Abstract
Problem: How do you design and populate a magic system for a fantasy RPG β choosing colleges, balancing combat vs. utility spells, and supporting both generalist and specialist builds?
Approach: Tim Cain answers a viewer's multi-part question about Arcanum's magic colleges, explaining his design process, goals, and what he'd do differently in future games.
Findings: Magic was designed after setting and story, which properly constrained its scope. Each college was meant to have combat, utility, and "showcase" spells, but the game shipped at alpha quality before balance could happen. The one-point-per-spell system undermined the generalist vs. specialist balance he was aiming for.
Key insight: Simple rules that let complexity emerge β inspired by Will Wright β are the best foundation for magic systems; Tim also points to Tyranny's spell-building system as a model he'd follow in future games.
Design Order: Setting β Story β Systems
Tim emphasizes a deliberate ordering: Arcanum's magic system was designed after the setting (magic vs. technology at odds) and story were established. Both influenced magic's design. He considers this the correct order for RPG development β systems should serve the fiction, not the other way around.
Because setting came first, Tim already knew magic needed to feel old, varied, and support classic fantasy archetypes alongside unconventional ones.
Filling Out the Colleges
The colleges emerged from character archetypes Tim wanted players to be able to build:
- Elemental magic (fire, earth, air, water) β for classic wizards, including summoning elementals
- Necromancy β split into white necromancy (healing, for clerics/priests) and dark necromancy (zombies, Speak with Dead)
- Nature spells β for druids and rangers
- Meta magic β magic that affects other magic, one of Tim's personal favorites
- Temporal magic β slowing, speeding, time stop
Speak with Dead was designed as a quest safety net: if a player killed an important NPC, they could still extract critical information from the corpse, avoiding soft-locks.
The Three-Spell Goal
Tim's design goal was for every college to contain at minimum:
- A combat spell β deals damage or debuffs enemies
- A utility spell β helps with exploration (finding traps, opening locks, invisibility, teleportation, healing)
- A showcase spell β an aspirational, powerful spell (usually the level 5 capstone) that made players want to invest in that college
Tim admits he didn't fully achieve this goal. Some colleges had weak or redundant entries, and there was no time to replace or rebalance them before shipping.
The Stat-Driven Complexity
Magic in Arcanum was deeply interconnected with character stats, achieving the "complex stat-driven game" design pillar:
- Constitution β fatigue and spell resistance
- Intelligence β which spells you could use
- Willpower β fatigue, spell resistance, and which spells you could learn
- Player level factored in
- Equipment choices affected your position on the magic/tech meter, which altered spell effectiveness
Tim considers this layered complexity a genuine success β there was substantial depth even before selecting and casting a spell.
Generalist vs. Specialist Balance
A major design goal was making both build styles viable:
- Generalists who dabbled across many or all colleges
- Specialists who deep-dived into one or two colleges
Tim believes the "one-point fiasco" (where a single character point unlocked a spell) worked against this balance, making it too easy to cherry-pick the best spell from each college rather than committing to a path. He has a dedicated video discussing this issue.
The Shipped Alpha Problem
Tim is blunt: Arcanum shipped as an alpha, not even a beta. Features were complete, but there was no balance pass and no time to evaluate whether specific spells should be replaced. Spells like Harm became obviously dominant because no tuning ever occurred. Given a few more months, spell replacement and balance would have been priorities.
Future Magic Design Philosophy
For future games, Tim identifies two guiding principles:
Will Wright's Rule
Let simple rules produce emergent complexity. Rather than hand-crafting dozens of bespoke spells, define simple per-college rules and let interactions between them create depth. Spells could interact with each other or be built from combinable building blocks.
Tyranny's Spell Building
Tim specifically praises Obsidian's Tyranny for its spell-crafting system, where players combine sigils (expression, accent, enhancement) to construct custom spells. He considers this the direction he'd take for magic in any future game he makes.
References
- Tim Cain. YouTube video. https://www.youtube.com/watch?v=YUTBnZLYSRk