alee/golfgame
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Fix draw-swap animation race and smarter CPU go-out decisions

Browse Source 
Increase post_draw_settle timing (1.1s→1.3s) and swap retry delay
(100ms→350ms) to prevent draw animation from being cut short by the
arriving swap animation. Also make CPU go-out decisions consider
opponent scores and avoid swapping high cards (8+) into hidden slots.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
adlee-was-taken 2026-02-23 20:57:59 -05:00
parent 9a5bc888cb
commit 17f7d8ce7a
2 changed files with 23 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
client/card-animations.js
View File

@ -1127,7 +1127,7 @@ class CardAnimations {
}); });
// Now run the swap animation // Now run the swap animation
this._runUnifiedSwap(handCardData, heldCardData, handRect, heldRect, discardRect, T, rotation, wasHandFaceDown, onComplete); this._runUnifiedSwap(handCardData, heldCardData, handRect, heldRect, discardRect, T, rotation, wasHandFaceDown, onComplete);
}, 100); }, 350);
return; return;
} }

27
server/ai.py
View File

@ -43,8 +43,9 @@ CPU_TIMING = {
# Delay before CPU "looks at" the discard pile # Delay before CPU "looks at" the discard pile
"initial_look": (0.3, 0.5), "initial_look": (0.3, 0.5),
# Brief pause after draw broadcast - let draw animation complete # Brief pause after draw broadcast - let draw animation complete
# Must be >= client draw animation duration (~1s for deck, ~0.4s for discard) # Must be >= client draw animation duration (~1.09s for deck, ~0.4s for discard)
"post_draw_settle": 1.1, # Extra margin prevents swap message from arriving before draw flip completes
"post_draw_settle": 1.3,
# Consideration time after drawing (before swap/discard decision) # Consideration time after drawing (before swap/discard decision)
"post_draw_consider": (0.2, 0.4), "post_draw_consider": (0.2, 0.4),
# Variance multiplier range for chaotic personality players # Variance multiplier range for chaotic personality players
@ -1301,12 +1302,28 @@ class GolfAI:
max_acceptable_go_out = 14 + int(profile.aggression * 4) max_acceptable_go_out = 14 + int(profile.aggression * 4)
# Check opponent scores - don't go out if we'd lose badly
opponent_min = estimate_opponent_min_score(player, game, optimistic=False)
# Aggressive players tolerate a bigger gap; conservative ones less
opponent_margin = 4 + int(profile.aggression * 4) # 4-8 points
opponent_cap = opponent_min + opponent_margin
# Use the more restrictive of the two thresholds
effective_max = min(max_acceptable_go_out, opponent_cap)
ai_log(f" Go-out safety check: visible_base={visible_score}, " ai_log(f" Go-out safety check: visible_base={visible_score}, "
f"score_if_swap={score_if_swap}, score_if_flip={score_if_flip}, " f"score_if_swap={score_if_swap}, score_if_flip={score_if_flip}, "
f"max_acceptable={max_acceptable_go_out}") f"max_acceptable={max_acceptable_go_out}, opponent_min={opponent_min}, "
f"opponent_cap={opponent_cap}, effective_max={effective_max}")
# High-card safety: don't swap 8+ into hidden position unless it makes a pair
creates_pair = (last_partner.face_up and last_partner.rank == drawn_card.rank)
if drawn_value >= HIGH_CARD_THRESHOLD and not creates_pair:
ai_log(f" >> GO-OUT: high card ({drawn_value}) into hidden, preferring flip")
return None # Fall through to normal scoring (will flip)
# If BOTH options are bad, choose the better one # If BOTH options are bad, choose the better one
if score_if_swap > max_acceptable_go_out and score_if_flip > max_acceptable_go_out: if score_if_swap > effective_max and score_if_flip > effective_max:
if score_if_swap <= score_if_flip: if score_if_swap <= score_if_flip:
ai_log(f" >> SAFETY: both options bad, but swap ({score_if_swap}) " ai_log(f" >> SAFETY: both options bad, but swap ({score_if_swap}) "
f"<= flip ({score_if_flip}), forcing swap") f"<= flip ({score_if_flip}), forcing swap")
@ -1322,7 +1339,7 @@ class GolfAI:
return None return None
# If swap is good, prefer it (known outcome vs unknown flip) # If swap is good, prefer it (known outcome vs unknown flip)
elif score_if_swap <= max_acceptable_go_out and score_if_swap <= score_if_flip: elif score_if_swap <= effective_max and score_if_swap <= score_if_flip:
ai_log(f" >> SAFETY: swap gives acceptable score {score_if_swap}") ai_log(f" >> SAFETY: swap gives acceptable score {score_if_swap}")
return last_pos return last_pos