Thursday, 7 February 2013

First impressions of AD&D 2nd Edition

So my Wednesday night group is just starting our new VOIP game, which will see Arthur unleashing us on a lovingly-crafted Advanced Dungeons & Dragons Second Edition game. This is actually my first time playing anything earlier than 3rd, outside of Baldur's Gate and the like. I do have a broad idea of how things work - I think I first encountered the rules in a copy of Dungeon bought to placate me on a family visit to relatives in New Jersey, of all things - but I don't have the kind of familiarity I'd like. On the plus side, it's a very knowledgeable group of people who are very good at explaining things.

Our first session was just getting things ready. We're using a new bit of software for VOIP, so had to play with settings and features a bit, and there's a new player that I hadn't met previously. Mostly though, it was chargen.

I read through the Player's Handbook last week to brush up on things, though quite soon it turned into skipping. A lot of the information there is pretty specific, and I don't really need to memorise lists of spells and equipment, or alternative ways of doing non-combat stuff. That's all nice to know, but it wasn't really worth poring over before the game's even started. Anyway, here's my first impressions of the game.

Obligatory THAC0 bit

THAC0, on reflection, doesn't seem as bad as it's often suggested to be. It's not very intuitive, and certainly compared to the simplicity of a Base Attack Bonus (which works in exactly the same way as every other modifier to your die roll) it's arcane. Having both THAC0 and Armour Class start high and improve as they decrease, in contrast to every other part of the game and most aspects of real life, displays a piece of logic that I find completely baffling. I can't imagine any reason why anyone would ever, ever choose to use THAC0 when other options were available.

That being said, I think some of the problem stems from a needlessly confusing explanation. Here's the game's description of THAC0 in use:

The first step in making an attack roll is to find the number needed to hit the target. Subtract the Armor Class of the target from the attacker's THAC0. (Remember that if the Armor Class is a negative number, you add it to the attacker's THAC0.) The character has to roll the resulting number, or higher, on 1d20 to hit the target.

Rath has reached 7th level as a fighter. His THAC0 is 14 (found on Table 53), meaning he needs to roll a 14 or better to hit a character or creature of Armor Class 0. In combat, Rath, attacking an orc wearing chainmail armor (AC 6), needs to roll an 8 (14-6=8) to hit the orc. An 8 or higher on 1d20 will hit the orc. If Rath hits, he rolls the appropriate dice (see Table 44) to determine how much damage he inflicts.

Now what if we imagine the same rules, explained by someone sober?

Roll a d20 and add the target's Armour Class. If you equal or exceed your THAC0, you hit.

Rath has reached 7th level as a fighter. His THAC0 is 14 (found on Table 53). In combat, Rath, attacking an orc wearing chainmail armour (AC 6), needs to roll an 8 (8+6=14) to hit the orc. If Rath hits, he rolls the appropriate dice (see Table 44) to determine how much damage he inflicts.

But it's still not fantastic. So what's THAC0 doing wrong? Well, a mixture of practical and descriptive issues, I think (and bear in mind this is mostly theoretical, partly based on Actual Play podcasts, and with a smattering of Baldur's Gate, rather than direct experience of the tabletop game).

  • You need to know the enemy's AC, which the DM must tell you each time, giving away useful information (and reducing any possibility of constructive fudging).
  • It requires subtraction, which people tend to find slower than addition - especially when negative armour class comes into play.
  • Amour class starts high, then decreases as it improves. Chainmail has AC6, but magic chainmail +1 has AC5, not AC7. Not only is this unintuitive, it's effectively contradicting itself. I can see where they were coming from (pluses are good, minuses are bad) but anytime you can avoid having to memorise arbitrary system-specific exceptions to normal use, you should.
  • Under the suggested description, you reverse any bonuses, deducting positive numbers from your THAC0, and adding the value of negative ones.
  • Similarly, a dagger +1 gives a bonus that must be subtracted from THAC0, not added to it.
  • You're expected to calculate THAC0 for each specific weapon you have, each of which must then be modified to fit the target.
  • THAC0 itself is, a lot of the time, a sort of secondary statistical entity, at least one step removed from anything concrete: it's the number you must roll to hit an Armour Class of 0, which you are virtually never doing. This makes it fiddlier to think about, in much the same way that standard deviations or a chi-squared x2 value are slipperier to think about than arithmetic. A Base Attack Bonus is a number you add to your die roll; a Bend Bars percentage is the percentage chance you have to bend some bars. These are much more intuitive.

