Skip to content

Update to v0.32 flags#179

Merged
mooskagh merged 1 commit intoLeelaChessZero:masterfrom
john-sp:update-flags
Nov 7, 2025
Merged

Update to v0.32 flags#179
mooskagh merged 1 commit intoLeelaChessZero:masterfrom
john-sp:update-flags

Conversation

@john-sp
Copy link
Member

@john-sp john-sp commented Oct 22, 2025

I gave gemini the output of lc0 -h and it updated the table to this.

Copilot AI review requested due to automatic review settings October 22, 2025 23:00
@github-actions
Copy link

🚀 Draft Preview: https://pr179-draft.lc0.org/

Copy link
Contributor

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull Request Overview

This PR updates the documentation to reflect the latest flags and parameters available in lc0 v0.32, with the content being updated by Gemini based on the output of lc0 -h.


Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.

|VerboseMoveStats | -v, --[no-]verbose-move-stats | false | Display Q, V, N, U and P values of every move candidate after each move. |
|FpuStrategy | --fpu-strategy | reduction | How is an eval of unvisited node determined. "First Play Urgency" changes search behavior to visit unvisited nodes earlier or later by using a placeholder eval before checking the network. The value specified with --fpu-value results in "reduction" subtracting that value from the parent eval while "absolute" directly uses that value. |
|FpuValue | --fpu-value | 0.33 | "First Play Urgency" value used to adjust unvisited node eval based on --fpu-strategy. |
|CacheHistoryLength | --cache-history-length | 0 | Length of history, in half-moves, to include into the cache key. When this value is less than history that NN uses to eval a position, it's possble that the search will use eval of the same position with different history taken from cache. |
Copy link

Copilot AI Oct 22, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Corrected spelling of 'possble' to 'possible'.

Suggested change
|CacheHistoryLength | --cache-history-length | 0 | Length of history, in half-moves, to include into the cache key. When this value is less than history that NN uses to eval a position, it's possble that the search will use eval of the same position with different history taken from cache. |
|CacheHistoryLength | --cache-history-length | 0 | Length of history, in half-moves, to include into the cache key. When this value is less than history that NN uses to eval a position, it's possible that the search will use eval of the same position with different history taken from cache. |

Copilot uses AI. Check for mistakes.
|DrawScore | --draw-score | 0.00 | Adjustment of the draw score from white's perspective. Value 0 gives standard scoring, value -1 gives Armageddon scoring. |
|ContemptMode | --contempt-mode | play | Affects the way asymmetric WDL parameters are applied. Default is 'play' for matches, use 'white_side_analysis' and 'black_side_analysis' for analysis. Use 'disable' to deactivate contempt. |
|Contempt | --contempt | | The simulated Elo advantage for the WDL conversion. Comma separated list in the form [name=]value, where the name is compared with the `UCI_Opponent` value to find the appropriate contempt value. The default value is taken from `UCI_RatingAdv` and will be overridden if either a value without name is given, or if a name match is found. |
|WDLCalibrationElo | --wdl-calibration-elo | 0.00 | Elo of the active side, adjusted for time control relative to rapid.To retain raw WDL without sharpening/softening, use default value 0. |
Copy link

Copilot AI Oct 22, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Missing space after period. Should be 'rapid. To' instead of 'rapid.To'.

Suggested change
|WDLCalibrationElo | --wdl-calibration-elo | 0.00 | Elo of the active side, adjusted for time control relative to rapid.To retain raw WDL without sharpening/softening, use default value 0. |
|WDLCalibrationElo | --wdl-calibration-elo | 0.00 | Elo of the active side, adjusted for time control relative to rapid. To retain raw WDL without sharpening/softening, use default value 0. |

Copilot uses AI. Check for mistakes.
|SmartPruningMinimumBatches | --smart-pruning-minimum-batches | 0 | Only allow smart pruning to stop search after at least this many batches have been evaluated. It may be useful to have this value greater than the number of search threads in use. |
|RamLimitMb | --ramlimit-mb | 0 | Maximum memory usage for the engine, in megabytes. The estimation is very rough, and can be off by a lot. For example, multiple visits to a terminal node counted several times, and the estimation assumes that all positions have 30 possible moves. When set to 0, no RAM limit is enforced. |
|MoveOverheadMs | --move-overhead | 200 | Amount of time, in milliseconds, that the engine subtracts from its total available time (to compensate for slow connection, interprocess communication, etc). |
|TimeManager | --time-manager | legacy | Name and config of a time manager. Possible names are 'legacy' (default), 'smooth', 'alphazero', and simple. See [configuration details](https://lc0.org/timemgr). |
Copy link

Copilot AI Oct 22, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Inconsistent quote usage: 'simple' should be quoted like the other time manager names ('legacy', 'smooth', 'alphazero').

Suggested change
|TimeManager | --time-manager | legacy | Name and config of a time manager. Possible names are 'legacy' (default), 'smooth', 'alphazero', and simple. See [configuration details](https://lc0.org/timemgr). |
|TimeManager | --time-manager | legacy | Name and config of a time manager. Possible names are 'legacy' (default), 'smooth', 'alphazero', and 'simple'. See [configuration details](https://lc0.org/timemgr). |

Copilot uses AI. Check for mistakes.
@mooskagh mooskagh merged commit 5de8508 into LeelaChessZero:master Nov 7, 2025
1 check passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants