kate/BadFish
1
0
Fork 0
mirror of https://github.com/sockspls/badfish synced 2025-05-01 09:13:08 +00:00
Code Issues Releases Wiki Activity

Update WDL model for current SF

Browse source 
This updates the WDL model based on the LTC statistics  (2M games).

Relatively small change, note that this also adjusts the NormalizeToPawnValue (now 361),
to keep win prob at 50% for 100cp.

closes https://github.com/official-stockfish/Stockfish/pull/4236

No functional change.
This commit is contained in:
Joost VandeVondele 2022-11-19 13:03:14 +01:00
parent d756d97a66
commit 3411631162
2 changed files with 3 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
src/uci.cpp
View file

@ -207,8 +207,8 @@ namespace {
// The coefficients of a third-order polynomial fit is based on the fishtest data
// for two parameters that need to transform eval to the argument of a logistic
// function.
constexpr double as[] = { 1.04790516, -8.58534089, 39.42615625, 316.17524816};
constexpr double bs[] = { -3.57324784, 22.28816201, -35.47480551, 85.60617701 };
constexpr double as[] = { -0.58270499, 2.68512549, 15.24638015, 344.49745382};
constexpr double bs[] = { -2.65734562, 15.96509799, -20.69040836, 73.61029937 };
// Enforce that NormalizeToPawnValue corresponds to a 50% win rate at ply 64
static_assert(UCI::NormalizeToPawnValue == int(as[0] + as[1] + as[2] + as[3]));

2
src/uci.h
View file

@ -35,7 +35,7 @@ namespace UCI {
// the win_rate_model() such that Stockfish outputs an advantage of
// "100 centipawns" for a position if the engine has a 50% probability to win
// from this position in selfplay at fishtest LTC time control.
const int NormalizeToPawnValue = 348;
const int NormalizeToPawnValue = 361;
class Option;