Optimal AC per level.

[Masakan]

So I just want to see if im clear on this.
AC is considered suboptimal because it is very difficult to raise to a point of value which is why people prefer miss chance and the like.
That being said for AC to be considered useful throughout all the levels.
You would generally have to raise it by 17+per character level. Meaning the Ideal AC to have at level 20 is 37.
And generally the only type of character capable of pulling this off is a Gish, because 9 times out of 10 they are gonna go abjurant champion/is dexterity focused(unless you are a self buffer in which case you will be focusing on buffing yourself to astronomical levels) and would fill in the gap by either fighting defensively constantly, using additional spells to boost it like dragonskin or stances that boost ac.

Anything else we could do?

[ranagrande]

Optimizing AC is fun.


It's pretty easy to get high AC even without items or magic.

Using the standard array for ability scores, a Dragonborn Gnome Undying Way Monk 2/Stoneblessed (Dwarf) 3/Dwarf Paragon 1/Forsaker 10/Deepwarden 2/Fist of the Forest 1/Dwarven Defender 1 with Vow of Poverty and Vow of Peace can get an AC of 66:

10 Base
12 Exalted
5 Deflection
14 Natural Armor
10 Constitution (replaces Dex)
10 Constitution
3 Wisdom
1 Dodge
1 Size

+10 more for 76 while in Defensive Stance.

Note that that involves using the standard array and starting with 11 Con. If you instead started with 18 Con, it would have an AC of 75, or 85 while in Defensive Stance.

[AvatarVecna]

The problem I generally have with optimizing AC is that, past a certain point, I've either neglected my Touch AC or my FF AC (or both), or I've got such a high AC that I can only be hit on a natural 20 anyway; getting your AC that high isn't exactly the easiest thing, but once you get there, going further is both easy and pointless. And resources that were going to be putting into pointlessly increasing your AC further can instead go into additional layers of defenses, like Damage Reduction, Fast Healing, Miss Chance, or the like.

What's the optimal AC per level? High enough that the most accurate monster in 3.5 (at about CR=ECL+5 or so) can't hit you except on a nat 20. How high is that? I dunno; pretty high, probably.

[eggynack]

There's probably some point of AC where the marginal value of picking up another point diminishes significantly, assuming constant cost of AC, but I don't think that's an especially useful way to think about the statistic. Different classes and builds have access to different things, and those things cost a wide variety of things. Thus, what I think you need to do is consider the most cost efficient AC source, and ask whether that's worth the cost, and if it is then you consider the next most cost efficient AC source, and so on until you hit a source that you consider to inefficient to pick up. Of course, different things cost resources that are fundamentally different, so you'll have to do some eyeballing to pick out the most cost efficient source at any given point, but it's a plausible process.

The end result is that some classes/builds have a high optimal AC, while others have a low optimal AC. So, for the example I tend to use, consider the druid. At the baseline, you have 9 or 10 AC, depending on dexterity, and maybe less if you optimized in a particular way. So, you go after the most cost efficient source of AC, wild shape, and if you're using a desmodu hunting bat form, then you have 20 AC right off the bat, gained in a fashion which took very little. Then you take the next most efficient source, luminous armor, and get up to 25 AC, or 29 against melee attacks. At this point, you're at something of a crossroads. You could leave things there, and stick with your perfectly fine 25/29 AC, or you could go a bit deeper with a monk's belt, greater luminous armor, and/or barkskin. None of these resources are as much of a no brainer as the first two, so they could either go on your build to provide significantly greater AC, or you could decide they're not worth it and stick with the also optimal number you already have.

The point is that few other classes will have the same sorts of choices associated with them, and perhaps none will have an identical cost/benefit analysis associated. So, for example, wizards also have luminous armor, but they might be less likely to use it because of a higher value from slots, less need for AC, and less access to the ability to heal the associated damage. Builds are inevitably going to have varying optimal AC, and that optimal AC will also vary greatly across level due to when the build can plausibly acquire various resources. None of that stuff I just said is useful for druids is going to be available at first level, after all, and a perfectly optimal druid build could be running an AC of 10, with a -2 from particularly bad dexterity and a +2 from armor. The AC graph is far from clean as a result, and that remains true even in some less extreme cases.

[Masakan]

Well heres how i did it and I honestly found this to be the only real practical way to do this.
So as I mentioned before Most gishes go abjurant champion, Shield will be a staple spell for you.
You will also be fighting Defencively, +2 to AC, +1 from tumble skill past 5 and 1 from the cautious trait. You would also want to get deadly defense just to maximize the benefit. So for the cost of 1 feat and some skill points, you effectively break even between AC gain from Attack roll loss, and get an extra 1d6 to your hit die. Not a bad trade.

Next you would get a Gloves of Dexterity+2 because you are going to get a mithril shirt no questions asked, Gain benefits from either Stance of Clarity(for Bosses) and Pearl of Black Doubt(for Mooks)
And Finally The Shield Bonus you get with shield combined with abjurant champion which basically gives you a +9 shield bonus that works against touch attack

So for Recap.
Base AC:10
Mithril Shirt:4
Dexterity Bonus:6
Fighting Defencively:4
Stance of Clarity:2(Bosses Only)
And Shield Bonus:9
For a total of 35 AC By Level 11.
You would effectively have tank status for little to no cost to yourself.
And this would be your first line of defense.

[LTwerewolf]

I typically use giants as the baseline for why ac isn't great. At level 11 against a cloud giant, you need 42 ac to only be hit on a 20 (and that's only if they're not charging or no magic is involved to increase their to hit). It's possible to get that ac by level 11, but in order to do so you're neglecting pretty much everything else, making them unlikely to actually attack you.

[Masakan]

I typically use giants as the baseline for why ac isn't great. At level 11 against a cloud giant, you need 42 ac to only be hit on a 20 (and that's only if they're not charging or no magic is involved to increase their to hit). It's possible to get that ac by level 11, but in order to do so you're neglecting pretty much everything else, making them unlikely to actually attack you.

Which is why I say gishes are the only ones who can pull ac builds off effectively. Oh you wanna ignore me? Ok "Lob a shocking Spark into their backside"
Besides, I think It's unfair to use Any sort of Giant as an example as anyone who tries to take a giant head deserves to be a big splat on the floor.

