What Third Edition gets right, part two

We’re looking at what each Edition of D&D gets right to see if they can all be mashed together to make the best darned edition of D&D, ever. This isn’t quite how the next edition of D&D is expected to work, but it’s a fun experiment. From our understanding so far the next edition of D&D will take a modular approach where the GM, group or even individual players can make their characters as complex or as simple as they desire. I’ll be blogging about relative complexity later this week.

We have already looked at Fourth Edition (parts one and two) and last time we covered the wonder that is the Open Game License. This time around let’s take a closer look at the Third Edition D&D rules themselves.

Starting power level

On a scale from 1 to 20 where your typical normal Peasant is a 1, the average 1st level Third Edition character hardly scrapes a 2. In contrast, a 1st level Fourth Edition PC would be a 4 or 5.

Third Edition D&D gets it right. At the very start of their careers, your heroes are barely differentiated from the common rabble. The only things which put them a step above their peers are higher-than-average stats (4d6 drop lowest > 3d6) and decent equipment. The PCs have barely enough hit points to take two solid hits, and need to rely on wits and teamwork rather than big flashy powers in order to survive. The world is dangerous because it’s dangerous, not because the flavour text says it is so.

Of course, there’s nothing to stop your game starting at a higher level. If you want your heroes to really be heroes (and on a par with their 4e counterparts) then start the campaign around 4th level and you’re there. The game needs levels 1-3 there though if you want that core D&D “starting from nothing, risking everything” experience. It’s one thing which 4e D&D sorely lacks, and Third Edition D&D gets spot on.

Later supplements and splatbooks ramped up the overall power level of the game to the point where even low-level PCs could be walking tanks, and this is one area where 3e fell down. I hope that the next edition of D&D does a better job of keeping a lid on the power curve than 3e did. 4e was pretty admirable in this respect, but even in that edition some power creep did manage to sneak in.

Vancian Magic

Is D&D really still D&D without Vancian Magic? Maybe, maybe not. I do think that D&D loses something when it’s not there, and there should be room in the rules for a Wizard class that has to commit their spell to memory and lose them when cast.

One of the strengths of Third Edition D&D is that it kept conceptually close to its roots. I love spells with Verbal, Somatic and Material components and some of my best gaming sessions began with the Wizard heading off on his own to search for rare spell components, then promptly needed rescuing by the rest of the party.

In another session the Wizard was cursed to cast a random memorized spell whenever he opened his mouth. That’s role-playing gold right there, and all thanks to the magic system in Third Edition D&D. Try that with Fourth Edition and you’ll just get blank looks.

What I would like to see in the next edition of D&D is the return of a Vancian Wizard class alongside an inate magic-user (Sorcerer?) who works in a similar way to a Fourth Edition class. This lets the GM choose which style of magic suits their campaign setting, and the player freedom to choose their preferred style of play. I feel that the 3e Sorcerer almost got it right, but the reliance on Spells per Day and a Spells Known list did little to sufficiently separate it from the Wizard class.

If I was designing the magic system for the next edition of D&D there would be a Vancian Wizard who could memorize a fixed number and level of spells per day, and other magic-using classes (Sorcerer, Cleric, Bard, Druid) would use spell points. These would be equal to their key stat (CHA for Sorcerer and Bard, WIS for Cleric and Druid) plus their class level. Spells cost 1, 3, 5, 7, 9, 13, 15, etc according to spell level but spells above your class level cost double. Only level 1 but want to cast that level 3 Fireball? That’s 10 points (level 3 spells cost 5 points, doubled because it’s above your level).

Spell points reset back to full points each day at times depending on the class (A Cleric of the Sun God’s points reset at dawn, for example) and can be affected by location, magic items, following (or disobeying) the laws of your god, etc. If your Sun God cleric fails to pray at dawn he only has half his spell points for that day, or a Druid with metal on his person is penalized by the item’s weight (don’t try wearing a metal breastplate if you’re a Druid – that’s -30 spell points!). Cantrips (0-level spells) cost no spell points provided your character has at least 1 spell point remaining.

This means that a 1st level Sorcerer with CHA 15 would have 16 spell points so could cast 16 1st level spells, or two 2nd level spells (presuming he knew any) and three 1st level spells, or one 3rd level spell and 6 1st level spells, or any mix of the above. At 2nd level he gains one more spell point but 2nd level spells are no longer double-cost so he could now cast five of them and two 1st level spells, or be free to mix them however he wants. Meantime the Vancian Wizard doesn’t need to mess with spell points at all but instead chooses his spells from his spellbook in advance.

