The Decline of Armor in D&D

[Kyutaru]

Armor has been a staple in D&D since the beginning, a critical element of fantasy where knights and rogues alike can do battle. It has changed tremendously over the years, simplifying over and over again till it holds little variation. It has been normalized to the point of almost not mattering.

In 5th edition, the simplest form, the difference between medium Half-Plate and heavy Plate is a single AC point when equipped by a character with sufficient Dexterity (an easily acquired 15). The heavier armor requires greater proficiency, twice the cost, longer don/doff time, and weighs more but for many that +1 AC is worth the trouble.

In 3rd edition it was slightly more complex with varying Dexterity caps that promoted different armor types being valuable for different characters, to the point that going completely unarmored was actually the most superior choice. Full Plate existed as the penultimate heavy armor and it was truly only worthwhile for someone dumping Dexterity, especially with the severe penalties associated with wearing it. Players focused on builds that ignored armor entirely in an edition where Finesse weapons meant you could double dip on the benefits of single stat stacking and multi-classing for low hanging fruit could net you ridiculous AC benefits from unarmored defense features. Armor was, effectively, useless.

But in 2nd edition... things were quite different. Armor had a personality all its own with rules that supported its uniqueness and strengthened the importance of selection, especially when it came to non-magical armor. Full plate, for example, was not merely "better plate". It had to be customized specifically to the wearer by an armorsmith which prevented randomly looted suits of it from being usable. But this hassle was worthwhile if you made use of the Weapon Type vs Armor table, which dictated that it was 4 points better against Slashing weapons and 3 points better against Piercing ones. The best armor in the game could potentially be even better! This even added a layer to weapon selection since in general Slashing weapons were terribly inaccurate against most armors, Piercing weapons were rather neutral or slightly bad, and Bludgeoning treated them as they actually were. When looking at the damage and speed differences of these weapon types it becomes clear that this was with purpose and balance in mind. Suddenly Slashing becomes ideal against monsters while Bludgeoning becomes excellent against armored humanoids, with Piercing straggling the middle for archers and rogues and animals everywhere.

Going back to 1st edition adds a second layer to this. Instead of damage types, all weapons had their own specific armor modifiers based on how effective they were. Additionally, armor was weaker against men and better against monsters, boasting completely different roll requirements depending on who was attacking it. Now there's this matrix forming of damage dealt, enemy type, armor modifiers, enemy size, attack speed, and proficiency that affects which weapon might work best for your character. All because of how they interact with armor.

There was only one difference between a Rapier and Shortsword in 3rd edition because they were both one-handed finessable 1d6 piercing weapons -- the crit range for the Rapier was superior in exchange for it being a slightly harder weapon to have proficiency for. Yet the 2nd edition version of the Shortsword had piercing/slashing, making it a viable pick due to (you guessed it) armor.

Where do you think they should have stopped? How detailed does the combat of the game need to be? Massive tables add depth but slow the game. Concise uniform rules and normalized totals speed calculations up but eliminate the very identity of these choices.

[MrSandman]

In my opinion, it should stop where you have two or three armour categories. For a game set in something similar to the European middle ages, you could do: unarmoured, gambeson and mail. If it's set in the late middle ages or renaissance, you could do: unarmoured, mail and plate.

Anything more complicated than that tends to create a lot of unnecessary detail and miss the point of how armour actually works.

[Mordaedil]

That 2nd edition rule was entirely optional I am afraid, much like the unarmed combat section. Most people just played it straight AC.

[Aotrs Commander]

But in 2nd edition... things were quite different. Armor had a personality all its own with rules that supported its uniqueness and strengthened the importance of selection, especially when it came to non-magical armor. Full plate, for example, was not merely "better plate". It had to be customized specifically to the wearer by an armorsmith which prevented randomly looted suits of it from being usable.

Which I suspect, most people simply ignored (we certainly always did) and is a ridiculously simply piece of fluff to add system independantly if you really wanted it.

(Though I have to then say why plate but not all other armour, because in reality leather armour made for Gandalf won't fit Gimli or Legolas and it won't even likely fit Aragorn, actually...)