[LTwerewolf]

Giants for me tend to be around the medium attack bonuses at any given level, not the top. There are plenty of things that have an easier time hitting than giants do, especially once you get magic involved. with your example, you're still not reaching the unhittable range, and need more investment to do so. Magic vestment brings you up to 37. You need further investment to get there for my baseline.

[eggynack]

Well heres how i did it and I honestly found this to be the only real practical way to do this.
Given that simply adding the monk's belt to my cited simple AC druid would trivially take it to around your given 35 AC, and could do so at a lower level and without fighting defensively, I find this claim rather doubtful. Wouldn't be difficult to do better either.

[Masakan]

Given that simply adding the monk's belt to my cited simple AC druid would trivially take it to around your given 35 AC, and could do so at a lower level and without fighting defensively, I find this claim rather doubtful. Wouldn't be difficult to do better either.
Ok we get it druids are broken as ****, you don't have to shove it down everyone's throats every time you try to make a point.

[eggynack]

Ok we get it druids are broken as ****, you don't have to shove it down everyone's throats every time you try to make a point.
The point is that your claim is an incorrect one. You seem to think that your way is the only practical way, and may argument is that it is not. I simply tend to use druids as the conduit to that point because I know them well and can make more well rounded and defended points using them than I would be able to through other classes. It's not like this stuff really depends on theoretical optimization as you seem to think. It's just kinda standard stuff that does its job well. I mean, we are optimizing AC here. Only so broken you can get.

Edit: Also, you really don't give yourself much room to stand on using magic for this. Seems hypocritical to use it in your construction and admonish me when I do the same.

[Masakan]

The point is that your claim is an incorrect one. You seem to think that your way is the only practical way, and may argument is that it is not. I simply tend to use druids as the conduit to that point because I know them well and can make more well rounded and defended points using them than I would be able to through other classes. It's not like this stuff really depends on theoretical optimization as you seem to think. It's just kinda standard stuff that does its job well. I mean, we are optimizing AC here. Only so broken you can get.

Edit: Also, you really don't give yourself much room to stand on using magic for this. Seems hypocritical to use it in your construction and admonish me when I do the same.

Ok first of all this is the kind of thing you need to ask clarity about, the one i suggested was the only one that I could come up with. I'm almost certain there are other ways to go about it, but this is the one that works for me.

[eggynack]

Ok first of all this is the kind of thing you need to ask clarity about, the one i suggested was the only one that I could come up with. I'm almost certain there are other ways to go about it, but this is the one that works for me.
You said you thought it was the only real practical way. Not, incidentally, that it's the way you thought was best for your particular desires but the best one you could find. Now, you presumably don't think it's the only practical way to accomplish this. That's pretty much the point in its entirety, I think. Also, I did post that thing about trivially getting 25 AC using only things available by 5th level, with better numbers also available with a little work, prior to your restating that you knew of no way to do this otherwise. Even if you were correct that you personally had no method the first time, I think it reasonable to expect that that would stop being true the second time.

[ILM]

So I just want to see if im clear on this.
AC is considered suboptimal because it is very difficult to raise to a point of value which is why people prefer miss chance and the like.
That being said for AC to be considered useful throughout all the levels.
You would generally have to raise it by 17+per character level. Meaning the Ideal AC to have at level 20 is 37.
And generally the only type of character capable of pulling this off is a Gish, because 9 times out of 10 they are gonna go abjurant champion/is dexterity focused(unless you are a self buffer in which case you will be focusing on buffing yourself to astronomical levels) and would fill in the gap by either fighting defensively constantly, using additional spells to boost it like dragonskin or stances that boost ac.

Anything else we could do?
I just want to point out your premise is off. If you take all the monsters in the SRD, the progression of their attack bonus means that you need an AC of 2xCR+9 on average to have a reasonable fighting change (i.e. they need a 10+ on their roll to hit you). It's actually remarkably consistent. 2xCR+9 means that at level 20 you'd need AC 49 to be competitive. Round it up to 2xCR+10 if it's easier to remember. Or if you prefer: Monster Attack Bonus = 2xCR.

[Masakan]

I just want to point out your premise is off. If you take all the monsters in the SRD, the progression of their attack bonus means that you need an AC of 2xCR+9 on average to have a reasonable fighting change (i.e. they need a 10+ on their roll to hit you). It's actually remarkably consistent. 2xCR+9 means that at level 20 you'd need AC 49 to be competitive. Round it up to 2xCR+10 if it's easier to remember. Or if you prefer: Monster Attack Bonus = 2xCR.

Huh...I was told that any higher than 40 was overkill.

[ILM]

Huh...I was told that any higher than 40 was overkill.
+40's the average attack bonus of CR 20 monsters in core. If you have AC 40, they don't even need to roll to hit you (on average; in practice, a Balor does need a 7+).

I think the point is that raising AC past 40 generally requires a decent investment of money and/or class features and/or spell slots, and that's the part that's not worth it because at CR 20 very few monsters' shtick is 'hitting PCs with a sharp object'. That Balor may have a 1 in 3 chance to miss with his sword, but I'd be more concerned about his at-will SLAs and summoning ability at this point.

[Crake]

Well heres how i did it and I honestly found this to be the only real practical way to do this.
So as I mentioned before Most gishes go abjurant champion, Shield will be a staple spell for you.
You will also be fighting Defencively, +2 to AC, +1 from tumble skill past 5 and 1 from the cautious trait. You would also want to get deadly defense just to maximize the benefit. So for the cost of 1 feat and some skill points, you effectively break even between AC gain from Attack roll loss, and get an extra 1d6 to your hit die. Not a bad trade.

Next you would get a Gloves of Dexterity+2 because you are going to get a mithril shirt no questions asked, Gain benefits from either Stance of Clarity(for Bosses) and Pearl of Black Doubt(for Mooks)
And Finally The Shield Bonus you get with shield combined with abjurant champion which basically gives you a +9 shield bonus that works against touch attack

So for Recap.
Base AC:10
Mithril Shirt:4
Dexterity Bonus:6
Fighting Defencively:4
Stance of Clarity:2(Bosses Only)
And Shield Bonus:9
For a total of 35 AC By Level 11.
You would effectively have tank status for little to no cost to yourself.
And this would be your first line of defense.

