What I miss about D&D 3.5.
The third (and a half) edition of Dungeons and Dragons was my first real roleplaying game system. I’d acquired a great deal of the Second Edition books and read them over, and I’d gone through the chronicles of The Adventurers in great detail, but the first time I put dice to tabletop was for a 3.5 game.
The basic resolution system of D&D 3.5 was the 20-sided dice. You rolled a d20, added the result to your bonus, and compared that to your target number. A strong, hale warrior with an excellently-crafted sword but little combat experience might have an attack bonus of +5 or so (+1 for base attack bonus based on level, +3 for their Strength bonus, and +1 for the masterwork quality of their weapon). A knight in full plate armor with a shield would have an Armor Class of 20. To determine if the callow youth could strike the knight, you roll a d20, add 5, and checked if the result was 20 or more.
The mechanical universe of D&D 3.5 starts with a few basic principles, and then applies them more-or-less evenhandedly. When characters overcome challenges, they get experience points, and when they get enough experience points, they gain a level. Higher-level characters are all-around better than lower-level characters. But monsters also have levels, more or less. The ogre that you fight has 4 levels of Giant (which are called racial hit dice, because they’re something innate to all adult ogres and not something they need to adventure for).
Now, this segues into another facet of the simulationism of 3.5E. There are no default bounds to where your numbers can end up. Take two characters, a fighter and a rogue. The Fighter uses the Good advancement for his attack bonuses, making them increase at one point per level gained. He’s also got a Poor reflex save, gaining only one-third his level on his rolls to avoid fireballs, entanglement in spider webs, and similar hazards. The rogue has an Average attack bonus and a Good reflex save, giving him .75 his level to attacks, and two plus half his level to Reflex saves.
Things don’t stop there, of course. The fighter, wanting to fight better, will be scrimping and saving to get the best magical arms and armor he can get, and will train in Feats that increase his attack and damage. Meanwhile, the rogue will be investing in gear to let him survive traps and hazards, and similarly invest in Feats. In this paradim, a Fighter can easily expect an attack bonus in the mid-thirties by maximum level, while the rogue will be looking at a bonus in the low 20s. Conversely, the rogue can easily hit a Reflex bonus in the mid-twenties by max level, while a fighter who doesn’t spend additional resources will have +6.
You can clearly see what will happen here. Anything the fighter is even moderately challenged to hit in combat is basically safe from the rogue’s attacks. And any Reflex-based attack that makes the rogue sweat even a tiny bit will hit the fighter every time.
Of course, it gets much, much worse. At least there is a chance that the numbers can line up and you can roll really well when it’s just a 90% chance of failure on your every action. D&D also embraces the fantastical and the magical. It’s a world where dragons soar across the sky, where rogues with rings of invisibility sneak through enemy lines, where music can soothe the savage beast (if the musician is a Bard with the right abilities and the beast doesn’t have that high of a Will save). If a dragon faces a party of adventurers, and that party can’t fly, then unless that party can ground the dragon or make some extremely effective ranged attacks, that party is dead. And that rogue with a ring of invisibility will win at any stealth challenge. It gets to the point where all of the higher-level monsters have gigantic multi-page stat blocks to cover all of their baseline immunities, so that players can’t pull out one weird low-level tactic to defeat a greater demon.
I miss 3.5 D&D because I’ve gotten what I can from it. I am an old hand at the rules now, and the failure of the system to provide a fun, scaling challenge for a wide variety of adventurers who don’t know the unwritten optimization curve sweet spot grates on me. But I miss learning that curve, and the excitement of finding I could combine obscure ability A with known-but-rarely-explored-spell B with specific world element C and produce amazing results.
D&D 3.5 is not broken, but it is trivially easy to break. It’s so easy to break that you can end up breaking the system without meaning to, just by choosing one class over another, or being richer or poorer in a particular class of magic item. It often breaks spectacularly, especially in the hands of players who know what they’re doing, and leads to Bards pulling pied piper maneuver on entire armies while the Fighter abuses the damage multiplier bonuses to deliver strikes which can sunder mountains, and the party wizard has just told all of reality to shut up and go away. It can break with all of the glitter and excitement of a crystal vase stuffed with a dozen fireworks. But for all its glitter, once you break the system, it remains broken.