As far as I can see, the switch from THAC0 to Base Attack Bonus must be one of the most sensible decisions in gaming history, at least in terms of making the game accessible.


The layout also suffers from some organisational issues. Most notably, it's quite hard to actually get all the character information beforehand. Saving throws are under the section on combat, not in character generation - in fact there's no real character generation section. Now, in fairness, they may have deliberately decided it wasn't necessary. For example, you may not strictly need to know your character's saving throws until you encounter the appropriate hazard. However, I did find it very disconcerting sitting with a character sheet, flicking back and forth as I tried to work out where the missing information was. I'm used to systems with a fairly all-encompassing chargen, where you can start at the beginning, fill in your sheet as you go, and then be ready to play. In contrast, AD&D seems almost to leave things so that you only work them out as necessary, which in some ways makes it quite like a Bethesda RPG. I'm not prepared to say that it's objectively a bad thing to do; I'm not remotely qualified. What I can say is that it leaves the game relatively difficult to absorb before you start play; and that that seems likely to cause delays when those rules do eventually come into effect.

Another issue is, the book points you towards specific chapters for things, but only displays chapter numbers on the first page of each chapter. Instead, it offers cutesy symbols on the remaining page: crossed swords are combat, a hand is wizarding, a hand with an ankh is priesting, a random assortment of objects that looks like something off a white elephant stall is class descriptions, a troll snarling in your face is encounters... There are two major problems with this approach. The first is that some of them really aren't very intuitive, but the names of the chapters are given alongside them if you open the pages properly. More pressingly, this in no way helps with the lack of numbers, since finding Combat or NPCs is no flipping use to you in finding Chapter 5: Proficiencies unless you also know what number every other chapter is. Yes, there is an index, and a contents section, but it's much nicer to just be able to flip through the book. This is particularly true when some chapters are only two pages long, and can be pretty tough to find. So, could do better.

I also found the presentation of options in some places cluttered and confusion - for example, there are different ways of calculating effects of encumbrance, and I spent quite a while trying to work out how they fitted together, before eventually spotting that they are alternatives rather than complimentary systems. The table for basic encumbrance doesn't mention movement at all, but is followed on the next page by a table listing detailed movement penalties for encumbrance. As I was trying to fill in a character sheet, I just worked through them. It's only on the third page - after descriptions of animal carrying capacities, stowage for containers, mount encumbrance and magic armour - they they get around to explaining how the basic encumbrance rules work with movement, at which point I very slowly realised something was up. But look, if you have a table called "Modified Movement Rates" that lists how encumbrance affects your movement, I don't think it's that unreasonable to assume that the latter is what you use. Stick an "optional" in the title or something, don't explain that bit in a sidebar that isn't even adjacent to it.

In terms of the burning question "is this laid out less usefully than Call of Cthulhu (6th edition)?", I'm yet to reach a final verdict. I suspect it's going to be a close call.

Saving Graces

The saving throws are really very arbitrary, and read almost exactly like "stuff what's come up in games I ran so far", rather than any kind of systematic attempt to create a system for avoiding effects. There's the random list of quite specific magical effects, with their own little subsystem, but then any other possible effect is just an ability roll. I have to wonder why it didn't occur to them to try and mesh the two.

There's a weak argument you could make based on the text in the book, which for the Everything Else saves says: "When a character attempts to avoid danger through the use of one of his abilities...", implying that the set saves cover passive resistance and everything else is an ability roll. But that's a bit misleading, since it should be possible to actively avoid quite a few breath weapons and some kinds of spell - anything from fireballs to hypnotic gazes to visible tendrils of transmuting energy. While I can see how you might evolve a system of level-based saves for miscellaneous effects, why split them down in this way? Why are Poison, Death and Paralyzation one save, and Petrification another? Why are Wands and Rods different from spells cast any other way, and why are both different from Poison anyway? Why are Warriors more vulnerable to Poison than Rogues are, and why are Wizards better at surviving Breath Weapons than anyone else?

Yeah, I'm not convinced.

Fast, Short and Lively

On the upside, I was pretty impressed with the speed of basic character creation. Filling in the sheet took quite a while as I rummaged through the book in search of saving throws, equipment weights or whatever. However, actually rolling up Oswyn was really simple. Six lots of die rolls, followed by a fairly simple choice of class: generally you won't be qualified for about half the classes as they have ability prerequisites, and if you have one or two high scores you really want to pick the matching class.