Throw away that mithril chain shirt and just cast greater luminous armor. As an abjuration spell that grants an armor bonus, abjurant champion will increase it, when combined with shield, those two spells alone give you 32AC.

[sleepyphoenixx]

Every caster can raise his AC to a decent level with a bit of effort.

Wisdom based casters go for a Monk's Belt with (Greater) Luminous Armor as a baseline, which is already pretty decent. Then you can add things like Barkskin, Shield of Faith, Halo of Sand, Sirine's Grace and Greater Magic Vestment, which adds even more. Druids get even more out of the Monk's Belt by casting Owl's Insight, but that hardly means it's useless for clerics. Druids also get even more natural AC from Wild Shape and a boost on Dex depending on form.

For arcane casters there's also a pretty wide selection of various bonus types, but the biggest boost probably comes from the Polymorph & Scintillating Scales combo. That can easily add +15-20 deflection AC all on its own, depending on the form. Even Alter Self has some forms with pretty impressive natural armor, like the Crucian or Tren for 8 nat AC. Even core-only there's the Lizardfolk for +5. If you're a non-humanoid it opens up even better forms.
Then you add Shield, Greater Mage Armor and Haste and you have pretty good AC even without Abjurant Champion.

Really, the only people who have trouble getting decent AC are non-casters, because AC from items gets expensive really fast, to the point where you can barely keep up even if you spend most of your WBL on it.
Of course in actual play it's expected that the casters buff their party members, which isn't usually calculated in when someone builds their character.
Without casters you'll have a lot of trouble though, which is a problem that miss chance doesn't have. And if you're not a melee combatant miss chance is usually enough, and it's significantly cheaper.

[ben-zayb]

At high levels, Gr. Magic Weapon + Defending Armor Spikes / Spiked Gauntlet / Hidden Blade / Elvencraft Bow is pretty cheesy, and people have been arguing that they can stack with each other, which is cheesier.

[Killer Angel]

That being said for AC to be considered useful throughout all the levels.
You would generally have to raise it by 17+per character level. Meaning the Ideal AC to have at level 20 is 37.


Leaving aside the current discussion, I'd say it's not exactly linear.
AC 18 at first level is sufficiently good. AC 37 at level 20 is pathetic.

[Hamste]

Optimizing AC is fun.

I have to ask why don't you just start dwarf? Surely you can get more ac if you switch out those three levels of stone blessed or in the very least it would make it better to play at those levels. Also what is the point of dwarf paragon? It seems to add literally nothing to the build. I ask because the way you have it set up levels 3-6 suck getting effectively no class features and both your 3rd and 6th feat are taken up by boring feats.

[Uncle Pine]

Getting your AC past 40 is ridiculously easy, the main problem being that people usually frown upon and/or disregard as too cheesy the most efficient methods.
For example, a +1 defending weapon costs about 8,300 gp (or 4,150 gp if you're crafting it yourself). Casting Greater Magic Weapon on it will net you +5 AC for a mere 1,660 gp/point of AC (or 830 gp/point of AC if you're crafting the weapon yourself).
Now buy two spiked guantlets, two boot blades, two elbow blades, two knee blades, two sleeve blades, a spiked armor and enchant all of them as above (total cost 91,300 gp, or 45,650 gp if you're doing everything by yourself). Then buy a lesser metamagic rod of chaining for 14,000 gp (you can probably split this with the rest of the party as it can be used to everyone's benefit, but 14,000 gp aren't that much anyway). Finally, cast (or have your party's Wizard cast) Greater Magic Weapon once using the rod.
Suddenly you have a base AC (normal, touch and flat-footed) of 5*11+10=65, before adding things like Dex, nat armor, armor bonus, etc.
Also note that this is something everyone, even your Big Stupid Fighter, can pull off. Moreover, for all the DM out there, note that having a high AC doesn't make someone unkillable: most of the nastier stuff (read: spells) don't particularly care about AC and if your player didn't invest in a couple of spellblades, throwing a chained dispel at him could make for a funny encounter.

[sleepyphoenixx]

Getting your AC past 40 is ridiculously easy, the main problem being that people usually frown upon and/or disregard as too cheesy the most efficient methods.
For example, a +1 defending weapon costs about 8,300 gp (or 4,150 gp if you're crafting it yourself). Casting Greater Magic Weapon on it will net you +5 AC for a mere 1,660 gp/point of AC (or 830 gp/point of AC if you're crafting the weapon yourself).
Now buy two spiked guantlets, two boot blades, two elbow blades, two knee blades, two sleeve blades, a spiked armor and enchant all of them as above (total cost 18,260 gp, or 9,130 gp if you're doing everything by yourself). Then buy a lesser metamagic rod of chaining for 14,000 gp (you can probably split this with the rest of the party as it can be used to everyone's benefit, but 14,000 gp aren't that much anyway). Finally, cast (or have your party's Wizard cast) Greater Magic Weapon once using the rod.
Suddenly you have a base AC (normal, touch and flat-footed) of 5*11+10=65, before adding things like Dex, nat armor, armor bonus, etc.
Also note that this is something everyone, even your Big Stupid Fighter, can easily pull off with little investment. Moreover, for all the DM out there, note that having a high AC doesn't make someone unkillable: most of the nastier stuff (read: spells) don't particularly care about AC and if your player didn't invest in a couple of spellblades, throwing a chained dispel at him could make for a funny encounter.

That's all assuming that your DM doesn't count the Defending enhancement as one source no matter how many you have - and boni from the same source don't stack, even if they're untyped. And that's not only a valid interpretation of the RAW, it's also the reasonable one, so it's likely what most DMs will use.

You've also messed up your math because 11 +2 equivalent weapons run about 90,000 gp, not 18,300, which is a good chunk of WBL even at 20.

[Necroticplague]

The problem with AC isn't that it's hard to boost (I should know, ended up with an AC in the lower 300's before). It's that it only covers some of the least harmful things out there. Normally, AC is just protecting you against losing HP. There are some exceptions with rays or touch attacks, but those are the relative minority. Meanwhile, AC doesn't do much to protect you from the nastiest things in the game: debilitating debuffs and SoLs. Raising your WILL and FORT, is thus of much greater import than your AC, as is acquiring immunities to such abilities.