But this hassle was worthwhile if you made use of the Weapon Type vs Armor table, which dictated that it was 4 points better against Slashing weapons and 3 points better against Piercing ones. The best armor in the game could potentially be even better! This even added a layer to weapon selection since in general Slashing weapons were terribly inaccurate against most armors, Piercing weapons were rather neutral or slightly bad, and Bludgeoning treated them as they actually were. When looking at the damage and speed differences of these weapon types it becomes clear that this was with purpose and balance in mind. Suddenly Slashing becomes ideal against monsters while Bludgeoning becomes excellent against armored humanoids, with Piercing straggling the middle for archers and rogues and animals everywhere.

It instead says to me "anything that is not a bludgeoning weapon is actually an inferior choice" since against half your opposition, it doesn't work as well. Unless bludeoning weapons got penalties to attacking monsters?

I did not, to be honest, pay even the slightest bit of attention to those optional rules; having great big table for Thac0 was tedious enough without having to add a load of modifiers for weapons attacking armour classes. It is not remotely worth the hassle, in my opinion. I didn't even want to apply those modifers to Rolemaster for their varient weapons, where they already have a huge varienty of attack tables for weapons already - because it was far too much like a pain in the arse. (And if I was going for a system with some serious attempt at realism, AD&D would NOT be the one of the two I would pick*)



(Also worth noting that in the real world, the main reason everyone didn't have full plate was because they coudn't afford it. Dividing armour into types is, well, principally a construction of fantasy RPGs and D&D in particular.)

Where do you think they should have stopped? How detailed does the combat of the game need to be? Massive tables add depth but slow the game. Concise uniform rules and normalized totals speed calculations up but eliminate the very identity of these choices.

The short answer is "there is no good solution." You can make it hugely complex and ultimately largely pointless (since you are likely the only one at the table that really cares, most players don't give a frack, beyond any extra hassle said mechanics impose on them) or you can make it simple and unobtrusive and live with the fact that there are going to be sharp differentiations and most of your armour is going to be redundant.

Pillars of Eternity attempted really hard to do away with the problem that there are only really three armour types in D&D (four if you count unarmoured), but introducing a granularity scale meant that, actually, basically, they reduced it to "use either the heaviest armour or no armour unless you really want some special ability froma specific piece of kit in the middle."

PoE2 did a little better in making them different, but only by basically making it back to three types of armour, and the difference between them only the differences between what damage reduction, essentially, you get between different attack damage (so like those optional rules mentioned earlier); which is fine in a computer game where I don't have to do any of the messing around. Which is, honestly, to the point that it might as well be flavour unless you wanted to work out which damage type you most often come up against and its relative penetration across the coruse of an entire RPG.

Rolemaster has twenty different types of armour if five different categories, but again, the difference was really "either AT 1 or 5 (no penalties) or the highest in the category you'd currently developed skill to maneuver in at the minimum penalty."



*Rolemaster DOES have rules for fitting armour and they were not limited to plate. These rules have ALSO been ignored as entirely as those in AD&D in the past thirty years. Heck, in 3.x we only pay attention as far as size categories for weapons and armour.

[Morty]

I'm not seeing any more complexity in 2E D&D treatment of armor. I see a lot of fiddly details that try to be realistic, but not complexity or granularity. D&D combat is primitive and there's no amount of modifiers that are going to change it.

[Grod_The_Giant]

It stopped because D&D has spent the last couple decades evolving away from its roots as a medieval wargame, and towards being a broader fantasy adventure system. There's only so much complexity you can fit in a game, and repeated editions of D&D have chose to focus that on the characters rather than their equipment.

If you want more simulationist combat, I suggest checking out Riddle of Steel-- it's a (modern) game designed by historical martial arts pros to capture the feel of actual sword-and-shield fighting, and by all accounts is quite good.

If you want to see what happens if you keep adding realistic modeling to a game until someone forcibly drags you away from your keyboard, on the other hand, look at Phoenix Command, a RPG combat simulation designed by gun nuts, and full of nightmares like this:

Spoiler: My eeeeeeyes!

Show

[Cikomyr]

Armors are not the ones who took it the hardest. Polearms did.

Ever looked at the comparative pole arm chart?

[Vogie]

I wouldn't say it's the decline of armor as a category - but it is the decline of specific armor. In 5e there aren't specific modifiers outside of whether it decreases speed or causes disadvantage stealth checks... in RAW.

However, you have the flexibility to create armor that does fit into those niche categories.

