Rambling Thoughts on 3rd Edition Dungeons & Dragons

This is sort of the second portion to my earlier post, but also an examination that an anon in the OSR general asked me to write, so here it is. After all those years playing 3.5, here's what I have to say about the system itself.

First of all, a caveat: This is in no way an “objective” review of 3.5, since we all bring our biases and heaps of experience both good and bad regarding a particular rule set. So, by its very nature, this examination is subjective, but may contain useful insights, or so one hopes.

So 3.5. The d20 mechanic is not in itself flawed. I’m not part of the hordes out there who have tossed the baby out with the bathwater in regard to roleplaying games or Dungeons & Dragons in particular. However, there are aspects of 3.5 that fail to use the d20 mechanic wisely, and creates a mess that is difficult to untangle.

Of great importance to the discussion is how 3.5 is intended to be played vs. how it is played in practice. 3.5 is intended to be played with three rules assumptions in mind:

1.) The DM is building the world based on the community demographics rules (beginning on pg. 135 of the 3.5 DMG and extending to pg. 168), and is not at all attempting to create a believable or living economy for the game world – instead they are creating one that serves the needs of the Players as Adventurers and Heroes. This includes, but is not limited to, the community wealth and GP limits, the generation of leveled NPC characters (those characters with levels in such classes as commoner, aristocrat, warrior, etc.), and the power structures that operate within said communities.

Allow me to elaborate for a moment on the GP Limit. This was always vaguely worded in the 3.0 and 3.5 DMGs, but in the Pathfinder Gamemastery Guide it finally reveals what should have been stated in plain English: On pg. 205 is Table 7-36, detailing the community statistics, and is similar to the one presented in the DMG on pg. 137. There are some differences, namely that Pathfinder establishes a lower overall limit on the availability of expensive items, and spells out that the Base Value is intended for magic items.

In the 3.5 DMG however, the GP Limit serves the same function, but has a clearly higher maximum for items (up to 100,000 gp in a metropolis as opposed to 16,000 gp for one in PF). In 3.5, to determine the maximum amount of “ready cash” a settlement has, one multiplies the GP Limit by 1/10th of the community’s population.

This works out to 125,005,000 gp for the smallest metropolis – an absurd number that would allow PCs to sell pretty much anything in the game world in a metropolis without impacting the metropolis negatively. This number also serves to limit the “maximum number of available items”, meaning that in a default metropolis, it is possible to buy 25,001,000 10' poles, or 2,500,100 greatswords (effectively 100 greatswords per citizen).

It’s what I like to call “sacrificing commonsense on the altar of game balance”. Whose "balance" though, is anyone's guess. Thankfully PF is more reasonable, and provides the “Purchase Limit”, which goes as high as 100,000 gp. It still translates to 2,000 greatswords, which is unlikely, but it is a small enough number that I'm willing to swallow my doubts.

2.) The DM is using the CR system to adjudicate XP and to generate treasure as appropriate for the PCs and their level. This is detailed on pg. 40, and also there is a discussion of it on pg. 48 of the DMG. There are variants for ad hoc experience, and also for freeform experience, but this is not the default assumption, since the rest of the game assumes that you will be using the CR system as written. There is quite a bit of advice regarding tailored encounters (encounters that either are weak to equipment the PCs have or are somewhat resistant to it to provide challenge) and status quo encounters (where monsters are simply placed in the game world, and it’s up to the PCs to make the tactically sound decision to engage them or not).

What this boils down to is that old meme about the 13 encounter treadmill (which is explained on pg. 41 of the DMG itself! - pic related below).

While clearly meant to be a guideline, and a starting point, the rest of 3.5 is geared toward this 13 encounter assumption, which is why prospective DMs are warned throughout the section that if they increase XP rewards, they must also increase treasure rewards lest the PCs become undergeared for their level. This is an important cornerstone to the way in which 3.5 is meant to be played. PCs are expected to have a certain amount of wealth commensurate with their levels, and if they do not, they will find themselves in difficulty against enemies who have defenses they cannot defeat, and attacks they cannot defend against.

 The Dungeon Master is completely and utterly responsible for this balancing act, and if they screw it up, the game suffers. And certain classes suffer more for the mistake.

3.) Wealth by Level. It’s often brought up on various forums and even /tg/ itself as one of the major failings of 3.5 and the reason for that is quite simple: It’s an easy shorthand for why the game’s underlying systems are flawed. Yes, the WBL table on pg. 135 is intended for creating PCs who are higher level than 1st, but it states, and I quote “Table 5-1 Character Wealth by Level is based on the average treasures found in average encounters compared with the experience points earned in those encounters.”