That's all assuming that your DM doesn't count the Defending enhancement as one source no matter how many you have - and boni from the same source don't stack, even if they're untyped. And that's not only a valid interpretation of the RAW, it's also the reasonable one, so it's likely what most DMs will use.
Normally, you'd be right, but Defending has a specific bit of text that allows for this.

A defending weapon allows the wielder to transfer some or all of the sword’s enhancement bonus to his AC as a bonus that stacks with all others. As a free action, the wielder chooses how to allocate the weapon’s enhancement bonus at the start of his turn before using the weapon, and the effect to AC lasts until his next turn.
Bolded for emphasis. Defending weapons stack with all other bonuses to AC. Including other untyped bonuses from Defending weapons.

[ben-zayb]

I have to ask why don't you just start dwarf? Surely you can get more ac if you switch out those three levels of stone blessed or in the very least it would make it better to play at those levels. Also what is the point of dwarf paragon? It seems to add literally nothing to the build. I ask because the way you have it set up levels 3-6 suck getting effectively no class features and both your 3rd and 6th feat are taken up by boring feats.+1 size bonus, and +Dex if Whisper Gnome is used. Dwarf Paragon is presumably for class skills or max ranks in Survival and Knowledge (Dungeoneering), which are needed for Deepwarden.

[Uncle Pine]

You've also messed up your math because 11 +2 equivalent weapons run about 90,000 gp, not 18,300, which is a good chunk of WBL even at 20.

You're right! I accidentally multiplied the cost/point of AC instead of the total cost/item there. It's been corrected. :smallredface:
Necroticplague already answered to your other point.

[Masakan]

Throw away that mithril chain shirt and just cast greater luminous armor. As an abjuration spell that grants an armor bonus, abjurant champion will increase it, when combined with shield, those two spells alone give you 32AC.

Sorcerer, Not wasting my time with arcane preperation

[sleepyphoenixx]

Normally, you'd be right, but Defending has a specific bit of text that allows for this.

Bolded for emphasis. Defending weapons stack with all other bonuses to AC. Including other untyped bonuses from Defending weapons.

I really don't see how "stacks with all other sources" gets around the "same sources don't stack" rule. Unless my english is more lacking than i thought something is either the same source or an "other" source, not both.

The rules text on the SRD only mentions "the same spell cast twice in succession" so there's maybe a bit of wiggle room if you want to ruleslawyer it, but i'd argue that applies equally to "the same weapon enhancement (Defending), no matter how many weapons with it you carry".

If your DM rules otherwise it's certainly a method to get higher AC, but your character will probably look very silly.

[Necroticplague]

I really don't see how "stacks with all other sources" gets around the "same sources don't stack" rule. Unless my english is more lacking than i thought something is either the same source or an "other" source, not both.

The rules text on the SRD only mentions "the same spell cast twice in succession" so there's maybe a bit of wiggle room if you want to ruleslawyer it, but i'd argue that applies equally to "the same weapon enhancement (Defending), no matter how many weapons with it you carry".

If your DM rules otherwise it's certainly a method to get higher AC, but your character will probably look very silly.
It's not "stacks with all other sources." It's "stacks with all other bonuses" (the object of the clause 'all others' is 'bonus', not 'source'). Even if two defending weapons are the same source (itself arguable), they are still two bonuses to AC, which the Defending property says can stack.

[Segev]

I really don't see how "stacks with all other sources" gets around the "same sources don't stack" rule. Unless my english is more lacking than i thought something is either the same source or an "other" source, not both.

The rules text on the SRD only mentions "the same spell cast twice in succession" so there's maybe a bit of wiggle room if you want to ruleslawyer it, but i'd argue that applies equally to "the same weapon enhancement (Defending), no matter how many weapons with it you carry".

If your DM rules otherwise it's certainly a method to get higher AC, but your character will probably look very silly.

This Defending Sword is a source of +5 AC that stacks with all other sources.
This Defending Spiked Gauntlet is a source of +5 AC that stacks with all other sources.

Each is a separate source of +5 AC. Each stacks with all other sources. Each therefore stacks with the other.

[Deadline]

AC's primary purpose (at least at later levels) seems to be discouraging Power Attack use by monsters, and limiting the number of iterative attacks they can successfully land. In other words, without significant resource expenditure, their primary attack (or most of their natural attacks) are going to hit you. So at higher levels, other mitigation strategies come into play. That's why miss chance, Mirror Image, and Abrupt Jaunt are so highly lauded. It changes the combat from "the big monster is going to hit me at least once for a lot of damage" to "the big monster probably isn't going to hit me at all". That's a pretty big deal.

We've had a few hyper optimized AC threads in the past, and the numbers you can get to are definitely impressive, but ultimately it suffers from "now the monsters busy themselves will killing your allies". It doesn't even matter if you can do moderate-good damage, if the monster is unlikely to hit your AC, then they'll do one of the following:


Target your Touch AC
Target your allies


And for most folks who are optimizing AC (i.e. the tanks), either of those effectively means they've wasted their resources.

Edit - ILM has a reasonable formula for AC, although it breaks down a bit in high-level rocket tag.

[Hamste]

+1 size bonus, and +Dex if Whisper Gnome is used. Dwarf Paragon is presumably for class skills or max ranks in Survival and Knowledge (Dungeoneering), which are needed for Deepwarden.

They later replace dexterity though so plus to dexterity isn't the reason and size bonus is only plus 1, though they do end getting con to ac multiple times which makes stone blessed make sense later but it hurts when you go 3 levels in a class that doesn't do much until later.

By level 16 (the first level they take deep warden) they have 48 plus int mod*19 skill points.

They need 5 climb, 5 heal, 5 jump, 5 k(dungeoneering), 5 survival.

Stone blessed requires 2 appraise and 5 craft by level 2 to enter at 3.

Fist of the forest needs 4 handle animal (but got 6 skill points from stone blessed)

Monk puts their 20 in craft (5), climb (5), jump (5), cross class appraise (2) and cross class k(dungeoneering) (.5).
Forsaker puts their 20 in heal (5), cross class survival (5), cross class k(dungeoneering bringing it to (2.5), handle animal (1)

That leaves 2.5 more k (dungeoneering) or 5 skill points over 4 levels worst case scenario. This assumes a neutral intelligence but seeing only wisdom and con are important stats for your defence that I see, that shouldn't be too hard.

