in that power.)

I'm sensing a considerable amount of evasion here. Okay, so since you won't take five minutes to enter data into four cells of a spreadsheet, tell you what. I'll do it for you. All you have to do is tell me what power, and also explain what you think the result is going to be.

I think they're broadly average, at least in the powers I chose. Empty clips may be slightly low damage, but it captures a range. If you feel like you're going to discover some game changing difference by putting in a beam rifle power, you can copy the sheet and then add another line to it. All you have to do is copy on the recharge, animation time, damage, and if applicable radius and arc from CoD -- the formulas I put in will calculate the proc rate etc. Let us know what you find (but I don't think you'll see anything wildly out of line with what's in there already).

What I was suggesting wasn't a code change. It's something that can be done within the existing power effects framework (unless there's something I'm not thinking of, but I'm pretty confident there isn't). It might have unintended consequences of one kind or another, but probably not actual bugs. I modeled it off of how Energy Transfer's health "cost" works, and that doesn't seem to have bugs.

You're right that is not counting incendiary ammo. I was deliberately trying not to choose powers that are "great." I wanted to give a sense of the performance of procs in "ordinary" powers.

It is true that back in the days of fixed proc rates, procs were better in AoEs than in ST powers, but the rest of this is wrong. We're looking at the damage increase per target, not the "chance of the proc firing somewhere on at least one target." Because in this subthread we're talking about damage procs (the other metric would be a good one to look at for self-buff procs). The reason that procs are better in AoE powers, in both the flat-chance and the PPM regimes, is that AoE powers are drastically lower-damage than ST powers, and procs with their flat damage will be more significant compared to the lower-damage AoEs unless this is compensated for by making procs trigger less often in AoE powers. The PPM regime does do this, but the reduction in proc rate is mild compared to the difference in damage. For example, Cosmic Burst and Empty Clips both have the same 10 second recharge time, but the proc rate of a 3.5 PPM proc in Empty Clips is 36.4% vs 70.41% in Cosmic Burst -- so half the rate. But the damage of Cosmic Burst is about 3x the damage of Empty Clips. If the HC team pegged AoEs at a "ridiculously low proc rate" such as 20%, the value of a damage proc in Empty Clips would be less than the value of a SO damage enhancement (albeit of course ignoring ED).

So, in an attempt to add a little bit of rigor to this whole conversation, I made a spreadsheet. https://docs.google.com/spreadsheets/d/1exC_bOWKCrjQ4eCjN12Cs5ETFPOpJCaBjmBnUIyMXHU/edit?usp=sharing This is not at all the final word on procs, but it shows how damage procs perform in a variety of powers in blast and melee sets, and specifically what they cash out to in terms of damage enhancement. I chose, for each AT, a "normal" ST power (a T2 power), a cone, a spherical AoE, and a heavy-hitter ST power. I used the same power for each AT except that I forgot that Stalkers don't get Dragon's Tail so I used Spine Burst for their spherical AoE instead. I tried to avoid powers that I thought were weird special cases like, say, Energy Transfer or whatever. I'm ignoring crits and scourge, so note that stalkers, scrappers, and corruptors will find procs to be less good than a naive reading of this spreadsheet would indicate. I'm including AoE sizes before Tanker Gauntlet bonuses, so the proc numbers are right, but the actual AoE sizes for Tankers are smaller than they are in play. Some takeaways: 1. For Blasters, in ST powers, a proc isn't really much better than an ordinary damage enhancement except that it ignores ED. For Scrappers, it's worse once you factor in crits. For Stalkers, about on-par. For other ATs, even in ST powers, the raw numbers of a proc are much better than you'd get from normal damage enhancement. 2. As @Luminara alludes to above, but is I think frequently forgotten in these debates, procs are much better in AoE powers than ST. A normal 3.5 PPM damage proc in even a Blaster AoE is like more than 45% damage enhancement! In Defender Empty Clips, each proc is worth the entire damage of the power. We often focus on things like pylon tests when talking about DPS, but procs are really good for AoE DPS. 3. It is certainly the case that most ATs will get a lot out of replacing any singleton spot in their build with a damage proc. Like, if you aren't missing the set bonus anyway, procs are way better than ordinary damage slotting except for ST powers in blasters, stalkers, and scrappers (maybe AoEs as well for scrappers). Is it worth the loss of set bonuses to 3 or 4 or 6 slot procs? I dunno, more complicated question. 4. Obviously, as noted by many others, but worth mentioning, any nerfs to proc damage will make it harder for the non blaster/scrapper/stalker classes to be competitive to the blasts/scraps/stalks in terms of DPS.