• Splint armor that reduces slashing damage taken?
• A shield that gives +1 AC against melee attacks but +3 AC against ranged attacks?
• Arrowbane Ring Mail that increases AC vs Piercing damage?

If you want to give your players the niche armor, give them niche armor.

[Aotrs Commander]

If you want to see what happens if you keep adding realistic modeling to a game until someone forcibly drags you away from your keyboard, on the other hand, look at Phoenix Command, a RPG combat simulation designed by gun nuts, and full of nightmares like this:

Spoiler: My eeeeeeyes!

Show

My eeeeeeyeglows...

Ye gods, I can't even begin to process how that would work and I'm a pretty good rules-smith...

This reminds me of another pertinent point.

Something my Dad ran across while talking to a chap about realistic wargames; the chap had developed a tank shooting system which was essentially like that, but had completely bollocks movement speeds which had no bearing on reality.

A simulation is oinly as accurate as its weakest link.

No amount of armour modifiers is going to make D&D combat more realistic as long as it inherently abstracts everything else. Even Rolemaster's insane second-by-second round sequence optional rule in one of the companions wouldn't give a very realistic combat (I genuinely wonder if anyone, even the writers, actually used that in a real game - though it would be more credible than... The above...)

[Morty]

It stopped because D&D has spent the last couple decades evolving away from its roots as a medieval wargame, and towards being a broader fantasy adventure system. There's only so much complexity you can fit in a game, and repeated editions of D&D have chose to focus that on the characters rather than their equipment.

And in all honesty, 5E's armor and weapon tables are still more complex than they have any reason to be.

If you want more simulationist combat, I suggest checking out Riddle of Steel-- it's a (modern) game designed by historical martial arts pros to capture the feel of actual sword-and-shield fighting, and by all accounts is quite good.

You can also check out Song of Swords, RoS's spiritual successor with cleaned-up mechanics and designers who aren't actively offended by the notion of magic not dominating non-magic by default.

[Grod_The_Giant]

And in all honesty, 5E's armor and weapon tables are still more complex than they have any reason to be.

Armor could definitely be reduced to Light/Medium/Heavy. I'd actually like a bit more detail on the weapons-- or at least have categories like Axe/Sword/Polearm with slightly more distinct differences...

(To the Homebrew forum!)

[NichG]

If there are ever two choices where you can calculate ahead of time which is optimal, merge them. For variables you can't control (what kind of enemy are we fighting today, what weapons do they use), take statistical averages. Repeat and see what survives.

To put it another way, if at the end of the day the tradeoffs can entirely be reduced to an estimated rate of sustaining damage and only that, then there should only be a single choice. If the tradeoffs involve other factors (cost, weight, movement speed) then there's room for roughly one additional distinct category per relevant factor before things get samey. Irrelevant factors (enchantment price far exceeds base item cost, weight is trivially manageable for characters who most need armor because melee is strength-based) don't count. Also abilities that fill the gap make categories irrelevant (Mage Armor).

I'd say there's roughly space for three - gear that sacrifices defense to enhance active ability (assassin's getup, enchanted robes - give skill bonuses or things like immediate action potion/poison access, but permits no armor bonus to AC), gear that sacrifices active ability for defense (full plate, modeled as a strength, dexterity, and movement penalty but grants significant DR and AC), and compromise gear with no specific benefits or drawbacks.

In something where weather is important (extreme heat or cold), you could have an extra category based on different rates of exhaustion.

[Morty]

Armor could definitely be reduced to Light/Medium/Heavy. I'd actually like a bit more detail on the weapons-- or at least have categories like Axe/Sword/Polearm with slightly more distinct differences...

(To the Homebrew forum!)

I am sceptical if 5E has enough granularity to create such detail. D&D never really has, but 5E removed what little it did have (like crit range).

[Grod_The_Giant]

I am sceptical if 5E has enough granularity to create such detail. D&D never really has, but 5E removed what little it did have (like crit range).