[Masakan]

AC's primary purpose (at least at later levels) seems to be discouraging Power Attack use by monsters, and limiting the number of iterative attacks they can successfully land. In other words, without significant resource expenditure, their primary attack (or most of their natural attacks) are going to hit you. So at higher levels, other mitigation strategies come into play. That's why miss chance, Mirror Image, and Abrupt Jaunt are so highly lauded. It changes the combat from "the big monster is going to hit me at least once for a lot of damage" to "the big monster probably isn't going to hit me at all". That's a pretty big deal.

We've had a few hyper optimized AC threads in the past, and the numbers you can get to are definitely impressive, but ultimately it suffers from "now the monsters busy themselves will killing your allies". It doesn't even matter if you can do moderate-good damage, if the monster is unlikely to hit your AC, then they'll do one of the following:


Target your Touch AC
Target your allies


And for most folks who are optimizing AC (i.e. the tanks), either of those effectively means they've wasted their resources.

Edit - ILM has a reasonable formula for AC, although it breaks down a bit in high-level rocket tag.

Hence why I think 39-40 is a good stopping point. Sure your AC is high, but low enough to make them think hey maybe he's worth going after, and you can get it at an almost minimal cost as a gish.

[ben-zayb]

They later replace dexterity though so plus to dexterity isn't the reason and size bonus is only plus 1, though they do end getting con to ac multiple times which makes stone blessed make sense later but it hurts when you go 3 levels in a class that doesn't do much until later.

By level 16 (the first level they take deep warden) they have 48 plus int mod*19 skill points.

They need 5 climb, 5 heal, 5 jump, 5 k(dungeoneering), 5 survival.

Stone blessed requires 2 appraise and 5 craft by level 2 to enter at 3.

Fist of the forest needs 4 handle animal (but got 6 skill points from stone blessed)

Monk puts their 20 in craft (5), climb (5), jump (5), cross class appraise (2) and cross class k(dungeoneering) (.5).
Forsaker puts their 20 in heal (5), cross class survival (5), cross class k(dungeoneering bringing it to (2.5), handle animal (1)

That leaves 2.5 more k (dungeoneering) or 5 skill points over 4 levels worst case scenario. This assumes a neutral intelligence but seeing only wisdom and con are important stats for your defence that I see, that shouldn't be too hard.

Yes, and the example was supposed to minmax, so being MAD for a 12+ Int and getting +1 size bonus to attack/AC are expected. Additionally, those extra skill points
you save can be put to skills that you will actually use, such as slightly better Spot/Listen. You also need Str as a tertiary stat if you want creatures to not ignore you as a threat, unless you plan to be just a giant wall of suck.

[Deadline]

Hence why I think 39-40 is a good stopping point. Sure your AC is high, but low enough to make them think hey maybe he's worth going after, and you can get it at an almost minimal cost as a gish.

Yep, although the problem there being that you are going to be getting hit. Which is ok if you are trying to Tank (albeit 3.5 is sorely lacking in modeling the whole Tank role). If you are relatively squishy and don't want to try and Tank, you'll probably want to shoot for a secondary layer of defenses (like a miss chance) to discourage attacks against you at all.

But yes, casters are better at defenses than mundanes. It's yet another role where they get to play "anything you can do I can do better". :smallsmile:

[Masakan]

Yep, although the problem there being that you are going to be getting hit. Which is ok if you are trying to Tank (albeit 3.5 is sorely lacking in modeling the whole Tank role). If you are relatively squishy and don't want to try and Tank, you'll probably want to shoot for a secondary layer of defenses (like a miss chance) to discourage attacks against you at all.

But yes, casters are better at defenses than mundanes. It's yet another role where they get to play "anything you can do I can do better". :smallsmile:

Why do you think I feel sorcerers are better at being gishes than wizards, access to spells wizards just do not have. Including say it with me now "Wings of Cover" The only down side is that you have to use it before an attack hits, but as long as it's not an AOE it effectively blocks ANYTHING.

[ComaVision]

Why do you think I feel sorcerers are better at being gishes than wizards, access to spells wizards just do not have. Including say it with me now "Wings of Cover" The only down side is that you have to use it before an attack hits, but as long as it's not an AOE it effectively blocks ANYTHING.

Wings of Cover also decreases the value of AC.

[Elkad]

I trust the DM to use appropriate monsters for the group and use a more flexible rule myself.

"within 5 points of the rest of the party". If I'm on the bottom, I'll try to add some miss chance on.

This avoids the "ignore the tank, kill the squishies" problem neatly.

[Masakan]

Wings of Cover also decreases the value of AC.

Except your not gonna just spam it, that's stupid.

[Red Fel]

I'd just like to point out that a Wizard - not a Sorcerer - who wants armor-based AC can pursue the Runesmith PrC, wear heavy armor, and ignore ASF from armor by casting from runes. Although the class says it can be taken by a Wiz or Sorc - and it technically can - runes can only be used for prepared spells.

So, yeah. Wiz has gishing options.

Going back to my corner now.

[Necroticplague]

I'd just like to point out that a Wizard - not a Sorcerer - who wants armor-based AC can pursue the Runesmith PrC, wear heavy armor, and ignore ASF from armor by casting from runes. Although the class says it can be taken by a Wiz or Sorc - and it technically can - runes can only be used for prepared spells.

So, yeah. Wiz has gishing options.

Going back to my corner now.

Sorcerors can pick up Arcane Preparation to benefit (also giving them access to the Luminous Armor line).

[Deadline]

Why do you think I feel sorcerers are better at being gishes than wizards, access to spells wizards just do not have. Including say it with me now "Wings of Cover" The only down side is that you have to use it before an attack hits, but as long as it's not an AOE it effectively blocks ANYTHING.