As it happened I had generally decent scores, but high Strength and Con, so a fighter was the obvious way to go. Also, to be fair, I wanted something straightforward for my first game in the system. Having picked a fighter, all there was to do was pick a couple of weapon specialisations and go shopping. No skill points to carefully apportion, with an eye on feat choices a few levels later. No feats to choose. Admittedly I actually rather like feats, but that's not the point: they do add another, really quite large, layer of complexity. Also, in fairness, we aren't using weapon proficiencies (only specialisations) nor non-weapon proficiencies, which does save some time. And we're aiming for a human-based campaign. We did roll up backgrounds to help with character inspiration, though. But yes, I was impressed than in about an hour and a half, we managed to test the VOIP software, introduce ourselves as necessary, establish some background, create characters, shop, start working out party backstory and do a fair bit of nattering.

So far, my time breakdown looks pretty much like this (in order of time spent, not in chronological order, in case you're really confused by my apparently bizarre approach to chargen):

  1. Fill in character sheet: ~30 mins
    1. of which ransacking rulebook: ~ 24 mins
    2. of which entering information: ~5 mins
    3. of which arithmetic: ~ 1 min
  2. Choose name: ~10 mins
  3. Invent character: ~5 mins
  4. Pick class and character abilities: ~2 mins
  5. Roll attributes: ~1 min
  6. Pick class: ~30 secs
  7. Pick race: ~1 sec

whereas a typical Pathfinder breakdown might look more like:

  1. Pick class and character abilities: ~30 mins
    1. of which reading feat descriptions, synergies and prerequisites: ~25 mins
  2. Choose name: ~10 mins
  3. Pick class: ~2 mins
  4. Fill in character sheet: ~5 min
    1. of which entering information: ~3 mins
    2. of which ransacking rulebook: ~ 1 min
    3. of which arithmetic: ~ 1 min
  5. Invent character: ~5 mins
  6. Pick race: ~1 min
  7. Roll attributes: ~1 min


So on the whole, I've got some reservations about the actual product. That doesn't really affect my anticipation for the game: as I said, I've got an experienced DM, everyone else has solid prior experience with the game, and I don't really expect to run into any problems. So I'm pretty excited.

And obviously, this is an early RPG, and it's not entirely fair to compare it to more recent developments and try to draw some kind of direct equivalence between them. Later editions of D&D, and pretty much everything else, have... well, at least had the opportunity to learn from AD&D 2nd edition, both in terms of actual rules, system coherence, and also fluffier issues of layout and organisation.

So yes, a not particularly deep non-review of a twenty-year-old RPG system. Woot. I can't imagine why this blog hasn't acquired a fanatical cult following yet.


  1. Re: THAC0 - It's interesting to compare this to the arrangement in 1st edition, where the Player's Handbook didn't contain the to-hit charts. Those were all bundled in to the Dungeon Master's Guide, along with saving throws, because Gary Gygax is the Old Testament God of tabletop RPGs and wanted to encourage GMs to be the same. Your GM could just let you know your saving throws and to-hit rolls if he or she wanted to, or they could just ask you to roll and add applicable bonuses and look it up themselves.

    FWIW, for the campaign my intention is to do something like that, simply to keep things speedy; you won't know the monster ACs, but then again if I've done my job properly you'll be able to tell roughly what standard of armour they are wearing and if you really wanted to you could suss it out fairly soon. Likewise, since I'll make sure to have your ACs in front of me, I won't be reporting monster dice rolls to you every time I roll a to-hit for them, I'll just narrate the effects of their attack and then tell you your character is dead.

    Re: Layout - Yep, no doubt it's a mess. It's somewhat friendlier once you are used to it - unlike, say, Weapons of the Gods, a game Dan and I played for months where I never sussed out the arrangement of the rulebook and I don't think anyone else did either. It's worth bearing in mind that the game was published at a time when the basic D&D game was also available, so the designers could be a little more sure that people coming to it knew their way around the system; all-noob groups were more likely to be playing basic before moving up to Advanced (or just ploughing on through the Expert/Companion/Master/Immortals sets or the Rules Cyclopedia), whereas a noob in a group of more experienced players could have a player or DM giving them a helping hand navigating the book. So I guess making the layout super-simple was a lesser priority back then.

