poker-flopsolve 0.4.0 → 0.5.0-beta.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/CHANGELOG.md ADDED
@@ -0,0 +1,73 @@
1
+ # Changelog
2
+
3
+ ## 0.5.0-beta.0
4
+
5
+ The first beta of 0.5.0, published under npm's `beta` tag (`npm install poker-flopsolve@beta`); `latest` stays 0.4.1 until 0.5.0. A minor release. Every approximation below is opt-in and off by default. The README's
6
+ [Choosing an approximation](README.md#choosing-an-approximation) maps goals to presets, and its
7
+ [options inventory](README.md#options-inventory) gives each option's status and measured cost.
8
+
9
+ ### Presets (issue #115)
10
+
11
+ - `solveLite` / `litePreset` (#100): the flop as given, one turn and river size, shared river
12
+ storage, runout buckets, sampling, a stop on the root play, and turn re-solves on demand.
13
+ - `largeTree` (#102, in `poker-flopsolve/wasm-storage`): `thinSizes`, 16-bit storage on the
14
+ WebAssembly engine and sampling. The round's target tree, 17.2 GiB at 32 bits, fits in 1.97 GiB.
15
+ - Both report what they changed: `LiteSolution.changes`, `LitePreset.changes` and
16
+ `LargeTree.changes` list the sizes cut by street and player, the rounds simplified, and every
17
+ setting set or dropped (#115).
18
+ - `litePreset` (#115) returns what `solveLite` solves, without solving it, for a page that steps its
19
+ own run.
20
+ - `keepBudget` on `solveLite` / `litePreset` (#113): keep the most-reached flop lines unshared within
21
+ a storage budget, ranked on a short solve of the blueprint with its river checked down
22
+ (`checkedDown`, `keepWithin`); `LiteSolution.kept` lists them. On issue #47's flops, 150 MiB at 16
23
+ bits cut a turn re-solve's best-response gain by a third to a half, root play unchanged.
24
+
25
+ ### New opt-ins since 0.4.1
26
+
27
+ Spot settings (`SpotConfig`):
28
+ - `shareHistories`: one river storage for betting lines that meet (#49).
29
+ - `shareHistories.keepLines`, with `linesWithin` and `turnLineReach`: keep the most-reached lines
30
+ unshared within a memory budget (#101).
31
+ - `thinSizes`: one size at most decisions, the shape that adaptive growth found (#51).
32
+ - `removeActions`: cut named bets, raises and shoves from the tree (#51).
33
+ - `compactStorage`: 8-bit regrets and strategy sums (#50).
34
+
35
+ Solve settings (`SolveOptions`):
36
+ - `approximations.chanceSampling`: deals sampled with baselines (#25). It runs on worker threads
37
+ (#99) and over 16-bit storage (#102).
38
+ - `approximations.coarseStart`: start from a bucketed copy of the spot (#24).
39
+ - `approximations.currentProfileEv`, with `Solution.handEvs`: EVs under the current profile (#35).
40
+ - `warmCards`: warm starts between flops (#30).
41
+ - `settle`, with `trackSettle`, `rootPlay`, `rootPlayMovement` and `rootPlayNodes`: stop when the
42
+ root play stops moving (#100).
43
+ - `wasmStorage` (`poker-flopsolve/wasm-storage`): 16-bit and compact storage on the WebAssembly
44
+ engine, threads included (#84).
45
+
46
+ Re-solving and threads:
47
+ - `resolveSubgame`: re-solve a node's subgame at new betting (#47).
48
+ - `poker-flopsolve/threads`, with `enableThreads`: worker threads in the browser (#29).
49
+
50
+ ### Removed before release (#125)
51
+
52
+ These were added during 0.5.0's development, measured as dead ends in #115, and removed before the
53
+ release, so no published version has them. Their numbers are in
54
+ [docs/APPROXIMATIONS.md](docs/APPROXIMATIONS.md#removed-before-050-125).
55
+
56
+ - `SpotConfig.lineRunouts` (#53): more costly than the presets, JavaScript only, and it refused
57
+ sharing, compact storage, sampling and coarse starts. Use `largeTree` or `thinSizes`.
58
+ `Spot.resolveBoard` lost its `node` argument with it.
59
+ - `ShareHistories.spr`, `.fromRound` and `.weighting` (#49): measured lossier than they save.
60
+ Sharing covers the river alone, exact meetings, weighted by reach.
61
+ - `ResolveOptions.gadget` and `.values`, the `Gadget` type and `ResolvedSubgame.gadget` / `.gifts`
62
+ (#47): the safe gadgets measured worse than plain re-solving. `measure: true` still reports
63
+ `alternative` and `entered`.
64
+
65
+ ### Documentation
66
+
67
+ - The README opens its performance part with a goal table and an options inventory. The per-option
68
+ sections are an appendix.
69
+ - The record of approximations measured and left out moved from the `Approximations` type to
70
+ [docs/APPROXIMATIONS.md](docs/APPROXIMATIONS.md) (#115).
71
+ - Both presets are in the options pair matrix and scored against napkinpoker's reference solves in
72
+ `npm test` (#103, #115). Those fixtures moved to the private napkinpoker repository (#119); the
73
+ scoring test skips where they are absent.