"Drastic changes" seem bad. Like, look, if (if!) procs are overtuned, then maybe they need 10% cut off their damage, or their PPMs turned down a little, area factor adjusted, or their max-chance-to-proc set to 80% instead of 90%. I'm sure someone will be on this thread shortly to say that if procs become 10% less effective, it's the worst thing in the world, Homecoming will be abandoned forever, their character will be literally unplayable, dogs and cats will be living together, mass hysteria. But... they'll be wrong. I think that far more than procs being overtuned, they are just a kludgey, bad design, but as I outlined in the last page, you can pretty easily correct a lot of that without making a giant difference in the way existing proc builds perform. @Bill Z Bubba's "drop all proc damage by 75%" is bad. That shouldn't happen. But there's plenty of room for substantive change to procs that doesn't upend the entire world.

If Luminara uses bolded words, then you have to do what they say.

Meh, people just brought it up, nothing particular about this except that it's been a while since we've had the arguments and there aren't big other topics to discuss right now.

I'd go the other direction -- make no recharge affect it, including local. The reason that global doesn't affect it was because of a (reasonable, IMO) objection that people had that "hey, I don't want to be telling people don't speed boost me because my procs will suddenly stop firing." There are lots of sources of external global recharge, and it's kinda conceptually sucky to not be able to plan anything because who knows, someone might hit Ageless. So I think the solution is just "everything is based on the base recharge of the power, doesn't matter what recharge you build off that." That could very reasonably include a nerf to proc rates -- lower base rates, but local recharge no longer affects them.

I have some bad news. (The chance for a proc to fire is capped at 90%.)

So, here's a suggestion that I do mostly endorse. The problem with procs as I see them is: First, they may need some fairly mild nerfs. Definitely not a 75% reduction in damage. Second, the interactions with recharge time are bad. They're terrible. They lead to all this proc bomb stuff, they lead to things that are supposed to be disadvantages for powers being advantages instead, they lead to the absolutely insane local/global recharge time break and focus the meta even more on global recharge. So, solution: 1. Procs go back to having a percentage chance instead of a PPM. 2. Powers have a multiplier to that percentage chance, that is noted in the in-game info for that power. 3. That multiplier is based mostly on recharge time and area factor, but it's like damage -- there's a formula for it, but it's an instantiated value in the power definition, not an instantaneously calculated formula. It doesn't vary depending on actual recharge time, and it can be set to a different number than the formula indicates, if the power designer feels like that's important to do. 4. We create the percentage chances and the formula to hit whatever overall damage numbers we feel is important. So for example, a normal damage proc might have a base chance to proc of 20%, and Neutrino Blast might have a "x1/2 proc rate," so that damage proc procs at 10%, while Cosmic Burst might have a "x3 proc rate," so the damage proc would proc at 60% in Cosmic Burst. Advantages of this solution: 1. It is vastly more legible to the players of the game who do not read the fora obsessively. They can see in-game one number, multiply it by another number, and know the proc rate. They do not have to know about whether a power has local or global recharge, whether it has a pseudopet or not, and then use a fairly complicated formula. One multiplication, no external sources of knowledge, done. 2. It can neatly sidestep the pseudopet problem. 3. It neatly solves the epic attack problem. 4. It allows individual powers that are "too good" or "too bad" with procs to be tuned individually. If we don't like DNA Siphon being able to be a proc bomb, we can just reduce its proc multiplier. 5. It returns fast-recharging to, ceteris paribus, being an advantage, not a mixed advantage/disadvantage, and vice versa for slow recharging. Disadvantages of this solution: 1. It means juggling one more number for every power, there are opportunities to get it wrong, and it's work to set up the right number in the first place. I think it's a manageable amount of work, if we use the current proc rate formulas as a design formula, but we are still going in and having to individually set the proc rates on hundreds of powers. 2. The current system to some extent "auto-balances" based on local recharge. With the new system, you'd have to just kinda set some values and then someone improving local recharge would make their procs better in that power. That doesn't bother me (it's how normal damage enhancement works, and the idea that this is a really meaningful balance in a meta of 180% global recharge is kinda laughable), but might bother other people. 3. Damage proc rate and non-damage proc rate are still conjoined. If you want a hold to be able to proc Lockdown +2 Mag at a decent proc rate, it'll also necessarily be able to proc the four Hold damage procs at a decent proc rate.

