>[!warning]
>This content has not been peer reviewed.
# Complexity and computation — RRT foundation
**Complexity** and **computation** (polynomial scaling, verification vs search, step-count as cost) are the ninth foundational area. They underpin the P vs NP application: Landauer and resolution cost; A2, A3.
---
## I. Classical side
- **P, NP:** Polynomial-time solvable vs polynomial-time verifiable; is "finding" ever much harder than "checking"?
- **Step-count:** Computation as a sequence of steps; each step has a cost (time, energy).
---
## II. Mapping to RRT / RST
| Concept | RRT / RST identity | Application |
|:---|:---|:---|
| **Step** | Translation; each step costs (Landauer). | A3 (Translation). |
| **Verification** | Bounded consistency checks; few steps. | P vs NP: "check" is cheap. |
| **Solving** | Search over states; potentially many steps. | P vs NP: "find" can be expensive. |
| **Polynomial time** | Workload (step count) bounded by $n^k$; substrate scaling. | A2; finite resolution. |
| **Resolution cost** | Cost of maintaining / discovering state; can exceed verification cost. | Landauer; A2. |
So: **complexity** (P vs NP) is reframed as asymmetry between resolution cost (solve) and verification cost (check) in the substrate's accounting.
---
## Links
- **Foundation index:** [[../Foundation index]]
- **P vs NP:** [[P vs NP/P vs NP (RST)]]
- **RRT:** [[Relational Resolution Theory (RRT)]]
- **Applications Roadmap:** [[../../Applications Roadmap]]