I'm aware of your opinion, and you are most welcome to it, but I disagree to an extent. At 20th level, the difference between a sorcerer and wizard is largely irrelevant, they operate pretty much the same (for huge lulz on that topic, look up the various threads by LordDrako, before he was banned). At lower levels, the Wizard gets at least as many if not more options than the Sorcerer does (Abrupt Jaunt at level 1 for attack avoidance). I'm guessing you are aware of most of these, and just really like Sorcerers. And that's cool, every class needs to feel loved. I'm a Dragonfire Adept/Warlock fan myself :smallbiggrin:

[Masakan]

I'm aware of your opinion, and you are most welcome to it, but I disagree to an extent. At 20th level, the difference between a sorcerer and wizard is largely irrelevant, they operate pretty much the same (for huge lulz on that topic, look up the various threads by LordDrako, before he was banned). At lower levels, the Wizard gets at least as many if not more options than the Sorcerer does (Abrupt Jaunt at level 1 for attack avoidance). I'm guessing you are aware of most of these, and just really like Sorcerers. And that's cool, every class needs to feel loved. I'm a Dragonfire Adept/Warlock fan myself :smallbiggrin:

I just like breaking the steryotype and becoming a more sociable spellcaster

[Deadline]

I just like breaking the steryotype and becoming a more sociable spellcaster

Be more of an Iconoclast than that! Be a WIZARD and be a sociable spellcaster! Break the mold with the mold! :smallwink:

[eggynack]

I just like breaking the steryotype and becoming a more sociable spellcaster
I'm not entirely sure how that's pertinent to the issue of wizards versus sorcerers for gishing purposes. There are ways that wizards are better than sorcerers at gishing, and there are ways the inverse is true. Notably, it seems that boosting AC is one of the former sorts of ways, due to sanctified spells and/or that prestige class. There's no need to introduce stuff external to the problem.

[Masakan]

I'm not entirely sure how that's pertinent to the issue of wizards versus sorcerers for gishing purposes. There are ways that wizards are better than sorcerers at gishing, and there are ways the inverse is true. Notably, it seems that boosting AC is one of the former sorts of ways, due to sanctified spells and/or that prestige class. There's no need to introduce stuff external to the problem.
Can't I make side comments?

[bekeleven]

Eggy predictably has druid covered, I'll just throw out the fact that Master of Many Forms has access to arbitrarily large natural armor bonuses. The War Troll at level 12 being a perfect example of a form that has high AC incidental to the fact that it's killing everything and takes nonlethal damage from basically all attacks. Sample AC for a war troll MoMF who's trying at all: 10 + 14 Natural + 4 Dexterity + 6 Armor + whatever else you want to throw in = 34+. With a bit of optimization you can push this past 40, but again, your regeneration is bypassed only by acid so why bother?Astute observers may note that a master of many forms with 1 form known is better than most optimized fighter builds.
Chameleon, my other one true love, is a lot harder to use before you've finished the class at level 15. At level 12, for instance, a mid-op chameleon is also sitting on 34 AC (Factotum 4/Monk 1/Chameleon 7 has +8 Armor, +4 Shield, +9 Int, and +3 Dex) with the option to spend an inspiration for 43 AC against one attack. Optimizing further can increase this AC to 40+.

At least, until the chameleon remembers he has access to shapechanging magics and turns into a war troll too. Then you get the best of both worlds for an encounter or two, an AC somewhere along the lines of 10 + 15 natural + 8 Armor + 4 Dexterity + 4 Shield + 10 Int + 1 Deflection = 52 AC, or 62 with an inspiration point.

I haven't calculated the theoretical maximum here, these are conservative estimates, the main limit being wealth by level or buff time (I'm assuming +1 natural armor, +1 deflection, +2 Dex, and +4 int).

[sleepyphoenixx]

Getting 100 AC or more may look impressive, but it's not doing much that AC 60 isn't if you're pre-epic. The same applies to lower levels, scaled appropiately.

For optimal AC values i generally just look at a dragon of the appropiate CR.
They get full BAB and high Str, with high HD for their CR, so they make a good (and quick) benchmark for the upper end of what to expect.

Once you're at a level where a dragon won't hit you 80-90% of the time investing more into AC is just a waste of resources that can be better spend elsewhere.
Even more efficient is getting less than that (around 50% is what i aim for) because that's the upper limit of what you'll face, so a part of your high AC is effectively wasted against most enemies and you'd be better off spending the resources on miss chance.

[eggynack]

Can't I make side comments?
I guess, but I'd figure you'd want to actually rebut the claims made, lest you risk your position of sorcerer gish superiority. Side comments are fine, but they tend to go best when paired with normal comments.

[Masakan]

I guess, but I'd figure you'd want to actually rebut the claims made, lest you risk your position of sorcerer gish superiority. Side comments are fine, but they tend to go best when paired with normal comments.
Ive said this before and I will keep saying it until i drill it into your head, never assume I speak in absolutes I never do that.
I Like sorcerer gishes yes I feel they would work very well since gishes tend to fall into the catagory of casting the same few spells anyway.
But I know you can get away with Wizard gishes, cleric gishes, Psionic Gishes, Bard Gishes if you build right. Hell you can do a warlock gish and it would work.
I do feel however that the spontaneity of said classes are more appropriate than prepared casters. But any casting class.....can be used for a gish build you just gotta know....what the **** you doing.

[AvatarVecna]

Ive said this before and I will keep saying it until i drill it into your head, never assume I speak in absolutes I never do that.

Hate to be a nitpicker...

[Masakan]

Hate to be a nitpicker...
God damn it I suck at wording ****!

[eggynack]

Ive said this before and I will keep saying it until i drill it into your head, never assume I speak in absolutes I never do that.
I Like sorcerer gishes yes I feel they would work very well since gishes tend to fall into the catagory of casting the same few spells anyway.
But I know you can get away with Wizard gishes, cleric gishes, Psionic Gishes, Bard Gishes if you build right. Hell you can do a warlock gish and it would work.
I do feel however that the spontaneity of said classes are more appropriate than prepared casters. But any casting class.....can be used for a gish build you just gotta know....what the **** you doing.
As Avatarvecna implied, you speak in absolutes all the time. For example, "Which is why I say gishes are the only ones who can pull ac builds off effectively." Moreover, whether you were staking out an absolute stance or not, you were staking out a stance when you said, "Why do you think I feel sorcerers are better at being gishes than wizards, access to spells wizards just do not have," which, assuming I'm reading your strange grammar correctly, means that you do think that sorcerers are better at gishing. Not your preference, but better. It's a claim whether you like it or not, and one that I'd think you'd want to defend, or failing that, one I'd think you'd want to cede.