So here is one view that I kind of generally hear people hinting at. I don't endorse this view, just saying that I think some people have it: "Procs in normal damaging powers are fine. My problem is that certain powers that do a trivial amount of, or no, damage or which have recharge times that are very different from the recharge time they 'ought' to for that power (such as epic attacks) can be loaded with damage procs and do a lot of damage." Proc rate is based (mostly) on recharge time. When the PPM system kind of mostly works, it basically scales a proc's effective damage to the power's effective damage, since (most) powers base their damage on their recharge time. So one way to satisfy someone with the above view would be to cut out the middleman. Instead of proc rate being based on recharge time, it could be based on actual damage. Some (depending on what you think of the problem) advantages here: 1. It cleanly provides a solution to both epic attacks and things like using DNA Siphon as a proc bomb. Epic attacks would have a normal proc rate for a power with similar damage, not a really high proc rate. DNA Siphon, which does trivial damage, would have a very small proc rate. Powers that do no damage at base would either have a 0% proc rate or some floor (5% or whatever). 2. It removes some of the current ambiguity about recharge time from the game, where like, if you have a shorter recharge time than you "should," is that an advantage? Well, kinda, but for procs it's a disadvantage. We could meaningfully say like, "Hey, an advantage this power has will be that it breaks the design formula by having a shorter recharge time," or "a disadvantage that this power has will be that it breaks the design formula by having a longer recharge time." 3. It would probably remove at least some of the current problems with pseudopets, which don't interact well with recharge-time-basis for PPM. 4. It at least starts to unwind some of the insane recharge time shenanigans that we get with PPM. But there are some significant flaws in this idea: 1. Recharge time is a (mostly) unitary value. Powers have A Recharge Time. A very limited number of powers have some complexity to their recharge time (like, it's unenhanceable, or the power can be insta-recharged in various ways), but that's nothing to the many, many, many ways that powers can deal damage. There really isn't a damage rating that a power has, there are just a number of effects, many of which are conditional in various ways. Writing code that decides what the damage rating of a power actually is for the purposes of proc rate would be challenging and would probably have a bunch of weird corner cases where the code's interpretation of "what damage does this power do" does not correspond to a reasonable person's intuition of "what damage does this power do." 2. We at some point have to grapple with the fact that there are 3 normal damage procs and 1 purple damage procs in Hold sets. The overwhelming majority of Hold powers do low or no damage. It is clearly by design for these hold procs to do damage, and it would be weird and bad if we made them into traps by making them have extremely low proc rates in most hold powers. Similarly for damage procs in immobilize and slow powers. 3. Also, what about non-damage procs? Surely nobody objects to people using Lockdown's hold in a Hold power, we don't want that to have a floored proc rate because the Hold in question does 5 points of damage. We could perhaps continue to have non-damage-procs be based on recharge time but damage procs be based on damage, but that creates yet more complexity in the proc system. Ultimately, I can't endorse this kind of solution (and I'm not sure I agree with the framing of the problem). But I think there is an important insight that the use of recharge time as a limiter on proc rate creates some real issues.

I don't think that CoD plays games like that (human-readable short-descriptions in game definitely do play those kind of games, but CoD works off the actual power definitions). For example, here's Concentration (the build up equivalent from Mental Manipulation): https://cod.uberguy.net/html/power.html?power=blaster_support.mental_manipulation.build_up&at=blaster It doesn't call out psi damage separate from other damage bonuses -- surely though Concentration does not fail to buff psi damage.