|
Post by johannhowitzer on Aug 4, 2020 0:58:13 GMT
Had an interesting conversation over a loot split today, in which people recommended I take and use the Vulpine Fane shield. The debate that followed made me think maybe I should go into some numbers on the forums, because it seems like there's some misinformation / placebo / taking for granted happening here. So it's probably good to give new players some accurate info so they don't shoot themselves in the foot.
How Armor Class Works
- AC maxes out its benefit at 21 higher than a creature's fully modified high-end attack bonus. This means against a creature with 100 AB, your best AC will be anything 121 or higher. At 120, the hit chance bottoms out, and at 121, you acquire full crit immunity, since threat rolls do not auto-succeed on a 20. Angus has 83 AB, so you need 104 if you're fighting him normally.
- Attack rolls are made per round from top to bottom. If Angus has five attacks at base 83 AB, he will likely get them in the order 83-83-78-73-68. The most important time to have good AC is when an enemy starts attacking you. Attacks of opportunity are also done at the highest number.
- Debuffs to enemy AB help, of course, but you must be able to count on those debuffs. Curse song, Prayer, Battletide, Bigby 5, etc. can all greatly improve your defenses against enemy attack rolls, but uncursed enemies from start of spawn, a very wide spawn, a sleepy Bard, or your own debuffed AC from an Illithid Temperer, a Falxugon, that trap by the Queen Spider... moral of the story, don't assume a perfect scenario. Angus, our example up to this point, can be tanked perfectly with only 88 AC, if you have a smart cleric and mage around. Or if you're, I dunno, a theurge or something, wink wink.
AC in the Hells
Now, of course nobody puts all their eggs in the AC basket, there are many other things to consider. In the case of the Vulpine Fane, it was a single extra level 8 and 9 spellslot, in exchange for a loss of 6 AC, from 142 to 136. Setting aside how bad this tradeoff is, let's consider just the two AC values. The argument made was that 136 is plenty, and you won't get hit if you just don't get flatfooted.
Hell mobs have around 116-120 AB, and paragon status can add up to 9 more. This means that the optimal AC, disregarding debuffs, is 150. An uncursed non-paragon with 118 needs an 18 to hit 136 AC, or a 20 to hit 142. That means taking one-third as many hits, right there, and a really common scenario. That's before we consider on-hit effects, like cornugon stun, swarm master eggs, etc. which are generally far worse than simple damage.
In a much worse scenario, say a superior Barbazu and a regular Falxugon, with the bard on the floor, suddenly 136 AC doesn't look so great. It becomes 133 AC with curse against 123 AB! That's a 55% chance to hit in the first round, or about 35-40% after conceal applies. Just a regular footsoldier raised to middle paragon status with one random present, not yet cursed, no special abilities, and you're eating every third hit. 142 AC would become 139 on curse, for 25% to hit or about 15% after conceal. That is less than half the hits taken on average.
The main takeaway from all this is that, while you can sacrifice a point or two here and there for big payoffs, don't underestimate the importance of AC, and don't sacrifice it for some tiny benefit just because you think the lower number will be good enough.
|
|
|
Post by FunkySwerve on Aug 4, 2020 4:07:10 GMT
- AC maxes out its benefit at 21 higher than a creature's fully modified high-end attack bonus. This means against a creature with 100 AB, your best AC will be anything 121 or higher. At 120, the hit chance bottoms out, and at 121, you acquire full crit immunity, since threat rolls do not auto-succeed on a 20. Angus has 83 AB, so you need 104 if you're fighting him normally. No. You're omitting debuffs to AC. I wouldn't get into flurries for this. That said, most added attacks are added at top ab, so your example is likely fine. As for the rest, I have no idea what you mean about the most important time to have good AC. I would say instead 'don't get flatfooted'. None of this makes sense. Each point of ac is another pip on a d20, for purposes of hitting, or 5%. But trying to translate that into damage is not as simple. Crit confirmation rolls, for example. There's much more in-depth analysis of a lot of this on the boards, if you hunt a while. We also ran a LOT of numbers when altering weapon crit ranges and multipliers. See here. Even those calcs involve some simplified assumptions. Again, you're only considering debuffs on mobs - don't forget those on you. ...area penalties. I agree with your conclusion. FWIW AC is only getting more important in Mechanus. Funky
|
|
|
Post by johannhowitzer on Aug 4, 2020 5:05:34 GMT
Which I got into.
I mentioned three examples in the text block you quoted right there.
Yes, 150 after penalties would be optimal for Hells. Notice the text "disregarding debuffs," once again in the text you quoted. I'd bust your chops more over these little slips, but then again, I've done this in the past myself, you're probably just pulling late hours or distracted or something, no big deal.
