You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit e5bc4a4
Browse filesBrowse the repository at this point in the historyBrowse files
Replaces and renders scoreboard/release-equivalence prose introduced by 271d687 and c5383e6, the hard-lane summary introduced by 112c134, and contributing-llm baseline prose introduced by b2a578c. Preserve release 1.0.1 grades independently from the promoted main baseline.
@@ -31,7 +31,7 @@ PineForge is a C++17 engine for backtesting and forward execution, with a C ABI
31
31
32
32
The separate PineForge compiler, [`pineforge-codegen`](https://github.com/pineforge-4pass/pineforge-codegen-oss), translates a PineScript v6 script into a C++ strategy that attaches the engine's Pine execution adapter (its releases 1.0.0 and 1.0.1 do; 0.10.4 predates the adapter); it owns translation, not execution semantics. TradingView comparisons measure this Pine path under the tested configurations. The [order model](docs/pages/fill-model.md) describes the current submodels and the remaining migration work; [Architecture](#architecture-kernel-vs-parity) states the boundary.
33
33
34
-
-**Proven, not promised.** All 7,989graded probes — 309open reference strategies, 680community-shared scripts and 61 probe scripts the maintainers wrote, on 18 market/timeframe lanes — grade *excellent* or *strong*against TradingView's own trade lists: **7,905 excellent, 84 strong, zero below strong**. The graded probes' TradingView trade lists hold 4,776,328trades; 17 probes with TradingView-side defects are excluded.
34
+
-**Proven, not promised.** All <!-- pf:scoreboard.graded|int -->7,989<!-- /pf -->graded probes — <!-- pf:inventory.corpusScripts|int -->309<!-- /pf -->open reference strategies, <!-- pf:inventory.communityScripts|int -->680<!-- /pf -->community-shared scripts and <!-- pf:inventory.probeScripts|int -->61<!-- /pf -->probe scripts the maintainers wrote, on <!-- pf:scoreboard.lanes|int -->18<!-- /pf -->market/timeframe lanes — grade against TradingView's own trade lists: **<!-- pf:scoreboard.excellent|int -->7,946<!-- /pf -->excellent, <!-- pf:scoreboard.strong|int -->43<!-- /pf --> strong, <!-- pf:scoreboard.belowStrong|int -->0<!-- /pf -->below strong**. The graded probes' TradingView trade lists hold <!-- pf:inventory.tvTrades|int -->4,776,328<!-- /pf -->trades; <!-- pf:scoreboard.anomaliesExcluded|int -->17<!-- /pf --> probes with TradingView-side defects are excluded.
35
35
-**Open runtime.** The engine and native live runner are Apache-2.0. The separately distributed [PineForge compiler](https://github.com/pineforge-4pass/pineforge-codegen-oss/blob/main/LICENSE) uses PolyForm Noncommercial terms with additional personal-trading permission; commercial use requires a separate license. Public reference strategies, benchmarks and validation tooling are available in their respective repositories; the community-script test set is not redistributed.
36
36
-**Fast.** In-process, no interpreter: median **36× faster than PyneCore** on the 200 strategies both engines time, measured at engine `35db01c8` on 2026-09-29 on an AWS c7a.8xlarge with PyneCore timed as a subprocess, its interpreter start-up included (PineForge runs a median 807k bars/s per strategy over its 201 slots with the bar magnifier on; [method](benchmarks/results/speed.md)). Parameter sweeps re-run a loaded `.so` with new inputs — no recompile, no fork.
37
37
-**Deterministic to the bit.** Two runs with the same inputs produce identical trade lists. Same on Linux and macOS.
@@ -226,38 +226,46 @@ live behavior or real broker fills.
226
226
227
227
## Validation scoreboard
228
228
229
-
**Measured 2026-09-29** on engine `35db01c8` with codegen-oss `70c2b4af` (the maintainers' baseline `pineforge-parity-baseline-20260929-engine-35db01c8`): **7,989 graded probes, 7,905 excellent + 84 strong**, none below *strong*, across 18 market/timeframe lanes. Their TradingView trade lists hold 4,776,328 trades. Engine v1.0.0, released 2026-09-30 from main `133f5714`, includes `35db01c8`, codegen 1.0.0 carries `70c2b4af`'s code, and the releases changed no grade. The 1.0.1 code was graded before it merged: codegen-oss `67892cda`, the code of codegen 1.0.1, with engine `b2a578ce`, whose library source is v1.0.0's, graded the same 7,989 probes with no probe changing tier (baseline `pineforge-parity-baseline-20261001-codegen-67892cda`); engine v1.0.1 changes only documentation since v1.0.0.
Release **1.0.1 still grades <!-- pf:releases[1.0.1].scoreboard.excellent|int -->7,905<!-- /pf --> excellent / <!-- pf:releases[1.0.1].scoreboard.strong|int -->84<!-- /pf --> strong until the next release**, on <!-- pf:releases[1.0.1].scoreboard.graded|int -->7,989<!-- /pf --> probes (baseline `<!-- pf:releases[1.0.1].scoreboard.id -->pineforge-parity-baseline-20261001-codegen-67892cda<!-- /pf -->`, <!-- pf:releases[1.0.1].scoreboard.date -->2026-10-01<!-- /pf -->). A main scoreboard advance does not change release results.