Buy hey, that’s just me. What do I know?

No Dragonborn or Tieflings

I quite like 4e’s Dragonborn, but not as a core race thankyouverymuch. The core races in the game should be the ones which are likely to be present in the majority of campaign settings. As D&D is a fantasy game first and foremost so that means Humans, Elves, Half-elves, Dwarves, Halflings and Half-Orcs (and Gnomes, if you insist). Having Dragonborn and Tieflings as core races in Fourth Edition meant revising existing settings such as the Forgotten Realms to a massive degree just to make them fit. Thanks, but no.

Save these races for Player’s Handbook 2, or whatever. Core races my ass.

Talking of races……

Savage Species

I consider Savage Species to be one of the most under-rated supplements ever released for any edition of D&D.

This supplement allowed you to play a Kuo-Toa. Or a Grimlock. Or an Ogre Mage. Or a Dragon. Or Flesh Golem. Or Zombie. Or any other creature at all from the whole of the Monster Manual and beyond. It included rules for creating Anthropomorphic Races (Catfolk ftw!) as well as turning powerful monsters into full classes with level advancement.

Quite simply, it blew my mind. Savage Species is D&D, wide open. I planned (but for several reasons never got around to running) a session where the PCs were Skeletons under the command of a particularly incompetent Necromancer (imagine The Office but with undead, and you’re there), and ran one scenario where the players were all Wizard’s Familiars trying to rescue their captive owners. In another session the players created the monsters inhabiting the dungeon and I ran a team of adventurers as NPCs trying to “cleanse” their home (I failed).

Much as Savage Species was glorious, ambitious and frankly downright awesome, it was not without flaws.

The problem was not so much with the supplement itself but with the whole Level Adjustment/Effective Combat Level thing. The rules for those didn’t work. A Level Adjustment was just nowhere near equivalent to the level it was supposed to be adjusting. You couldn’t, for example, play an Ogre Mage (5d8 HD, +7 Level Adjustment) alongside a party of 12th level heroes and have a hope in hell’s chance of keeping up. In the end we just ignored Level Adjustment completely. Sure, your Ogre Mage is going to be more powerful than your 5th level comrades for a while, but it will all balance out in the end.

What Third Edition D&D, and Savage Species in particular, did was help us to think of monsters as having classes. The PCs could face off against a group of Orc Rangers or a White Dragon Necromancer, and that just rocked. Sure, the implementation was time consuming – creating custom monsters and adding class levels took ages – but the end result was often very satisfying, at least for the five rounds they lasted in combat.

I would love to see a return to something like this in the next edition of D&D, but in a simplified and less time consuming form. Fourth Edition D&D tried (and partially succceeded) with the Class Templates in the DMG but like so many things in that edition of the game it wasn’t well followed through when new classes were released, or the material was hidden behind the D&D Insider paywall. Did they ever release Psionic class Templates, for example? D&D Insider lacks a good enough index (or decent organization at all, for that matter) for me to find out.

What would be interesting would be open up the whole “a race as a class” thing. Your hero could be a Level 1 Elf just like in good ol’ Classic D&D, but also take levels in a profession-based class at later levels. Your Level 3 Elf/Level 2 Ranger could stand alongside a Human-1/Fighter-4  as an equal. Likewise, a Level 20 Goblin would be a God-King among Goblins, and a true foe to be feared.

Imagine a version of D&D where every monster had a sliding class level range so the GM could easily scale each and every monster encounter, and the players could even mix-and-match races to create unique race combinations. Taking Orc-1/Human-1 gives you the archetypal Half-Orc, but what about a Gnome-1/Goblin-1? Gnoblin, anyone? It’s like a Half-Orc in miniature!

Seriously though, monsters with levels rocks both sides of the gaming table, hard. It will need to be implemented a darned site better (and by better, I mean nowhere so damned time consuming) than Third Edition’s system, but I believe it can, and should, be done.

Phew. We’re going to need a Part Three for this. After that we’ll take a sidestep and look at Pathfinder before moving on to AD&D and end with the grand-daddy of them all.

Till next time!

You may also like...