Yeah, I oversimplified, kind of on purpose just to make the overall point, probably should've just left that out, as I barely scrape into the math and it's still the most math in this post. Wasn't really necessary, bad call. I'll edit the post to avoid confusing people.
|
|
|
Post by simpetar on Aug 4, 2020 8:34:56 GMT
The argument made was that 136 is plenty,... Yes, for hells at least. ...and you won't get hit if you just don't get flatfooted. No. Hell mobs have around 116-120 AB, and paragon status can add up to 9 more. This means that the optimal AC, disregarding debuffs, is 150. Yet people do just fine even without AC in 150s. Aside from the fact that only handful builds can even reach that number and even fewer can do so without compromising their stats elsewhere... does that mean that optimal AC in limbo is 164, to avoid being hit by elite feral elf archers? Doubtful. In a much worse scenario, say a superior Barbazu and a regular Falxugon, with the bard on the floor, suddenly 136 AC doesn't look so great. It becomes 133 AC with curse against 123 AB! That's a 55% chance to hit in the first round, or about 35-40% after conceal applies. Just a regular footsoldier raised to middle paragon status with one random present, not yet cursed, no special abilities, and you're eating every third hit. 142 AC would become 139 on curse, for 25% to hit or about 15% after conceal. That is less than half the hits taken on average. Once again, characters have been through this exact situation and lived to tell the tale, yes, with 136 AC. I don't think anybody would disagree that higher AC is better, but survival does not boil down just to that one number. AC doesn't exist in vacuum. One of my own tanks completed limbo 2 hard with only 3 deaths and 129 AC. Hint, it was a staffy, case in point. The main takeaway from all this is that, while you can sacrifice a point or two here and there for big payoffs, don't underestimate the importance of AC,... Similarly, don't overestimate it ...and don't sacrifice it for some tiny benefit just because you think the lower number will be good enough. Except in the cases when you know it will be enough, I guess?
|
|
|
Post by johannhowitzer on Aug 4, 2020 11:18:35 GMT
So you've pretty much agreed with me on everything, then... The argument made was that 136 is plenty,... Yes, for hells at least. ...and you won't get hit if you just don't get flatfooted. No. That was a conjunction. 136 is plenty, but you will still get hit sometimes even if not flatfooted, so the argument I went on to make isn't that 136 is not enough, but that more will have a beneficial effect, and should not be ignored if available, in favor of a small benefit like one extra spellslot. Again, my argument wasn't that 150 is necessary, it was refuting the idea that 136 and 142 are functionally equivalent, by showing that up to 150, each additional point does something and should be considered when possible. "Optimal" here just means "provides maximum benefit," not "necessary for survival." I said as much one paragraph earlier: "Now, of course nobody puts all their eggs in the AC basket, there are many other things to consider." This post was purely a refutation of the notion that the difference from 136 to 142 AC is not zero, nor even negligible. And you hold up staffmaster as a counterexample, as if I didn't factor conceal into my math in that quote right above your words there.Yes, like once you hit say 90-95 AC against Angus and you have a black blade or another party member around, you know you can stand in melee since he'll go into whirlwind and you won't be taking very many hits and can just heal when needed. But this post wasn't about maxing only one stat, it was about the notion of leaving 6 AC on the table when the tradeoff is bad, under the logic that the 6 AC isn't doing anything. It was to show that the 6 AC does in fact do something and should be considered, not discarded out of hand when readily available. So yes, it seems like you agreed with me pretty much across the board here, all of your counterarguments seem directed at points I wasn't making. Overall what I'm taking away from both of your posts is that people have more or less the same ideas about AC I have, which is reassuring. I really just wanted to address this for newer players, so they're not assuming there's no point in ever going above a certain AC value, and potentially leaving extra defense on the table.
|
|
|
Post by tomaan on Aug 4, 2020 18:27:53 GMT
Not going to nitpick your individual points, but I think it's important to note that this only talks about direct attacks under ideal conditions - how often does that happen on HG? I think it's important to understand how that 136 AC is derived, how/where your play-style can leave you vulnerable and how you can address those vulnerabilities. For example, if you are a run-in-and-bash-em player, you will get caught flatfooted at some point, so you have to plan for/around it. And you also can't forget there are touch attacks in the server, and AC on those is calculated a lot differently. I guess the point I'm trying to make is that AC is really just a number; without some context, it's really hard to evaluate it. [EDITED]
|
|
|
Post by Enius the White on Aug 4, 2020 19:27:56 GMT
This is definitely useful information for newer players. I think Johannhowitzer's key point is that the comparatively large HG AC number is nowhere near linear in efficacy. Only the last 20-30 points usually really matter, run and situationally (buff/debuff) specific. In fact, generally speaking, on a level/gear appropriate run, even 2-4 points of AC can have an inordinately important impact on survival. The crit confirmation that is described is... critical.
The soak/conceal/crit reduction focus for defence that Simp touches on can, of course, also be solid, usually requiring strong and carefully selected gear in and approaching the end game.
And, clearly, there are myriad other survival considerations starting with disables, instakills, KD, etc, but as far as AC goes, the information provided is quite helpful.
|
|
|
Post by johannhowitzer on Aug 5, 2020 5:20:21 GMT
The argument "what if you don't need it" is a counter-argument you can make about anything in the game, really. Do you need breach immunity in first map of Dis? Well, not normally, but you might encounter a random. Crit immunity against the Zhedi in Ssithrak? Not if you brought a caster with small wall and plan ahead, or leave a golem alive to block for you. Need acid immunity for Deadpool? Not technically, you can get away with a couple energy buffers, heals, and a fast kill.