I dunno, I could see something like

Weapon Group	Property	Light Weapon (Damage)	One-Handed Weapon (Damage)	Two-Handed Weapon (Damage)
Axes	None: Axes are balanced weapons	Handaxe (1d6)	Battleaxe (1d8)	Greataxe (1d12)
Clubs	Brutal: Clubs have a -1 penalty to attack rolls	Mace (2d4)	Warhammer (2d6)	Maul (2d8)
Spears	Reach: Spears can be used to attack targets 5ft away	Dart (1d4)	Glaive (1d6)	Pike (1d10)
Swords, Piercing	Finesse: Use Dex to attack	Dagger (1d4)	Shortsword (1d6)	Estoc (1d10)
Swords, Slashing	Accurate: Swords gain a +1 bonus to attack rolls	Kukri (1d4)	Longsword (1d6)	Greatsword (1d10)
Flails	Tangling: Grant Advantage on Shove attempts to knock a foe prone	Fighting Chain (1d4)	Whip (1d6)	Morningstar (1d10)

[Max_Killjoy]

Regardless of how armor itself is statted out in D&D, there's always that underlying core oddness of trying to combine "how hard is this character to hit" with "how hard is it for a hit to hurt this character" into a single thing, and having that thing be entirely passive (as in attacker rolls, compares to static value, hits for effect or does not hit for effect).

[The Glyphstone]

I dunno, I could see something like

Weapon Group	Property	Light Weapon (Damage)	One-Handed Weapon (Damage)	Two-Handed Weapon (Damage)
Axes	None: Axes are balanced weapons	Handaxe (1d6)	Battleaxe (1d8)	Greataxe (1d12)
Clubs	Brutal: Clubs have a -1 penalty to attack rolls	Mace (2d4)	Warhammer (2d6)	Maul (2d8)
Spears	Reach: Spears can be used to attack targets 5ft away	Dart (1d4)	Glaive (1d6)	Pike (1d10)
Swords, Piercing	Finesse: Use Dex to attack	Dagger (1d4)	Shortsword (1d6)	Estoc (1d10)
Swords, Slashing	Accurate: Swords gain a +1 bonus to attack rolls	Kukri (1d4)	Longsword (1d6)	Greatsword (1d10)
Flails	Tangling: Grant Advantage on Shove attempts to knock a foe prone	Fighting Chain (1d4)	Whip (1d6)	Morningstar (1d10)

Why is a Morningstar considered a two-handed flail-type weapon? Not disputing the conceptual idea of weapon-types, just questioning that specific choice.

[NRSASD]

I'm nibbling away at making a slightly more complex armor/weapon system, where the heavier armors provide categorically better protection than light armor but are easier to hit. Things like resistance to specific damage types or maybe even damage mitigation (like the heavy armor master feat).

This gives me the chance to make weapons have more meaning, as currently there is no difference between a battleaxe and a longsword. But the overarching goal is to keep things simple. We'll see if that manifests the way I hope it does.

[awa]

Why is a Morningstar considered a two-handed flail-type weapon? Not disputing the conceptual idea of weapon-types, just questioning that specific choice.

for some weird reason a lot of games (mostly computer) confuse spiked flails with morning stars


I run a fairly complex system where armor is variable DR, but also converts some of the excess damage into subdual. They also make it harder to get critical hit.

Armor use is limited by your combined str+con mod the logic being while a common solider just needs to wear their armor to the battle field a pc needs to be able to swim a river or climb mountains in their armor and thus needs to be able to wear it like a second skin.


Weapons like axes have the Armor piercing trait reducing the effectiveness of armor

weapons like clubs have the impact trait which deals bonus subdual damage that is harder to block with armor

weapons like daggers have keen that makes it easier to get critical hits (which grant a large amount of armor penetration)

Their is a lot more to it but that is the relevant part.

This system would not work with a straight port to say 3rd edition where feats like power attack would break it.

[Willie the Duck]

Get this quick aside out of the way:

It instead says to me "anything that is not a bludgeoning weapon is actually an inferior choice" since against half your opposition, it doesn't work as well. Unless bludeoning weapons got penalties to attacking monsters?

Although later parts of 2e added some better bludgeoning options (and some printings had morningstars dance around between B and P), the primary penalty for bludgeoning weapons was that they did significantly less damage. A longsword, for instance did 1d8 vs small and medium opponents and 2d6 vs large and larger opponents. For a Warhammer, those numbers were 1d4+1 and 1d4.

Anyways, to the meat of the discussion:

Where do you think they should have stopped? How detailed does the combat of the game need to be? Massive tables add depth but slow the game. Concise uniform rules and normalized totals speed calculations up but eliminate the very identity of these choices.