21 Responses

  1. Siskoid says:

    I’m loving this series, Greywulf, and I can’t wait for your posts on 2nd and 1st editions, which are the ones I started out with and know best. I’m thinking that any D&D in my future may well make use of your observation as a starting point for house rules!

  2. Jennifer says:


    My favorite character of all time was a pixie warmage. I had to balance her carefully to ensure that she would be able to keep up with the human wizards in the party even with her lower class levels, but it was so worth it. And what she lacked in power, she made up in surviveability – tiny bonus to AC, plus permanent flying and improved invisibility? Yes please! To this day, I am begging my group for a chance to run her again.

    • greywulf says:

      I created a Pixie Warlord in Fourth Edition D&D a few weeks back and I’m dying to give him a playtest. Good to know there’s another Pixie fan out there! :)

      • Jennifer says:

        I just like things weird. ;)

        That said, have you looked at the SW Saga rule set at all? We just started playing it (again) and it’s very interesting to see how it took stuff from both 3.5 and 4e, and topped it off with some stuff that was completely unique.

  3. Quirky DM says:

    Savage Species was awesome and became one of our most used books in 4E.

    For your sorcerer idea and spell points, maybe you need to REALLY take a page from 4E and give them “mana surges”- aka, healing surges for spell points. That way, you can limit their full potential in any given encounter, but still let them go at it all day long without worrying about conserving their spells for the right encounter.

  4. Jonathan says:

    Savage species, yes. I must respectfully disagree with you about the ‘rightness’ of 3E starting power level. At least, it wasn’t right for everybody and it certainly wasn’t for me. I do not have fond memories of first level as a caster. My most vivid memories involved my character bleeding on the floor completely out of spells and thinking “jeez, this makes no sense. If I’m this weak, I would never have left wizard school!”

    But then, I’m a 4E guy… who likes weird races in the core book (because then my DM can’t tell me that they aren’t allowed easily (because this happened to me far too many times in 3e with DMs telling me that my favorite race or class was badwrongfun)) and who likes my character feeling what I’d call ‘reasonably competent,’ Or ‘at least smart enough to not leave training before being able to beat goblins.’ ;) All kidding aside, I think my biggest problem was never with 3E or 3.5 itself, but rather with trying to play it at the wrong tables. I just want the chance to play the Warlock!

    While I didn’t care for vancian magic (see: favorite class = warlock), I’d like to see it supported as an option for the people who did like it.

    While I hated level adjustments (and would say that race design was something 4E also did right) I do have fond memories of Savage Species. I still wish that there had been something like it for 4th Edition. It’s strange and wonderful and had rules for playing a dragon or a deva or anything else I could want. Good times. Easily my favorite book ever.

    • greywulf says:

      Lol! I quite like the mental image of a Wizard bleeding on the floor and wishing he’d never left wizard school :)

      Some gamers criticize 3e for being too low powered at 1st level, and some say that 4e is too high powered. Rather than trying to find a middle ground I think the solution is to recognize that different groups play the game in different ways and deliver rules that cater for both. Provided solid character generation rules which allow the PC to begin play as a wet behind the ears farmer’s son (with his father’s rusty longsword), an experienced veteran of the wars, or anything in between. Put the decision where along the scale the campaign is set, and it’s a win all round.

      Can they do that? I think so, yes. Wait until I write about Mutants & Masterminds and Levers later this week…..

      • Jonathan says:

        I like that idea. I really liked Chatty’s 0 level rules for 4th edition; I’d kind of like to see something like that as an official part of the core or an early book.

        M&M being my favorite game I’m not playing (inevitably I wish I was), I look forward to reading that post on it later. :)

  5. Geoff Smith says:

    I think there were race levels in 3e. Called them Paragons, IIRC.

    I’ll hit up the negatives, as I guess that’s my thing here. If I were much more articulate and had actually looked at the rules in a while I’d try to be your foil, but I’m just not that good :) I’ll just be a nit-picker :)

    3rd ed had, as it’s greatest flaw, power creep. It just got ridiculous. Normal high level PCs pretty much broke the game, and with the addition of the splats it just got out of control quick. It also had a huge issue that it could be min-maxed really well, and people who were really, really good at it could produce PCs that were simply in another league. This could discourage and frustrate those who didn’t min-max because they didn’t know about it or weren’t any good at it, and those who tried to make interesting thematic/RP PCs that didn’t have sheer power as their number one design decision. This is one of those things I thought 4e did much better, or at least tried to.

