Skip to content
Permalink
Browse files

Retire CloseEnemies. Bench: 3520923

  • Loading branch information...
31m059 committed Dec 27, 2018
1 parent 69dc556 commit 15166f0d1c926facb4595b5c0d6a597b5b78aa5b
Showing with 0 additions and 4 deletions.
  1. +0 −4 src/evaluate.cpp
@@ -153,7 +153,6 @@ namespace {

// Assorted bonuses and penalties
constexpr Score BishopPawns = S( 3, 7);
constexpr Score CloseEnemies = S( 8, 0);
constexpr Score CorneredBishop = S( 50, 50);
constexpr Score Hanging = S( 69, 36);
constexpr Score KingProtector = S( 7, 8);
@@ -482,9 +481,6 @@ namespace {
if (!(pos.pieces(PAWN) & kingFlank))
score -= PawnlessFlank;

// King tropism bonus, to anticipate slow motion attacks on our king
score -= CloseEnemies * tropism;

if (T)
Trace::add(KING, Us, score);

2 comments on commit 15166f0

@snicolet

This comment has been minimized.

Copy link

replied Dec 27, 2018

Apart from the full simplification of CloseEnemies, could you also try a couple of tests with other values for the bonus? Like S(6,0), S(7,0) and S(9,0)? :-)

@31m059

This comment has been minimized.

Copy link
Owner Author

replied Dec 27, 2018

@snicolet I would be happy to do so. Do we think that this bonus is sensitive enough that differences of only S(1, 0) will meaningfully affect performance?

More importantly, what bounds should we use for tweaks? The old standard [0, 4] at STC and then the new [0, 3.5] for LTC? (Given that complexity-adding patches are LTC [0, 3.5], LTC [0, 4] doesn't seem to make sense...)

Please sign in to comment.
You can’t perform that action at this time.