Why exactly have I called out the latter part of the sentence? Because this is what DMs get wrong the majority of the time, because they either forget about it, or ignore it, and thus are not playing as intended. They don’t follow the CR system as presented, they either give too much treasure or too little treasure compared to the XP they’re handing out, and suddenly all hell breaks loose.

Take for example a player invited to a game in progress, who is instructed to make a character of a similar level with the existing party. If less treasure, but normal XP has been handed out over the course of the game, the new player will have more treasure for their XP than the rest of the party, and therefore be able to afford better goodies in the form of hand-picked magic items. On the other hand, a player invited to a game where more treasure, but normal XP have been handed out will actually begin play at a disadvantage if the WBL table is followed. Already, things begin to break down, even assuming all classes are equal in power level - which is demonstrably false.

So, with these assumptions in mind, we should examine the classes.

Let’s compare some classes and see where they stand by putting a (mostly) arbitrary point tally to their abilities.

HD

• d12 = 5 points
• d10 = 4 points
• d8 = 3 points
• d6 = 2 points
• d4 = 1 points

Attack Bonus

• Good = 3 points  
• Average = 2 points
• Poor = 1 points

Saving Throws

• Good = 2 points
• Poor = 1 points

Skills

• Good (8x) = 4 points
• Decent (6x) = 3 points
• Average (4x) = 2 points
• Poor (2x) = 1 points

Feats/Class Features

• 2 points apiece

Spells

• 2 points per slot*

Using this rather rough and not at all scientifically accurate gauge of class power, I have written out several classes in the 3.5 PHB and DMG for you to compare. All classes are being judged on their sum total of abilities at level 20.

PC Classes

Fighter

• d10 HD (4)
• Good BAB (3)
• 2x Poor Saves (2)
• 1x Good Save (2)
• Poor Skills (1)
• 11 feats+1 class feature (ability to take Weapon Specialization)+proficiencies (counting as feats, so Simple, Martial, Light, Medium, Heavy Armor, and Shields and Tower Shields) total = 19 (38) 

Final Point Tally = 50 points

Monk

• d8 HD = (3)
• Average BAB (2)
• 3x Good Saves (6)
• Average Skills (2)
• 4x Feats+roughly 19 class features (46)

Final Point Tally = 59 points

Ranger

• d8 HD (3)
• Good BAB (3)
• 1x Poor Saves (1)
• 2x Good Save (4)
• Decent Skills (6)
• ~17x class features/feats (34)
• 12x spell slots (24)

Final Point Tally = 75 points

Wizard

• d4 HD (1)
• Poor BAB (1)
• 2x Poor Saves (2)
• 1x Good Save (2)
• Poor Skills (1)
• 5x feats+1x class feature (12)
• 40x spell slots (80)

Final Point Tally = 99 points

NPC Classes

Adept

• d6 HD (2)
• Poor BAB (1)
• 2x Poor Saves (2)
• 1x Good Save (2)
• Poor Skills (1)
• One class feature, also has proficiency in simple weapons (counted as a feat) (4)
• 17x spell slots (34)

Final Point Tally = 46 points

Aristocrat

• d8 HD (3)
• Average BAB (2)
• 2x Poor Saves (2)
• 1x Good Save (2)
• Average Skills (2)
• Zero class features, but has proficiency in simple and martial weapons and all armor and shields (counted as feats) (14)

Final Point Tally = 25 points

Commoner

• d4 HD (1)
• Poor BAB (1)
• 3x Poor Saves (3)
• Poor Skills (1)
• Zero class features

Final Point Tally = 6 points

Expert

• d6 HD (2)
• Average BAB (2)
• 2x Poor Saves (2)
• 1x Good Save (2)
• Decent Skills (3)
• Zero class features, but has proficiency in simple weapons and light armor (counted as feats) (4)

Final Point Tally = 15 points

  

*I originally had calculated spells as 1 point per spell slot, so (spell slots)*(spell level), but I felt that it was too messy and that comparing spell slots to feats and treating spells as powerful single use "feats" was much more useful.

Obviously, this methodology is flawed. Not all feats, and not all spells are created equal. Not all spell or skill lists are equal. However, this was mostly to demonstrate that while the fighter has many permanent abilities, and freedom to choose those abilities, by 20th level he has fewer tricks than a wizard of equal level, even though the wizard may “lose” abilities as the adventure goes on, those abilities are so powerful that they can often shut down an entire encounter with a single spell, which is arguably more powerful, and more flexible than the fighter’s permanent abilities that are never exhausted.

