The Infinite Atlas discussion - Reacting to Zizaran's video
Summary
- Comparison of player agency in Path of Exile 1 versus Path of Exile 2 endgame.
- Identification of “objectively terrible” map layouts like Sun Temple and Molten Vault.
- Proposals for increasing agency, including “Horizon Orbs,” triple-choice map nodes, and layout blocking.
- Discussion on incentivizing bad maps through target-farmable uniques, biome-specific loot, and thematic rewards.
- Suggestions for improving Atlas mobility via Tower mechanics and Kingsmarch “terraforming.”
- Hopes for procedural generation variety in biome sizes, similar to Minecraft’s world generation.
The Conflict of Player Agency
There is a notable difference in player agency between Path of Exile 1 and Path of Exile 2. In PoE 1, players choose the exact layouts they want to run. In PoE 2’s Infinite Atlas, players are largely forced to run the maps the game provides. While the biome system is a positive addition, there is a need for a healthy compromise that allows for more player choice without necessarily returning to the PoE 1 “Strand only” meta.
Identifying and Rerolling Bad Layouts
Specific maps are considered significantly worse than others due to their layout and mechanic compatibility:
- Mire: Often cited as bad, though arguably better than others.
- Sun Temple: Features multiple levels with RNG-based corner objectives, leading to excessive backtracking.
- Molten Vault: A maze-like layout involving levers and a boss arena that limits space for league mechanics like Abyss to spawn.
A potential solution is the introduction of a Horizon Orb. This would be a common currency (similar to an Exalted or Chaos Orb) that allows a player to click a map layout on the Atlas and reroll it to a different type within the same biome. This maintains the thematic vibe of the biome while allowing players to avoid layouts they find frustrating.
The Triple-Choice Node System
Another proposal for agency involves changing how map nodes are selected. When clicking a node, the game could offer a choice between three random maps from that biome (e.g., Sump, Backwash, or Mire).
- To minimize clicking, the game could remember the last choice.
- A selectable Atlas Keystone could be added: “Jesus Take the Wheel,” providing 1% more Quantity in exchange for removing the choice and automating the map selection.
- GGG could use data from these choices to identify which maps have a near-zero pick rate, signaling that those specific layouts need a fundamental rework.
Incentivizing Maps with Rewards
Instead of just fixing layouts, developers could add specific incentives to less popular maps:
- Map Buffs: A chance for bad maps to spawn with inherent buffs like increased Rarity, Magic Find, or faster movement speed.
- Target Farming: Assigning specific unique items or higher drop weights to certain map bosses or layouts.
- Biome-Specific Uniques: Since players move through biomes, certain uniques could be tied to the forest or ocean biomes rather than specific maps, allowing for target farming even without total layout control.
- Thematic Drops: Vaal-themed maps (like Vaal Factory or Molten Vault) could have a chance to drop Vaal currency or provide access to corruption benches, similar to the Temple of Atzoatl.
Atlas Strategy and Consistency
Strategy testing requires consistency. To accurately measure the profitability of a mechanic, one needs to run the same layout with the same variables multiple times. The current Infinite Atlas makes it difficult to generate reliable data because layouts and tablet rolls are constantly changing. Providing a way to skip or reroll bad layouts is essential to prevent player burnout during long sessions.
Mobility and Map Blocking
Several ideas exist for navigating the Atlas more efficiently:
- Layout Blocking: A UI option to “unfavorite” or block one layout per biome so it never spawns.
- Terraforming: Using the Kingsmarch town system to “craft” expeditions that bridge gaps on the Atlas, effectively skipping undesirable paths.
- Tower Functionality: Cleared Towers currently feel less important. They could be updated to allow players to travel to any node within the Tower’s range without needing a specialized unique tablet. Alternatively, make the “travel anywhere” tablets significantly more common.
Biome Generation and RNG
There is a hope for more “magic” in the procedural generation of biomes. Rather than every Breach or Abyss biome being a standardized size, they should have variable RNG. Discovering a massive Abyss biome covering 80 maps should feel like a rare and exciting event, similar to finding a unique biome in Minecraft. This adds a sense of discovery to the Atlas that persists over years of play.