    Re: saving throws - yeah, "stuff what's come up in games so far" was as far as I can tell how they came about in OD&D and nobody really bothered to rationalise them until 3rd edition.

    Re: character gen speed - I hear anecdotally that a lot of people find all-random systems like the TSR-vintage D&D systems much easier to introduce new players, because new people can get very, very easily overwhelmed with all the choice 3rd ed and later editions give you. And of course once you do learn your way around the rulebook it becomes much faster - based on your figures (which look right to me) the AD&D gen process could come to under half an hour, whereas with Pathfinder I don't see it going down below the 50-plus minutes you've worked out there unless you've spent a massive amount of time studying the feats or you're using a build guide.

    1. ...painstakingly written commend is blatted by software restart.

      The 1st edition arrangement sounds completely insane, but I suspect it's mostly down to a gulf in thinking. Coming to it from recent games, where players have a lot more direct control and sometimes the GM/player barrier is fairly blurry, keeping that kind of information from them seems weird. But I can see, if you come at it from a refereed wargames background, then for one thing you might actively want to conceal tactical information from players, and for another you might just not see any particular reason to share detailed mechanics with people. Especially if (as I get the impression) Gygax was viewing D&D as a game a knowledgeable GM would Run For people with no gaming experience whatsoever, rather than a more collaborative thing.

      I'll just narrate the effects of their attack and then tell you your character is dead.

      Well, can’t say you didn’t warn us...

      Re: Layout – it seems to be one of those things that’s quite easy to spot flaws in, or at least perceived weaknesses, but incredibly hard to really pin down how to do well. Cthulhu has a lot of problems, Hellcats had some definite issues, and I seem to remember After Sundown was a bit all over the place as well. I suppose part of it is that any but the most basic game is going to have interaction between different subsystems, and there’s not necessarily any one right place to put some of the information.

      Re: Chargen – I think Pathfinder and the like take ages the first time you’re presented with them, because there’s an overwhelming choice of skills, feats, spells and so on. After a couple of goes, even if you don’t try to learn the details, you understand how they work, and are better at filtering out class-irrelevant options, or things you won’t be qualified for at 1st level. A bit more, and you can recognise the more straightforward and common options, which is fine if you’re not that bothered about mechanising things too much. However, to get the most out of the game both fluff- and crunchwise, you do really need to either learn the options, or spend a lot of time on chargen. On the plus side, good presentation can make it a lot easier by clearly showing you prerequisites, dependency chains, and generally helping to filter out some of the noise. Some books do better than others at that, though some kind of tool or app that could actively filter for you would be a good idea. I have somewhere a sizeable file, lovingly prepared, that collates the feats from all the major Pathfinder rulebooks, correctly alphabetised and indented for feat chains... Quite a job. I’d do something with it, only I’m not sure what, or indeed what I’d be allowed to do. Plus, I mean, my Pathfinder game is defunct now.

  2. The other thing to remember about AC is that it evolved out of a tabletop wargame where Armour Class was literally an Armour Class - each type of armour had a value and there was no overlap, so AC1 was "first class armour" and AC2 was "second class".

    I suspect that in that context "add AC to dice roll and check vs THACO" was also just as sensible as "Add AB to dice roll and check vs AC".

    1. Sorry, not quite getting you in the second paragraph there.

    2. I mean that in a wargame, you know your opponent's stats (or they can tell you) so "D20 + AC vs THAC0" was a perfectly straightforward mechanic, a lot of what gets confusing about it comes in when AC becomes secret.

      Similarly in a tabletop wargame having a static to-hit based on your skill which is modified by circumstances (like enemy armour) is fairly common. If you think about it, it's sort of how shooting works in Warhams - BS3 means you need a 4+ to hit, and then you put modifiers on top of that.

      Sorry, I didn't express that very clearly at all.

    3. Ah, I see. True, but I still can’t fathom how they ended up with “THAC0 – AC vs. d20” as the mechanic. No, wait, I suppose if the GM is looking after both THAC0 and AC that does sort of work. But I can’t see how you’d wind up with that mechanic in wargaming unless it was all done in secret by referees, which isn't my experience at all.

      From what I’ve found online, one of the reasons things seem weird is that the original mechanic was not a scaling armour value, but just a list of armour types that happened to be given numbers. These worked on a chart vs. different attack types, so there really wasn’t any arithmetic in there, just pure comparison. This post highlights the Chainmail rules where different weapons hit different armours on various numbers. It seems like at some point people switched to using number labels for the armour types (each number presumably covering several possible types), though they could perfectly well have used letters instead for what they were doing.