I did not, to be honest, pay even the slightest bit of attention to those optional rules; having great big table for Thac0 was tedious enough without having to add a load of modifiers for weapons attacking armour classes.

I'm not seeing any more complexity in 2E D&D treatment of armor. I see a lot of fiddly details that try to be realistic, but not complexity or granularity. D&D combat is primitive and there's no amount of modifiers that are going to change it.

It stopped because D&D has spent the last couple decades evolving away from its roots as a medieval wargame, and towards being a broader fantasy adventure system.

Something my Dad ran across while talking to a chap about realistic wargames; the chap had developed a tank shooting system which was essentially like that, but had completely bollocks movement speeds which had no bearing on reality.
A simulation is oinly as accurate as its weakest link.

I think oD&D did what made sense -- it was still a derivative or a fantasy subsection or a wargame (where deciding to equip your troops with pikes vs swords and shield vs axes and shields vs lances and horses was one of the primary tactical choices one could make), and therefore it made sense that these choices should play a part in this new game. It wasn't strictly necessary when playing this odd dungeon-crawling variant, and the supply lines and switching out troops were gone, but there quickly became this logic that the PCs would have a bunch of retainers and hirelings, so you could still carry around different arms and armors depending on situation (plus all weapons did 1d6 damage, so I can see the desire for some level of variation). However, as it became clear that a lot of what you would fight would be monsters (not enemies in armor), plus once a major goal (other than gold/xp) was to find that magical weapon +1 (2, 3, eventually +5 holy avenger or somesuch) and the similar armor and shield, a whole lot of the reason for wanting that variation and decision making (are you really going to switch out of your plate+2 to put on chainmail +1 simply because you think you might run up against opponents armed with weapons that do worse vs chain?) dried up.

So at that point you end up with a whole lot of complexity desperately seeking purpose. As others have pointed out, realism is not a good candidate, given the other weaker links in the realism chain for D&D. B/X-BECMI and 2e if you ignore the S/P/B optional rules is fairly good -- heavier/more expensive armor is generally better if you can afford the expense/encumbrance with caveats. Those caveats being that unarmored is for mages; leather and studded are for thieves/rangers (with each individual making different choices on unarmored, leather, or studded depending on how important their thief skill %s were to them); leather and hide are for druids; chainmail is for elves to give fighter-mages a special variety of; and everything else is for non-ranger warriors and clerics. That at least gives a fairly reasonable reason for most of the armors to exist.

3e did change all that. OP's position that armor was a horrible option or only for Dex-dumping seems a little overboard. Plenty a decent character had a 16 Dex and mithril plate at high levels to achieve their front-line prowess. However, the overall movement away from 'as heavy as your purse/carrying capacity/class features will allow' was something of a spanner in the works. I get the 'why' (so people could play swashbucklers alongside knights), but the implementation had some hangups.

Overall, I think the answer to 'they should have stopped' probably would have been right after the LBB introduction of oD&D. Once they figured out how people were playing the game, they should have seen that the weapon vs. armor table was not a huge part of most people's games, and dropped the complexity-in-search-of-purpose like a bad habit... or figured out a specific system to replace it with an emphasis on having a specific "why" to the complexity.

[Beleriphon]

I like Star Wars Saga Edition's take on armour. It adds to Reflext and Fortitude scores BUT without investing heavily in armour use a character cannot add their dexterity or unarmoured defenses to Reflex while wearing armour. So an unarmoured character usually had a better Reflex score, but much worse Fortitude, never mind the other benefits stuff like a sealed space suit provides outside of a Star Destroyer. Sure a character could achieve astronomically high Relfex scores by investing in wearing armour, but that means specifically building for that with the talents, feats, and what not.

I have no issue with the way armour works in D&D, specifically 5E. I like that there are some goofy options, and some good options. It sets up a series of choices, sure players are going to pick the best option hey possibly can for the costs they can afford in game, but that's not unusual. A lot of the other options are genuinely there for NPCs builds. If you look at some of the monsters they have armour in their stat block, and its not always the best option.