232
+
233
+
The quantities above render from the public [facts tokens](https://github.com/pineforge-4pass/pineforge-release/blob/7238bb3d1a30c68cc857197cadf883bce29680f0/facts/facts.json). Maintain them with `lab facts render --repo . --facts <local facts file or pinned raw URL>`; `lab facts check` with the same inputs reports drift. Grades are registry-derived; the authored-script and closed-trade inventory is explicitly sourced to a historical public README for the identical population, not to registry row or slug totals.
230
234
231
235
| Board | Test set | Result |
232
236
|---|---|---|
233
-
|**Public** — [open corpus](https://github.com/pineforge-4pass/pineforge-corpus)| 312 reference strategies on BINANCE:ETHUSDT.P 15m, Apache-2.0, reproducible by anyone | this repository's sweep: **311 excellent + 1 declared anomaly**; the 309of them in the measured population: **309 excellent**|
234
-
|**Closed test**| 7,680probes of 741TradingView scripts across the 18 lanes: 680community-shared scripts (7,179probes), private under TradingView's Terms of Service, and 61 probe scripts the maintainers wrote (501probes) |**7,596 excellent + 84 strong**|
237
+
|**Public** — [open corpus](https://github.com/pineforge-4pass/pineforge-corpus)| 312 reference strategies on BINANCE:ETHUSDT.P 15m, Apache-2.0, reproducible by anyone | this repository's sweep: **311 excellent + 1 declared anomaly**; the <!-- pf:scoreboard.corpusProbes|int -->309<!-- /pf -->of them in the measured population: **<!-- pf:scoreboard.scopes.corpus.excellent|int -->309<!-- /pf --> excellent**|
238
+
|**Closed test**|<!-- pf:inventory.closedProbes|int -->7,680<!-- /pf -->probes of <!-- pf:inventory.closedScripts|int -->741<!-- /pf -->TradingView scripts across the <!-- pf:scoreboard.lanes|int -->18<!-- /pf -->lanes: <!-- pf:inventory.communityScripts|int -->680<!-- /pf -->community-shared scripts (<!-- pf:inventory.communityProbes|int -->7,179<!-- /pf -->probes), private under TradingView's Terms of Service, and <!-- pf:inventory.probeScripts|int -->61<!-- /pf -->probe scripts the maintainers wrote (<!-- pf:inventory.probeScriptProbes|int -->501<!-- /pf -->probes) |**<!-- pf:scoreboard.scopes.closed.excellent|int -->7,637<!-- /pf -->excellent + <!-- pf:scoreboard.scopes.closed.strong|int -->43<!-- /pf --> strong**|
The hard lane holds the 309 public-corpus probes and 700 closed-test probes (640 community scripts, 60 maintainer-written probe scripts). 17 more probes on it reproduce TradingView-side defects and are excluded from grading.
254
+
| NASDAQ:AAPL · 15m | 632 | 631 | 1 | 0 |
255
+
| NASDAQ:AAPL · 1D | 464 | 464 | 0 | 0 |
256
+
| NSE:NIFTY · 15m | 314 | 313 | 1 | 0 |
257
+
| NSE:NIFTY · 1D | 225 | 225 | 0 | 0 |
258
+
| NYSE:F · 15m | 637 | 625 | 12 | 0 |
259
+
| NYSE:F · 1D | 505 | 502 | 3 | 0 |
260
+
| OANDA:EURUSD · 15m | 653 | 647 | 6 | 0 |
261
+
| OANDA:EURUSD · 1D | 453 | 453 | 0 | 0 |
262
+
| OANDA:XAUUSD · 15m | 695 | 692 | 3 | 0 |
263
+
| OANDA:XAUUSD · 1D | 456 | 456 | 0 | 0 |
264
+
|**Total**|**7,989**|**7,946**|**43**|**0**|
265
+
266
+
<!-- /pf -->
267
+
268
+
The hard lane combines public-corpus and closed-test probes for BINANCE:ETHUSDT.P 15m; zero regression is allowed. TradingView-side defects are excluded from grading.
Copy file name to clipboardExpand all lines: docs/pages/contributing-llm.md
+2-2Lines changed: 2 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -19,8 +19,8 @@ a **kernel** that matches triggers, prices fills, books lots and settles, and a
19
19
**Pine adapter** that reproduces TradingView's execution semantics on top of
20
20
it. A separate project transpiles PineScript into C++ that attaches the
21
21
adapter. The engine's value is that its output is *byte-reproducible* and graded
22
-
*trade for trade against TradingView* on a fixed population (8,006 probes at
23
-
the current baseline; 7,905 of the 7,989graded are excellent, 84 strong), so
22
+
*trade for trade against TradingView* on a fixed population (<!-- pf:scoreboard.population|int -->8,006<!-- /pf --> probes at
23
+
the current baseline; <!-- pf:scoreboard.excellent|int -->7,946<!-- /pf -->of the <!-- pf:scoreboard.graded|int -->7,989<!-- /pf -->graded are excellent, <!-- pf:scoreboard.strong|int -->43<!-- /pf --> strong), so
24
24
almost every rule below exists to keep a change from quietly moving a byte.
0 commit comments