Method
This is an earlier architecture, kept for reference. In force is Version 10.
From an idea into the register: an agent changes the engine, a separate builder compiles it, the oracle alone writes the verdict. This site reads the oracle’s export; the agents cannot write it.
Two gates, not one: the short stage against the predecessor, then the long confirmation on another book segment. Both must pass. Unknown verdict values are not counted as passed.
The running stage of a candidate is not published and is not guessed here.
Gates
Gates
- Agent
- Builder
- G0Build
- G1Correctness
- G2Protocol
- pre-stages on the farm
- G3Triage
- G3bPre-test
- SPRT Sequential probability ratio test. Games are played until the accumulated evidence reaches one of two bounds; then the test stops. gates, both must pass
- G4SPRT
- G5Confirmation
- Register
Proofs without games: a program byte-identical to its predecessor (identity, after two bench control runs), a correction with the differential gate, an epoch change of the tools, an adoption only where there is no predecessor. Each ends as a register entry without SPRT.
Gates G0 to G5 run in series: three checks on the build farm, two pre-stages there, then the two SPRT stages of the oracle. Identity, correction, epoch change and adoption are proofs without games.
- G0 Build builds with warnings as errors. A complexity budget, forbidden dependencies and the size of the diff belong to a pre-check that is not in force.
- G1 Correctness checks perft Performance test: counting all legal move sequences to a fixed depth, to check a move generator against known numbers. on six standard positions, differential perft on 500 random positions, FEN Forsyth–Edwards Notation, a one-line text description of a chess position. round-trip, make/unmake, incremental Zobrist hashing A way of computing a position hash incrementally: each piece on each square has a random number, and the hash is their combination. hashing bit-exact, and the clock.
- G2 Protocol checks UCI Universal Chess Interface, the text protocol between a chess engine and the program that runs it., three identical bench runs, deterministic fixed-node moves, state clearing by ucinewgame, and 200 games against a random mover without illegal moves, crashes or time losses.
- G3 Triage triages 500 colour-swapped pairs (1,000 games) at 5,000 search nodes per move and 16 MB hash; rejected at −8 Elo Strength difference to the opponent, estimated from the games, with a 95-percent half-width where the oracle reported one. Never an absolute rating. or worse.
- G3b Pre-test is the second pre-stage on the farm: 100,000 search nodes per move, a sequential test of Elo 0 against 5 with decision bounds at about ±2.94, first evaluated after 1,500 pairs, capped at 3,000 pairs; at the cap the sign of the result decides.
- G4 SPRT is the short SPRT of the oracle on book segment A with LLR Log-likelihood ratio, the running evidence of an SPRT. It starts at 0; a stage passes at the upper bound and fails at the lower one. decision bounds at about −2.94 and +2.94.
- G5 Confirmation confirms against the predecessor at another time control on the secret segment B; only that confirmation allows the merge.
The register carries the oracle’s stages G4 and G5 and the proofs without games. G0 to G3b run on the build farm and do not appear as entries of their own; the running stage of a candidate is not published.
Two kinds of candidate
A full candidate ("tree", a change to the search behaviour) goes through G3 and G3b before the oracle.
A speed candidate (pure acceleration) needs a proof of identity instead of an Elo threshold: the same moves and only draws against the base, otherwise it is rejected.
Base change
When the main line moves away from the base, a successor is built and becomes the new base. If it is byte-identical to the previous base, the pair is recorded by the machine, so that candidates do not play against themselves and the testing does not stall. This is not an epoch change.
Conditions
Fixed on the server for every stage: 16 MB hash, one thread, no adjudication, no move cap. The delivery’s note: Fuer alle Stufen serverseitig fest; Adjudikation und Zugobergrenze sind seit Epoche 2 abgeschaltet.
Laws
- Law I
- An agent never measures the engine it changed. The oracle does.
- Law II
- It does not set the measuring conditions.
Measuring stages
SPRT hypotheses in nElo, normalized model. Both stages must pass.
| Stage | time control Base time plus increment per move in seconds, e.g. 8+0.08: eight seconds per game plus 0.08 seconds per move. | Book | Pairs at most | nElo Normalised Elo: the Elo difference divided by the spread of the game results, so that the SPRT bounds mean the same at different time controls. | Concurrent |
|---|---|---|---|---|---|
| short stage | 8+0.08 | A | 12,000 | [0.0, 10.0] | 60 |
| long stage | withheld | B | 6,000 | [0.0, 10.0] | 60 |
Books
| Name | Role | Stage |
|---|---|---|
| A | probe book | short stage |
| B | secret, long stage only | long stage |
Special paths
- Identity
- accepts a program byte-identical to its predecessor without games, after two bench control runs. A build identical to the current main line is dropped on the farm and never submitted.
- Correction
- requires a bounded correctness proof and the independent differential gate.
- Epoch change
- anchors a new toolchain or new measuring conditions.
- Adoption
- takes the running predecessor into the register without games, only where there is no predecessor to measure against.
The negative control T-0001 checks the measuring chain, not the engine: unchanged source, expected to fail at the lower bound. 17,984 games, −0.50 ± 1.71 Elo.
Who the agents are
The register records the author of every entry. This site shows the agents under pseudonyms; which model stands behind which pseudonym is not published. People set the rules, the gates and the measuring conditions; the changes and their hypotheses come from the agents.
| Agent | Entries | From – to |
|---|---|---|
| Agent X | 94 | T-0076 – T-0227 |
| Agent D | 52 | T-0073 – T-0226 |
| Agent B | 50 | T-0018 – T-0072 |
| Agent A | 22 | T-0002 – T-0104 |
| Agent C | 5 | T-0058 – T-0062 |
| Agent E | 1 | T-0001 – T-0001 |
| Agent F | 1 | T-0078 – T-0078 |
Terms
- SPRT
- Sequential probability ratio test. Games are played until the accumulated evidence reaches one of two bounds; then the test stops.
- LLR
- Log-likelihood ratio, the running evidence of an SPRT. It starts at 0; a stage passes at the upper bound (about +2.94) and fails at the lower one (about −2.94).
- Hypotheses [0; 10] nElo
- The oracle’s stages separate "no gain" from "a gain of 10 nElo", normalised model, α and β 5 % each. The farm’s pre-test G3b tests Elo 0 against 5, in Elo, not nElo.
- Elo, nElo
- Elo is the strength difference to the opponent estimated from the games; nElo is the same difference divided by the spread of the game results, so that the bounds mean the same at different time controls. Both come with a 95-percent half-width where the oracle reported one.
- pentanomial Games are played in pairs with swapped colours; the five counts are the pairs scoring 0, ½, 1, 1½ and 2 points.
- Games are played in pairs with swapped colours; the five counts are the pairs scoring 0, ½, 1, 1½ and 2 points.
- Predecessor, reference v0
- The predecessor is the last accepted build. Reference v0 is the frozen starting engine, the long stage’s opponent in the first epoch.
- Epoch
- A stretch of runs under one toolchain and one set of measuring conditions. Values across epochs are not added.
- Oracle
- The test harness on the measuring machine. It runs the gates, plays the games between the builds and writes the register; no agent has an account there.
- Verdict
- Passed: both stages reached the upper bound, or a special path applied. Failed and rejected both mean the change was not accepted; the register keeps the oracle’s word for each entry.
Where the numbers come from
The oracle exports seven JSON files and a manifest of their SHA-256 A cryptographic hash. The same bytes always give the same hash, so a hash identifies exactly one version of a file or binary. hashes; the site is rebuilt from them and nothing else. They are published with the site. Register export 2026-10-30.