[Keltest]

I liked the way Pillars of Eternity handled armor: it was just straight up damage reduction, but heavier armor gave you a bigger delay between turns/lowered your initiative depending on which rule set you were using. Wearing plate armor made you nearly invulnerable to anything not specifically designed to fight it, but you would always act last in turn based, or have really slow action speeds in real time. Meanwhile the wizard in the backline wearing just a robe could get off two or three spells in the time it would take you to get one attack done while wearing full plate, or could lead every round with a strong CC or damage spell with their good initiative.

[pendell]

How much of those details become important once magic becomes involved? It's been joked about in the strip, but a +12 leather bikini is going to be superior to plate. In a world where magic items> non-magical, then the non-magical details aren't really important, since you're going to discard them just as soon as you can loot something better from a creature with the appropriate treasure type.

Respectfully,

Brian P.

[HouseRules]

I dunno, I could see something like

Weapon Group	Property	Light Weapon (Damage)	One-Handed Weapon (Damage)	Two-Handed Weapon (Damage)
Axes	None: Axes are balanced weapons	Handaxe (1d6)	Battleaxe (1d8)	Greataxe (1d12)
Clubs	Brutal: Clubs have a -1 penalty to attack rolls	Mace (2d4)	Warhammer (2d6)	Maul (2d8)
Spears	Reach: Spears can be used to attack targets 5ft away	Dart (1d4)	Glaive (1d6)	Pike (1d10)
Swords, Piercing	Finesse: Use Dex to attack	Dagger (1d4)	Shortsword (1d6)	Estoc (1d10)
Swords, Slashing	Accurate: Swords gain a +1 bonus to attack rolls	Kukri (1d4)	Longsword (1d6)	Greatsword (1d10)
Flails	Tangling: Grant Advantage on Shove attempts to knock a foe prone	Fighting Chain (1d4)	Whip (1d6)	Morningstar (1d10)

Swords is more called Blades.

There's an emphasis on piercing, called daggers, and include the polearm version, called spears.

There's a balance on piercing and slashing, called straight swords, and include the polearm version, called staffsword.

There's an emphasis on slashing and it's usually one-edged, called knife, and include the polearm version, called glaive. Japanese have Naginata.
Long Knives are longer than many short swords.

[Grod_The_Giant]

Why is a Morningstar considered a two-handed flail-type weapon? Not disputing the conceptual idea of weapon-types, just questioning that specific choice.

Mostly because I didn't want to reuse the term "flail" and couldn't come up with anything else quickly. (I'm imagining something like the Witch King was using in Return of the King). Might be better to have a separate "unique" category for things like nets, flails, and lances, though, come to think.

[The Glyphstone]

Mostly because I didn't want to reuse the term "flail" and couldn't come up with anything else quickly. (I'm imagining something like the Witch King was using in Return of the King). Might be better to have a separate "unique" category for things like nets, flails, and lances, though, come to think.

"Snares" could work for stuff like nets and flails. Though an 'other' category is probably good for the weird stuff.

[Tvtyrant]

Armor could definitely be reduced to Light/Medium/Heavy. I'd actually like a bit more detail on the weapons-- or at least have categories like Axe/Sword/Polearm with slightly more distinct differences...

(To the Homebrew forum!)

I felt like 4E did this really well, with different accuracy bonuses and feat trees for the different weapon types. It is also the most combat oriented of the editions though.

[Great Dragon]

@Grod: the weapon chart is nice.
I'm not sure if it would really help much in my current games, though.

Like some of the other members here, I also ignored the Weapon vs Armor rules for AD&D 1&2.

5e D&D Armor is a little annoying.
Reference

PCs with High AC are able to take on CRs around Three times their level! (Like a Level One PC with AC 18-19 can easily take on most CR 3 Monsters.)

Spoiler: Hobgoblins

Show

I just got done running "CR 1" Hobgoblins against a Level Two party. With the Hobgoblins having AC 18, they were a very hard fight for the Party!

I felt that base (CR 1) Hobgoblins should have AC 14, Veteran Hobgoblins AC 16 (CR 3), and Captain Hobgoblins AC 18. (CR 5)

Especially since monsters aren't really supposed to easily match PCs: just have different or interesting Abilities.

And, I felt that not every Hobgoblin would have access to Chain Mail (with shield).

Now, Armor Cost keeps all the classes to Scale Mail (50 gp. AC 14 +2 Dex +2 shield = 18) at best, until around 3rd level.

Studded Leather is 45 gold. (AC 12 +2 shield + [5] Dex = 19 max).