[AvatarVecna]

God damn it I suck at wording ****!

It helps to get into a habit of generalization around here: it's always a fair assumption some ******* is gonna climb out of the woodwork every time you declare something "always" or "never" just to correct you. I don't mean anything personal by it, but reading that sentence made me laugh out loud, and I just couldn't let it go.

So that I actually contribute something useful to the thread, let's just put it this way: at every ECL, there's a certain point where pumping AC higher isn't worth the resources invested into it to pump it higher, and those resources will be better spent elsewhere. This point (where AC is as high as it needs to be) can be reached in a number of ways by a number of builds. If I was to put a formula to it, I'd say something like AC=(ECLx2)+20, with a decent split between FF and Touch AC the whole way. Past that point, raising AC higher is probably pretty pointless for somebody of your ECL, but I can't say I've actually done the math on this, it's just what I feel would be the most a person could really want at any particular level.

[ZamielVanWeber]

One of the biggest upsides of arcane gishing is the ability to defend yourself against attack without having to resort to AC. (Greater) Mirror Image is a fantastic defense, as are Greater Blink and Greater Invisibility. All of these have the upside of simply negating attacks: effects such as True Strike need to know what square you are in for Greater Invisibility and Greater Mirror Image and Greater Blink aren't concealment, so True Strike does not work against them either.

[ShneekeyTheLost]

The problem here is that it is tough to scale AC to the same degree that you BAB. It's pretty trivial to have a BAB of 50+ around level 10ish. It's decidedly not trivial to have an AC of 70ish by that point. Which brings up the problem...

AC, to have the effect you are describing, needs to be 19 higher than the BAB an offensive character would need. So not only are there fewer options to increase AC than BAB, but you need it significantly higher as well.

Furthermore, there's plenty of ways to simply bypass AC as a defense, such as Save or Lose effects, touch attacks, or just plain 'no save, no attack, just die now' effects.

AC only works on beatsticks, which are the least dangerous of the various threats you will have to deal with in your career as an adventurer. Damage sucks as a method of defeating your opponents unless you deal so much damage that you can reliably one-shot kill anything you swing at because if you don't actually kill your opponent, the effect you have had on his ability to return the favor is precisely zero. Save or Lose effects are similar, but at least you can stack your DC's *FAR* easier than you can stack saves.

The reason miss chance is seen as a better idea is because it stacks on top of AC. If he hits your AC, then you check your miss chance. It's a second way to ensure that even if he lands a natural 20, he still might end up not generating a hit. Getting a decent AC isn't too hard, and will fend off the slings and arrows of random mooks. But for anything serious, you're going to want layered defenses and immunities to the more common methods of 'save or lose'. Depending on AC to the exclusion of everything else only means that you're most vulnerable to the most dangerous of opponents.

[ben-zayb]

One of the biggest upsides of arcane gishing is the ability to defend yourself against attack without having to resort to AC. (Greater) Mirror Image is a fantastic defense, as are Greater Blink and Greater Invisibility. All of these have the upside of simply negating attacks: effects such as True Strike need to know what square you are in for Greater Invisibility and Greater Mirror Image and Greater Blink aren't concealment, so True Strike does not work against them either.

Another advantage to Miss Chances (via Incorporeality, Concealment, etc.) is that there's no way AFAIK to optimize percentile miss chance rolls, so even a CR(+NI) challenge still has the same 20% or 50% chance to miss you without a way to completely negate it.

[AvatarVecna]

Another advantage to Miss Chances (via Incorporeality, Concealment, etc.) is that there's no way AFAIK to optimize percentile miss chance rolls, so even a CR(+NI) challenge still has the same 20% or 50% chance to miss you without a way to completely negate it.

Depending on the source of the Miss Chance, it's possible to negate the miss chance (Ghost Touch Weapons, True Seeing, Blindsight, etc.), but yes, if you're having to roll, you're basically stuck with the roll since it's not a d20 roll; there's ways to reroll d20s, but not percentiles AFAIK.

[ben-zayb]

Depending on the source of the Miss Chance, it's possible to negate the miss chance (Ghost Touch Weapons, True Seeing, Blindsight, etc.), but yes, if you're having to roll, you're basically stuck with the roll since it's not a d20 roll; there's ways to reroll d20s, but not percentiles AFAIK.

I did a pretty poor phrasing at the end but that's what I meant.:smallredface:

[Masakan]

The problem here is that it is tough to scale AC to the same degree that you BAB. It's pretty trivial to have a BAB of 50+ around level 10ish. It's decidedly not trivial to have an AC of 70ish by that point. Which brings up the problem...

AC, to have the effect you are describing, needs to be 19 higher than the BAB an offensive character would need. So not only are there fewer options to increase AC than BAB, but you need it significantly higher as well.

Furthermore, there's plenty of ways to simply bypass AC as a defense, such as Save or Lose effects, touch attacks, or just plain 'no save, no attack, just die now' effects.

AC only works on beatsticks, which are the least dangerous of the various threats you will have to deal with in your career as an adventurer. Damage sucks as a method of defeating your opponents unless you deal so much damage that you can reliably one-shot kill anything you swing at because if you don't actually kill your opponent, the effect you have had on his ability to return the favor is precisely zero. Save or Lose effects are similar, but at least you can stack your DC's *FAR* easier than you can stack saves.

The reason miss chance is seen as a better idea is because it stacks on top of AC. If he hits your AC, then you check your miss chance. It's a second way to ensure that even if he lands a natural 20, he still might end up not generating a hit. Getting a decent AC isn't too hard, and will fend off the slings and arrows of random mooks. But for anything serious, you're going to want layered defenses and immunities to the more common methods of 'save or lose'. Depending on AC to the exclusion of everything else only means that you're most vulnerable to the most dangerous of opponents.

Diamond Mind Saves....that is all.

[AvatarVecna]

Diamond Mind Saves....that is all.

Firstly, a one-line argument that isn't even really a sentence is usually going to be pretty terrible, unless it's close to a universal truth.