    I was also not a huge fan of the reliance of magic items on the system. Trying to play a low/no-magic world really made the uneven leveling experience much worse. I did like that there was an ability to make magic items, and to make pretty much anything you could think of. This was good, as it was very difficult to make magic items in earlier additions. What was bad was the prevalence of magic items that added so much combat advantage. I would almost like to have magic items add more color, or do things like change damage types (a fire sword did just as much damage as a normal sword, but did it with fire damage either in part or entirely). They did add some of the ability to, say, level a single weapon with (I believe it was called) Heirloom weapons? This would be good to be part of a core system. The sword-using class got better independently of the weapon they were using, and would allow them to potentially use the same sword their entire career.

    Caster differentiation.
    4e could have potentially made this much better, but wound up going to opposite direction where a lot of the classes just felt really similar, just with different ‘power sources’. In 3rd there was little difference between how the different kinds of casters (baring something like Psionics with points, or the warlock), and all the casters really felt similar. Sorc, Cleric, Mage, bard all were basically the same class with different spell lists. They could have gone further and had a unified caster class, I guess, though with the flexibility of the d20 class system, you really could design custom classes fairly easily. This is one way to go with it. The one I would prefer, however, is that they are very different. Clerics, mages and druids should play very differently, and have their own mechanics. Sometimes I think this is overly complicated, but I think the slight differences they threw in are almost worse. Either throw the casters together and differentiate them based on spell selection, or make them different classes all together.

    Character creation time. I mentioned before that the new system should have a ‘quick start’ ability, so just sit down and start playing without needing to spend a lot of time building a PC. This just got worse if you were building a high level PC/NPC.

    I guess the last thing I’ll harp on here is the exp system. This was a big departure from 1st and 2nd ed, in that everyone needed the same amount of XP to level. I thought this was weird at the time, since it really should take more xp for the different classes to level (they tried to make this reasonable by trying to even out the leveling increase, giving fighters as much when they leveled as mages, but it just never really worked out that way, especially for front-loaded classes like Paladins and Rangers). Where this shines is multiclassing (one of the big places where 4th really falls down, IMO) and facilitates the creation of pretty much any class/character concept you can imagine! I can also understand that the designers wanted the party to advance in tandem, but wanted to hold on to the XP sacred cow. I think they should dump it all together and have a party XP/advancement system with individual XP an optional thing, or return to 1st/2nd ed and have each class progress at its own rate.

    • greywulf says:

      I’m nodding along with pretty much everything there apart from the unified XP system. I think that was a huge improvement over AD&D’s differing XP tables, for precisely the reason you give: simple multi-classing. Given the choice I prefer Fourth Edition’s multi-class options over Third Edition’s, but that’s just me.

      When I get onto Pathfinder I will be looking at XP tables in more detail. I do like how Pathfinder offers a choice of slow, normal, or fast XP advancement depending on how you want to play the game. That’s something the next edition of D&D definitely needs to “steal” back :)

  6. Geoff Smith says:

    Skills! Forgot about skills! While an improvement over NWPs from 2nd, they just had some of the same problems. Mostly that some were vastly overpowered, and some had little apparent use whatsoever. I might have split them into functional and color skills, and given some points in each. A class like Expert might get all color skills. What they missed out on was really expanding skills directly into combat. It could have been interesting to see leveling only give you skill points which you then paid to increase the abilities you wanted (with similar level restrictions). Again, a classless system would have been too much of a departure, and I can see that. It also could have been very complicated…

  7. Eric says:

    I really like your point on the sliding levels for monsters. It just has so much potential for both players and dms. I was never a fan of race as class in general, but you are spot on about that.

  8. Alphastream says:

    I really disliked 3E’s EL system and the work required to create monsters that could challenge the party. The rules were so stringent as to make it very difficult for the DM. You were shackled. Another problem was spellcasting, which added amazing complexity to any spellcasting monster. I waited an entire year to DM for Living Greyhawk because I was afraid to run spellcasting monsters… seeing those lists of spells was daunting, and it was easy to flounder as you flipped through books at the table, mid-combat.

    In 4E the templates had some of those same problems. And they are unnecessary. Far easier to just add a monster theme power that communicates role or flavor. Without the shackles you were free to just add powers or modify. Need that orc to be a warlord? Give them a power that moves all the orcs in a burst around it, then a minor recharge power that lets another orc make a basic. Easy, simple, fun.

