Arcanum: Postmortem Part 2 β€” The Original Attribute System

Abstract

Problem: How did Arcanum's attribute system evolve from its original design to what shipped, and what experimental ideas were left on the cutting room floor?

Approach: Tim Cain walks through the original 12-attribute, three-column design document for Arcanum, explaining why it was cut down to 8 attributes across two columns, and what problems arose during development.

Findings: The original system featured a third "Spiritual" column (Passion, Zeal, Faith, Temperament) alongside Physical and Mental, organized into Power/Speed/Resistance/Appearance rows. It was scrapped due to balancing nightmares, tester confusion, and the fundamental problem that more attributes means more attributes players don't care about.

Key insight: Attribute systems built from a clean foundational matrix (categories Γ— nature) produce more coherent designs than ad-hoc "let's throw in Luck" approaches β€” but elegance on paper doesn't guarantee playability, and sometimes fewer, well-connected attributes serve the game better.

The Original 12-Attribute Design

Arcanum shipped with 8 attributes in two columns (Physical and Mental) across four rows. But the original design had 12 attributes in three columns: Physical, Mental, and Spiritual.

The four rows were renamed too β€” instead of Power/Resistance/Prowess/Appearance, the original used Power/Speed/Resistance/Appearance:

Physical Mental Spiritual
Power Strength Intelligence Passion
Speed Dexterity Cleverness Zeal
Resistance Constitution Willpower Faith
Appearance Beauty Charisma Temperament

"Cleverness" was originally called "Witty" β€” how fast you could think on your feet. Tim tried to make an acronym from the initials but never succeeded.

The Spiritual Column

The Spiritual attributes represented the non-mental parts of your inner being β€” how emotional, deep, or calm you were. Temperament, for instance, has nothing to do with intelligence: some people are just calm and some aren't. Tim wanted to capture that dimension of a character.

Easter egg: Some of these spiritual attributes were later snuck into The Outer Worlds, though they didn't survive testing there either.

Why It Was Cut

Too Many Derived Statistics

With 12 base attributes, the number of derived stats exploded β€” and they were unbalanced. Strength and Dexterity naturally connected to far more derived stats than something like Zeal.

Skill Distribution Was Impossible

Arcanum shipped with 16 skills across Combat, Social, Thieving, and Tech. Distributing 16 skills across 12 attributes in a way that felt balanced and intuitive proved unworkable.

Perk Gating Became Confusing

With 12 attributes to potentially gate perks behind, designers constantly argued: should this perk require Intelligence or Willpower? Passion or Zeal? The distinctions were too fine-grained for practical game design.

Tim notes he's since changed his mind on attribute-gated perks entirely β€” he now dislikes gating perks behind stats players can't change during gameplay.

Players Min-Maxed Ruthlessly

The more attributes in a system, the more attributes exist for players to dump. Testers would pick a few favorites, pump them to maximum, and leave the rest at rock bottom β€” creating absurd characters who were impossibly strong but had zero endurance, or brilliant but completely un-clever.

Tester Confusion

"Faith" and "Spirit" sounded religious even though Tim didn't intend them that way. Passion and Zeal were constantly confused and conflated β€” testers couldn't articulate the difference.

The Perception Problem

The original system had no Perception attribute, and adding one proved surprisingly hard. Physical perception splits into sight (ranged attacks, spotting hidden things) and sound (hearing traps, detecting enemies) β€” two distinct senses. Mental perception was something like "observation" β€” not just seeing a differently-colored book, but realizing it might be a hidden switch. And spiritual perception? Tim had no idea what that would even mean without introducing psychic abilities.

The solution was to collapse back to Physical and Mental columns, slot Perception in across from Dexterity as the mental "Prowess" stat, and accept that it was an imperfect fit. Tim admits he never really liked it.

Escalating Attribute Costs

Tim originally designed attribute costs to escalate β€” higher values cost more points. This elegantly solved the min-maxing problem: being great at one thing and terrible at others cost the same as being good at several things. It also worked well with Arcanum's auto-perks at attribute level 20. But players still abused it by pushing multiple attributes to 20 while dumping others to 1, since there was no penalty for minimum stats.

The Geometric Shapes Experiment

This experience led Tim to experiment with geometric shapes as attribute constraints β€” a way to introduce structure without arbitrary rules like "you can't drop more than one stat below X." He tried this concept in The Outer Worlds but it didn't survive playtesting. He still believes there's a good idea hiding in it.

Many point-buy systems bolt on ad-hoc rules (minimum floors, maximum caps, limits on how many stats can be extreme). Tim felt shapes could replace all of that with a single intuitive constraint. He hasn't cracked it yet: "I'm just not smart enough to figure out what that idea is. Maybe one of you will."

The Damned-If-You-Do Problem

Tim closes with a candid observation about attribute design in RPGs: reuse familiar attributes and players call you boring and stuck in the past. Innovate with new ones and players say you're changing things for the sake of change. There's no winning β€” you can only try to find something that feels fresh without being alienating.

References