And this is before factoring in magic items.

The CR System, Monsters, and Fighters

Something that has troubled me for a long time with 3.5 was the book Savage Species. And this is not because it allows monstrous PCs or unbalances the game. No, it actually troubles me because it revealed a fourth assumption that is bound up in the CR system. On pg. 13 of Savage Species, there is a small discussion about how to adjudicate level adjustment for monsters, and the basis for comparison is the fighter. I grant that I may be taking it out of context, but the main thrust of that paragraph is that if an ogre with a level of fighter is superior to playing an 8th level fighter, than the ogre needs to be given a bigger level adjustment to bring it into parity with the fighter class.

Moreover, if you examine the monsters in the Monster Manual, you’ll notice that ACs tend to keep up with the Fighter’s base attack bonus. I’ll use giants as an example:

Hill giants have a CR of 7 and an AC of 20. Frost giants have a CR of 9 and an AC of 21. Fire giants have a CR of 10 and an AC of 23.

A 7th level fighter (using whichever point-buy method you like, shooting for 18 Str, half-orc race, and all ability increases into Str) would have an attack bonus of 7+5+2 (class+Str+magic weapon) = +14. Which is a pretty reasonable chance of hitting an AC of 20. However, a single hill giant might easily murder the lone fighter, seeing has how it has +16 to attack and a greatclub that does 2d8+10 damage, and a pile of hit points besides (102). Obviously, the hill giant is not meant to challenge a fighter one on one. Its challenge rating is supposed to be a “challenging” encounter for a party of four 7th level characters.

I suspect (but cannot prove) that the Monster Manual was purposely balanced against the fighter class. If this is true, then they balanced the monsters with the CR system against one of the (arguably) weakest classes in the game, and as a result, when you allow a strong class like the wizard to run rampant with more XP and more treasure than is normal, they utterly dominate the game, and the fighter is relegated to playing a game of catch-up by buying expensive magic items in an attempt to replicate some of the abilities wizards enjoy almost as a matter of course through class features.

Obviously, the Tier System exists, and is a community attempt at patching 3.5’s system to be more fair. Don’t allow classes below or above certain Tiers, say the supporters. However, I maintain that it doesn’t change the underlying fact that the game is built on a CR system that is calculated against the assumption of a weak class being in the game, which in turn informs the treasure generation, and in turn affects the availability of magic items in the game. Moreover, the Tier System is not “playing as intended”, it is the premier example of how the game is played in practice.

A Conclusion & Ruminations on OSR

Ultimately, if one has fun with 3.5, there’s nothing wrong with that, and if you approach the game understanding its flaws you can manage to do more or less anything with it. My contention is that you might as well play something that better suits your play style than a system you actively have to cut apart and fix up as best you can. Which is primarily the reason I play OSR games – and I am not for moment suggesting that OSR is the “right” or “proper” way to play RPGs. It simply suits my tastes and my needs as both a player and a DM.

I DMed 3rd edition in its varieties and flavors for about 14 years. I am well aware of what makes it tick, and I’ve come away weary of the minutia of builds, and balance, and ivory tower game design (the now infamous apology by Monte Cook wherein he detailed how he and the other 3rd edition designers created “trap options” that would encourage “system mastery” like Timmy Cards in Magic The Gathering - see pic related below).

It’s not enough that the fighter is weak, and merely stays that way as he levels, but he also is given a plethora of options with little to no guidance, and three of those options are Endurance, Run, and Toughness. None of them make the fighter good at his particular job in the party, and very few experienced players would take those feats without good reason (a DM who springs night encounters regularly, for example, would make Endurance reasonably worth taking). Those reasons though, are incredibly difficult to predict at the outset.

Tellingly, the fighter's role isn't even defined due to the seemingly large, but ultimately narrow choices the fighter is presented with.

One of my gripes with 3.5 is that it gates off portions of the rules, and while it seems like a good idea to have Power Attack, Cleave, and Great Cleave as discrete powers, it ultimately harms the fighter, and he is made irrelevant by spells and spell-like abilities that can either emulate a feat or are simply superior to those feats, never mind that the spell can only be used a handful of times before it becomes unavailable. What matter when a wizard can provide safe harbor with spells, and the party can rest without molestation?

Skills too present problems, and the skill synergy system in 3.5 was supposed to help the low skill point classes, but ultimately benefits the classes with loads of skill points to push around and maximize their synergy bonuses.

I have so much more to say on this, but I’ve already run five pages according to my word processor program. I also have places to be and things to do, but hopefully this provides some insight into why someone might not be fond of 3rd edition Dungeons & Dragons.