    Vancian magic has a really cool story to it. I missed that when I first looked at 4E. But I don’t miss being forced to bring books to the gaming table, nor to have so many table rulings on spells, nor wizards using darts and crossbows. Worse, the thrice-cursed Pearls of Power that let wizards regain spells really broke the game and hurt Sorcerers as a class. I think the concepts of Vancian magic and 4E powers can work together, however. I’m eager to see how D&D Next will do this.

  9. OgreDM says:

    Your “Race as Class” idea is one of the central themes of the d20 OGL game “Legend” from the rule of cool guys – http://www.ruleofcool.com
    They call it “Tracks” and the whole class system is built around it.
    I think it’s a pretty neat idea. I’m looking forward to introducing my players to the system.

    Good series of posts, looking forward to the finish!

  10. Chris Burgos says:

    One point of contention I would have with you is the Race issue. The new races were a step forward, not backward in the case of 4th edition. 4th Edition setup new classes and therefore needed new races to go along with those classes. 5th edition needs to continue with that. DnD is not a game that should compare edition A to B, it needs to continue to innovate and you cant do that by pulling old features into a new edition. DnD 5e can only hope to gain new customers, too many 3.5 players are stuck in 3.5 land (pathfinder) and they wont likely give that up. With good innovations on a strong system, 5E will begin to capture more eyes and ears. It will need strong marketing and the OGL but a return to the old is not a smart move.

    • Siskoid says:

      Chris, you make a good point about how each successive edition is diluting the customer base. I think it has something to do with how quickly they’re coming now. I happily jumped on the 2E bandwagon back in the day to refresh my 1E stuff, but really haven’t since. At some point, you’re gonna have YOUR D&D and that’ll be that, especially if you invested a lot of money in that edition’s products.

      • migo says:

        I had a similar feeling going from 2e to 3e. I started in the mid 90s, but the game was 11 years old (22 considering the minimal differences between 1e and 2e) and it was time for a new edition. 3.5 came 3 years later and unfortunately most people switched over, rendering my earlier 3e purchases relatively useless. 4e coming out 5 years after that was still on the short side, and Essentials getting released 2 years after that made me even more apprehensive. I was thinking there might be another 4 years after Essentials given how things seemed to be going, but we’re looking at less.

        That’s the main thing that has me apprehensive about 5e. Since Hasbro got their hands on D&D, the editions have been coming out way too quickly. The next edition needs to come out when people are tired of the previous edition and are asking for a new one. I’m not sure that has actually happened with 4e (or even 3.5 – I just never liked it rather than getting tired of it, I’m not sure if people who actually liked it are feeling fatigue yet).

        The quick re-release of the edition is probably what third edition got most wrong.

    • Jonathan says:

      Agreed. As a rule, I love new races, and I really hope “D&DNext” isn’t a step backwards. If it’s all inclusive (somehow), great, but I don’t want to see them lose the innovative improvements 4E actually had. Ever. The whole reason I’m not playing Pathfinder (aside from the “wrong people” problem) is that while 3.5 was where I started playing, I would not willingly go back.

      … Though I had great fun with a 0th edition retro-clone once …

  11. Oz says:

    Glad I’m not the only one who had issues with how they did the races in 4E. It was like WotC was more concerned about using new races to establish their IP rather than give the players the core races they expected. I was also put off by how the original classes were spread over 3 PHBs with a proliferation of classes filling up the space.

  12. Lugh says:

    You make some really excellent points.

    One of the things I loved most about 3e was that you could choose your starting level. We tried doing that several times in 2e, and it was an absolute mess every single time. You want to play zero to hero? Start at 1st. You want to play a more even game? Start at 4th and use something akin to the E6 variant (capping levels at a certain point and only gaining new feats thereafter).

    I loathe the Vancian magic system with a fiery passion. So I’ll pass on that topic.

    The EL/LA thing was a good idea with bad execution (like many things in 3e, honestly). I really do think that they deliberately hosed the power levels of the LA races to strongly encourage people to play the core races. Basically, you pay a “snowflake tax”. We’ve found locally that reducing LA by 1 across the board and giving a slight boost to what you get with racial hit dice evens things out nicely without too much work.

  13. It’s a D&D trope but I can’t say I miss Vancian magic one jot. And don’t daily powers kind of do that anyway?

Leave a Reply