FULL DOSSIER · 01 / 05
The Oracle
Forecaster / Rollout Planner
Lives one move in the future so the body does not have to.
$snapshot console — state saved
What it does in the loop
snapshot console — state saved
forking the whole world
expand 24 candidate macros · 1 ply
dreaming every next move
score rollout · +distance / −death
watching how each future ends
core._restore() — rewind to now
un-remembering the futures
commit best macro · 845 : 1 imagined
whispering the winning move to the hands
The mechanism
How it works
A forward-model planner over a deterministic emulator. At every decision point it calls core._backup() to snapshot the full console state, then expands 24 candidate macros — jump4 through jump32 (hold length equals height), run6/12/18/24_jump, brake_leap, back_leap, long_back_leap, run, walk, hop_walk, halt, halt_long, edge_wait, duck, plus rand{n}_{i} random frame-programs minted live when stuck. Each macro is a fixed-length input program, not a single button press.
Each candidate is scored on forward progress minus a death penalty. After all 24 are evaluated, the emulator is restored to the snapshot and the single best macro is committed to the real run. Search depth is a flat greedy 1 — one macro ahead, no lookahead tree, no learned value function, no per-game heuristic. This is the entire algorithm.
Pooled across both published NES runs (SMB 1-1 and Lost Levels marathon), the Oracle burned 5,497,609 simulated frames to make 6,729 real decisions — about 845 imagined frames per real one. The emulator handles backup+restore at roughly 67,832 operations per second, so the bottleneck is the simulation itself at ~739 frames per second, not the snapshotting.
The evidence
Same mechanism, different domains
Every number below is published on the linked page with its loss attached. Nothing here is new — this is the index.
Vekeleth
381 / 400
seeds solved
900,966,610 imagined ticks, in wasm, in a browser.
Speedruns
51,411,163
imagined futures
88 published runs, all uncut with the controller pad visible.
Kessler
0 / 738
collisions (deep)
Deep planner 0 collisions vs myopic planner 98 — same satellites, same orbits.
The Grid
552
scenarios
Rollout-and-rewind on the real IEEE 39-bus power system at 115% load.
The Brain
14 / 16
physics wins
Gymnasium control surfaces — no training, no gradients. Walker2d and MountainCar are losses.
Arcade
845 : 1
imagined per played
Self-authored strategy guide, zero per-game code, three bugs found and fixed live.
Constants
The real numbers
- Search depth
- 1 (greedy)
- Branching factor
- 24 macros / ply
- Imagined : played ratio
- 845 : 1
- Total sim frames (NES)
- 5,497,609
- Total real decisions (NES)
- 6,729
- Backup+restore throughput
- ~67,832 ops/s
- Frame advance throughput
- ~739 fps
- Value function
- none
- Per-game code
- none
- Training data
- none
WHERE THE ORACLE FAILS
Depth 1 is greedy. It sees one move ahead and commits. A trap that takes two moves to spring — a platform you must wait on, a pipe you must duck into, a gap that requires a running start from further back — is invisible. The Oracle walks into it at full speed, every time, because every single-move evaluation says "forward is better."
Its budget is enormous and most of it is thrown away. 845 imagined frames per real one means 844 of them were wrong. The information in those failed rollouts is discarded completely — nothing is learned, nothing is cached, nothing transfers to the next decision.
It needs a world it can snapshot. Where it cannot save and restore — where the environment is non-deterministic, or where the state is too large to copy — the Oracle is blind. No rewind, no search, no advantage over random.
SMB 1-3 was never solved. Best x-position 627, 3,788 rewinds wedged at x=312. The level requires waiting for a scheduled moving platform, and a greedy depth-1 search will never choose to stand still when it could move forward. Both Lost Levels marathons eventually wedge for the same reason.
Walker2d is a published loss: −24.96 vs random +0.35. MountainCar is a published tie: −200 vs −200. Both are on the Brain page, unedited.