At 750 gp for Half-Plate (AC 15 + 2 Dex and +2 shield = 19), even the Fighter and Paladin are not really scary with Splint Mail (AC 17 +2 shield = 19) for 200 gp.

Plate (AC 18 +2 shield = 20) costing 1,500 gp means that this is normally not available until about 5th level.

*****

Now, as always, the biggest part is optimization at Character Creation.

Anyone with 20 Dex (Halflings, Elves, and Goblins) doesn't need Heavy Armor, since they can get an AC 19 with just Studded Leather and shield. (55 gp) From 1st level!

But, starting out with access to Medium Armor means that even with a 14 Dex, they can still rock out with that same AC 19 as soon as they get 760 gp. (About 3rd level?)

True, Heavy Armor plus shield can get AC 20:
But, it's not like there's an Assembly Line Factory cranking Plate Armor out!

But, 5e is still waaay better then 3.x!
3.x: you could stack your +5 Full Plate (13), +5 tower shield (+9), +5 Dex (mithril plus nimbleness), +5 Amulet of Natural Armor, and +5 Ring of Protection; +1 for Dodge Feat = AC 38. At about 17th level.

Where the only Monsters that are really able to hit that on a regular basis are things like Great Wyrm Dragons, Storm Giants (Mountain Giants? and maybe Cloud Giants?) and the Tarrasque.

True, another 20th Level Warrior type with 30 Str and a +5 weapon could still hit that with a roll of 3+ on their first attack, 8+ roll with their second attack, 13+ roll with their third attack, and 18+ roll with their fourth attack.

But then we're getting into PvP territory, there.

[Grod_The_Giant]

Where the only Monsters that are really able to hit that on a regular basis are things like Great Wyrm Dragons, Storm Giants (Mountain Giants? and maybe Cloud Giants?) and the Tarrasque.

Uh, looking at CR 17 monsters...

• An Aboleth Mage can't hit you with its tentacles, but that doesn't matter because it's a 10th level Wizard.
• An Old Brass Dragon hits you on a 3+.
• A Mature Adult Bronze Dragon hits you on a 4+.
• A Marilith needs to roll a 13 to hit you, but it's also rolling 6 times.
• A Formian Queen can't make physical attacks, but can cast 8th level spells, so, you know, good luck.
• A Frost Giant Jarl needs an 8 to hit you with its first attack and a 13 with your second, which is just about fair.
• A Very Old White Dragon hits you on a 3+.

The Tarrasque has a +57 to hit.

[awa]

Uh, looking at CR 17 monsters...

• An Aboleth Mage can't hit you with its tentacles, but that doesn't matter because it's a 10th level Wizard.
• An Old Brass Dragon hits you on a 3+.
• A Mature Adult Bronze Dragon hits you on a 4+.
• A Marilith needs to roll a 13 to hit you, but it's also rolling 6 times.
• A Formian Queen can't make physical attacks, but can cast 8th level spells, so, you know, good luck.
• A Frost Giant Jarl needs an 8 to hit you with its first attack and a 13 with your second, which is just about fair.
• A Very Old White Dragon hits you on a 3+.

The Tarrasque has a +57 to hit.

actually the maralith also has a tail attack and at will magic weapon so it needs a 12 with 7 attacks so its even less fair.

That said a lot of these monster are assumed to be using power attack so this at least weakens that, on the other hand you spent a lot of resources getting your ac that high. Of course as in all things 3.5 being a caster allows you to have a vastly higher ac far more easily if that is your thing.

edit I love how (I assume) we both immediately looked up the srd on the CR17 monsters

edit 2 Ac is useful in the early game but its hard to keep it up in the late game and increasingly ineffective as the spells get more and more powerful

[Great Dragon]

Thanks, Grod and awa!!

I lost all my 3.x books while moving, I'm stuck on my phone and not able to access PDFs, and no more room for downloads.
(And I can't post and read the SRD at the same time)

The ones I listed were what I could dredge up from memory.

I had forgotten that the Tarrasque had a +57 to hit!!! (And four attacks? Claw x2, Bite, and Tail?)

I was not listing Spellcasters for attacking the Super Armored PC, since I considered that to be separate.

@Beleriphon:
My GM tried SW Saga, but ended up going with Revised Core for ease of play.

We do like that SW Core Armor grants DR.