Secondly, I assume you're referring to the line of maneuvers that let you substitute a Concentration check for a saving throw; all of these use up your immediate action, so you can only use (at maximum) one of these once per turn. The rest of the time, you're stuck with your saving throws...not that those are hard to optimize either. Even without taking into account the fact that you're using up at least one prepared maneuver to use these things, they're still worse than immunities because they only work once per round, and they only allow you to succeed on the save.

EDIT: Had to add something; bolded it to make it obvious

[Yogibear41]

I find that comparing your AC at a given level to what a true Dragon of a similar CR has for its to hit bonus is a good gauge of what is a good AC at each level. I know its far from perfect but those guys generally have a pretty good to hit chance for their CR, especially as they get older and start getting those really high strength scores. If you can make a dragon miss you on a 19 on the dice roll you are in pretty good shape, heck if you can get them to miss on a 15 or worse that is pretty darn good. I have a level 7 character that has an unbuffed AC of 30 and a CR 7 Young Red Dragon can still hit him on a 11 or better with its primary attack, without getting buffs.

[eggynack]

Diamond Mind Saves....that is all.
I somehow doubt that would work against touch attacks, and other no save effects. I also thought your writing was somehow completely absent of absolutes, a thing not really in evidence here.

[Deadline]

Diamond Mind Saves....that is all.

It burns your immediate action, and you can't use it again until you refresh your maneuvers. If you want to use it every turn, you are likely a Warblade and doing nothing but making a single melee attack each turn. In other words, it shuts you pretty much completely down in order to mitigate against spells. It's impressive, to be sure, but it's not the end-all be all (boosting your saves and making them naturally is still generally better, that way you save your maneuvers for the save or lose spells you really need to avoid - that's why builds like the Sorcadin exists).

[ShneekeyTheLost]

Furthermore, Diamond Mind Saves are not relevant to the AC discussion at hand, and actually proves my point that you are wise to consider your saves instead of your AC because most things that are going to shut you down are going to typically have a save (if you are lucky) rather than an attack roll.

The melee equivelant would be Setting Sun. In fact, Baffling Defense + the stance Shifting Defense does a pretty amazing job instead of AC against a single opponent. In brief, the Counter Baffling Defense lets you use a Slight Of Hand skill check to defeat an attack roll. We all know how easy it is to jack skill checks, they aren't subject to auto-fails, and you can always find one of several ways to simply Take 10 on it. The stance Shifting Defense gives you a free 5' move on your opponent's turn when they miss. With no limit of how many you can take. What this does is you generate a miss, then move out of your opponent's reach, keeping him from following it up with even more attacks. So not only do you defeat one attack, you defeat his entire full attack in one shot. Combo with the training dummy for a 10' step instead of 5' to be even more certain that you are well out of reach.

Granted, it is only effective against a single opponent, so best used in boss battle type scenarios, but it does obviate AC against a single opponent.

[Deadkitten]

It burns your immediate action, and you can't use it again until you refresh your maneuvers. If you want to use it every turn, you are likely a Warblade and doing nothing but making a single melee attack each turn. In other words, it shuts you pretty much completely down in order to mitigate against spells. It's impressive, to be sure, but it's not the end-all be all (boosting your saves and making them naturally is still generally better, that way you save your maneuvers for the save or lose spells you really need to avoid - that's why builds like the Sorcadin exists).


You can recover all expended maneuvers with a single swift action, which must be immediately followed in the same round with a melee attack or using a standard action to do nothing else in the round (such as executing a quick, harmless flourish with your weapon). You cannot initiate a maneuver or change your stance while you are recovering your expended maneuvers, but you can remain in a stance in which you began your turn.

Not really directing this directly at you, you just happen to be the most recent post on the matter but if you go with the heavily implied interpretation that you can only ready a maneuver once then you can only use the save replacing maneuvers as a warblade every other turn since immediate actions take up your next turns swift, and you need both your standard and your swift action to recover maneuvers.

1/Turn every other turn is definitely something a DM can manage if they want you to fail a save.

So what I'm saying is that I essentially agree with you, except that it is actually slightly worse than even you thought it was.

And at this point I realized there was no real point to me even posting. :smallsigh:

[AvatarVecna]

Not really directing this directly at you, you just happen to be the most recent post on the matter but if you go with the heavily implied interpretation that you can only ready a maneuver once then you can only use the save replacing maneuvers as a warblade every other turn since immediate actions take up your next turns swift, and you need both your standard and your swift action to recover maneuvers.

1/Turn every other turn is definitely something a DM can manage if they want you to fail a save.

So what I'm saying is that I essentially agree with you, except that it is actually slightly worse than even you thought it was.

And at this point I realized there was no real point to me even posting. :smallsigh:

It's only that bad if you have to make the same save; there's a different maneuver for each one, so if (on three turns in a row), the DM called for a different save, you could substitute Concentration for all three of them, as long as you had all three maneuvers prepared. They would all be expended at that point, though, so that's a bit of an issue. And of course, if the DM calls for two saves in a row of the same type (within two turns of each other), you have to rely on your regular save for the second one.

Honestly, with how easy saves are to optimize, you're better off optimizing so that you can succeed on a 2-20, and then dipping Cleric for the Pride domain (reroll a nat 1 on a save, but must take the second result); that makes it a 1/400 chance of failure, assuming your save is high enough to beat the DC on a 2. You can make your odds even better by also having Choose Destiny active (expensive item, and rare spell, but worth it: it's like 5e advantage on every d20 roll); with Pride Domain, Choose Destiny, and a save good enough to succeed on a 2, you have a 1/160000 chance of failure.

...which is still worse than immunity to the effect, especially since immunities are likely easier to get.

[Deadline]

Not really directing this directly at you, you just happen to be the most recent post on the matter but if you go with the heavily implied interpretation that you can only ready a maneuver once then you can only use the save replacing maneuvers as a warblade every other turn since immediate actions take up your next turns swift, and you need both your standard and your swift action to recover maneuvers.

1/Turn every other turn is definitely something a DM can manage if they want you to fail a save.

So what I'm saying is that I essentially agree with you, except that it is actually slightly worse than even you thought it was.

And at this point I realized there was no real point to me even posting. :smallsigh:

Ah, right, I derped and forgot that swift action requirement. So you can use the same save maneuver once every other round.