      So far as I can tell, “add AC to die roll and check vs. THAC0” wouldn’t have made sense then, because AC wasn’t a numerical value, just a label.

      Eventually the idea of AC as an actual number came in to presumably streamline things, and they introduced THAC0, except they didn’t quite take it far enough. So to me it looks like the real problem was that nowhere in that evolution process did anyone step back and say “hang on, if we’re changing all these elements, is there a more logical way to model it?”

    4. Bear in mind also that THAC0 came in actually quite late - it's a 2nd ed. innovation (previous editions just gave you big tables). There may have been a desire to maintain backward compatibility with the extraordinary amount of 1st edition stuff they'd already printed and would ideally like to sell off to 2nd ed players.

  3. And now I understand how THACO works. Thank you.

  4. Re: Thac0, I've been played the 2ed for years. You may have got the Thac0 a little bit wrong. Tha DM doesn't need to tell yoo the enemy AC at all.
    If we take a look at the combat resolution step what the player is expected to do for attacking is:
    - roll the d20
    - add to the d20 result all his bonus (strenght, specialization, magic..) and subtract all his penalaties (possibly encumbrance, some penaltise for specific ingiuries from the Combat&Tactics manual, magic..)
    - take the resulting number and subtract it to the Tach0.

    the number you get is the AC your PG hits. If this is equal o lower (that is better) that your opponent's you hit, otherwise you miss.

    You simply tell the DM "My PG hits AC 5"
    The DM will tell you if you hit or miss, not the AC of the opponents.

    Your party will still be able to figure out the opponents AC if the combat last enough to let you see some good and bad roll.

    1. However I agree with you that is quite far from being intuitive or even very logical.

    2. It’s interesting that your description is so different from theirs, but on reflection I think it’s because their description is actually aimed at the DM. But theirs does require either the player or the DM to have information that is held by the other one, which was my point: if you’re trying to follow their instructions as a player (which is how I assume things work unless explicitly told otherwise) you need to know the AC to subtract it from your THAC0. If you’re the DM, you need to know the THAC0 to subtract from. Similarly, if you’re trying to follow my much simpler explanation, you need to know the AC to add it to the die roll.

      I’m not entirely convinced that THACO-(d20+modifiers) is a huge improvement myself, but of course different things work for different people. How it tends to work in practice in our games is:
      - roll 1d20
      - apply modifiers
      - Arthur tells you if you hit or not

      On the plus side, despite being very clunky people have managed to play with it for decades and the game became a huge phenomenon, so hey, it could clearly be worse.

    3. "- roll 1d20
      - apply modifiers"

      What we use to do is a further step where you subtract this number from your Thac0, doing this you het the AC you hit. If this is equal or lower than the enemy one (the DM check this) you hit.

      If you are the DM you already knows your player's opponent AC, so you just have to confront the AC number hit by you player and the AC of the opponent.

      I clearly recognize that this is not the best solution they could came out with, but however there's no need to "pass" any wierd information about AC, Thac0 or other stuff between players and DM.

      Possibly my English is not good enough to be clear (not native speaker), but here is an example:

      Let say we have a 5th level warrior, with Thac0 16 (from tables), specialist with the long sowrd that is the weapon he is actually wealding. He also is pretty strong since he has 17 in hi Strenght attribute, that means a +1 to hit and +1 to dmg. Note that I haven't said anything about his opponent since it is not relevant right now.

      The players roll a "9" with his d20. He adds up hi specialization bonus (+1) and his strenght bonus (+1), se he gets to 11. He subtract this number from his Thac0 and finds out that he hits AC 4 (15 -11).

      The DM checks the opponent AC and sees that it is 5. He tells the player that the hit is true and good. Had the opponent AC been 3 or lower the worrior's attack would have been a miss.

    4. Oh, I was forgetting about it. I went back reading the PHB copies I have, the Thac0 expanation is really worse that I remember..

    5. Ah no, your original explanation was fine. It's just we play over roll20 anyway, and Arthur remembers our THAC0, so he can tell from the rolls that come up onscreen whether or not we hit his monster's AC. Which means I never have to do the maths myself.

      Yours is definitely clearer than the PHB option, but it's still a bit more mental arithmetic than I want to do regularly.