The main point of this post is to dispel the notion that the extra 6 AC is worthless, by showing some common situations where it can be very helpful. I think that did come across well enough, despite all the rabbit trails we ended up on.
|
|
|
Post by FunkySwerve on Aug 6, 2020 3:17:01 GMT
Which I got into. I mentioned three examples in the text block you quoted right there.Yes, 150 after penalties would be optimal for Hells. Notice the text "disregarding debuffs," once again in the text you quoted. I'd bust your chops more over these little slips, but then again, I've done this in the past myself, you're probably just pulling late hours or distracted or something, no big deal. Only one slip, and that's because I spent all of five minutes responding in an attempt to be helpful. Given the number of errors in your post, I would not be in a hurry to 'bust' anyone's chops to any extent. Funky
|
|
|
Post by Tru3Fals3 on Aug 6, 2020 9:36:24 GMT
The crit confirmation that is described is... critical. this terrible pun made skim reading the dumpster fire of replies well worth it, i thank you sir
|
|
|
Post by evildeath123 on Aug 7, 2020 22:26:38 GMT
At the risk of derailing the thread...
A part of understanding AC is also understanding how unimportant it is to staying alive.
Mitigating the damage when AC is bypassed is much more important. Whether your AC is 120 or 150, you’re going to get hit sometimes.
I don’t know of ANY build where AC alone is enough defense.
I used to run hells/abyss with an XDD with nowhere near ‘max’ AC.
Knowing damage types, damage effects, etc., was for me, much more important than AC in staying upright.
There are many things that make an XDD less than optimal as a front line tank, but having Elemental damage immunities all at 150%+ and the main exotics for the encounter at or near 100%+, and covering conf/sleep/stun/etc., are much more important than AC.
In my opinion, max AC is way down the list of what is ‘required’ to stay alive making the original point of the post, trading away AC for other ‘stuff’ from different gear set ups, much more biased to the properties other than AC.
Edit: should have read all the replies properly before posting... 🤭
|
|
|
Post by Paradoom on Aug 9, 2020 6:47:42 GMT
I think the general point of johannhowitzer for the AC is a good start for getting a grasp of it and shows the importance as what you should strive for. In server reality it is by far more complicated as the discussion shows and noone wants to have beginners fall into an AC only trap. *I like it*
For starters: you will have to cover absolute damage reductions for phsyiscal, elemental and exotic damage types on top of immunity. Conceal increases the survival alot and as tank ask for it. Maxed out parry also helps to reduce critical hit damage unless you are critical hit immune. Maxed out hide increases the conceal from druid´s and rangers camouflage granted conceal even beyond that of the conceal spell.
In my experience, as tank, all this does not help you very much to survive in the late game content, if you don't have a high enough AC. And 130+ (or rather midst 130s, that assumes level 60 double demi) is what you should go for. Yes yes there are exceptions...
|
|
|
Post by johannhowitzer on Aug 12, 2020 0:50:35 GMT
Good lord, Funky, you have an impressive knack for taking friendly banter as insulting. Yeah, I made one error in my attempt to not overwhelm newbies, which I removed on your criticism. And the rest I was just having some fun with, I figured you'd made the post in too much of a hurry, given how much of my post you skipped over. Take it easy, none of this is personal.
|
|
|
Post by FunkySwerve on Aug 12, 2020 15:46:13 GMT
Good lord, Funky, you have an impressive knack for taking friendly banter as insulting. Yeah, I made one error in my attempt to not overwhelm newbies, which I removed on your criticism. And the rest I was just having some fun with, I figured you'd made the post in too much of a hurry, given how much of my post you skipped over. Take it easy, none of this is personal. I didn't take it personally, but it was more obnoxious than friendly. I zipped through your post tossing out some helpful info, and was thanked by you griping about how I failed to notice that the thing you said at the outset, while still incorrect, was elaborated on four paragraphs later. Was I gravely wounded by your words? Nope. Annoyed by the log/speck dichotomy of it all? Definitely. Clearly there is a disconnect in expectations. I write for a living, and do it so it can be understood on first read. I have neither time nor patience to sift through a meandering and ultimately self-contradictory stream-of-consciousness post trying to figure out what the author may or may not have understood or meant to convey. Bottom line, if you want to convey or garner information effectively, you have some work to do. If not, hey, keep doing you. But don't expect me to sit quietly while you rattle off a half-dozen errors and then focus on the thing I overlooked in your post. I mean, seriously. It's not just annoying, but also self-defeating, assuming your goal was to glean useful insight on game mechanics. Funky
|
|
|
Post by johannhowitzer on Aug 13, 2020 10:40:32 GMT
Okay sure, whatever. I mean, you quoted me six times, and three of those times the text you quoted held exactly the information you were nitpicking. If you're going to "zip through" so hard you're making a post that contains half the info you think is missing, of course I'm going to have a little fun with it, if that's obnoxious to you, I guess that's too bad. My post seems to have had the outcome I was going for anyway, and your reply did help improve it. No hard feelings.
|
|