Quiescence search #39

andtsa · 2025-04-27T11:56:34Z

No description provided.

Signed-off-by: Andreas Tsatsanis <[email protected]>

github-actions · 2025-08-21T11:17:06Z

d6 benchmark results:

d6 on TARGET branch:
--------------------
�[1m�[32m    Finished�[0m `test` profile [unoptimized + debuginfo] target(s) in 0.07s
�[1m�[32m     Running�[0m tests/d6.rs (target/debug/deps/d6-a7f1cab7364c11cd)

running 1 test
[*] creating engine at version 0.6.2
[*] tt: created VL table with 8 entries.
[*] Entering UCI mode
engine_info (0ms): info string entering uci mode
engine_info (1ms): info string table resized to 8 entries.
[*] uci: option use_tt set to on.
engine_info (1ms): info string table resized to 1024 entries.
engine_info (1ms): info string table resized to 1024 entries.
engine_info (14ms): info depth 1 seldepth 1 multipv 1 nodes 40 nps 78330 hashfull 0 tbhits 0 time 0 score cp 11 pv g1f3
engine_info (14ms): info depth 2 seldepth 2 multipv 1 nodes 860 nps 418204 hashfull 19 tbhits 0 time 2 score cp 25 pv g1f3 g8f6
engine_info (14ms): info depth 3 seldepth 3 multipv 1 nodes 2852 nps 434606 hashfull 311 tbhits 0 time 6 score cp 10 pv g1f3 g8f6 d2d4
engine_info (54ms): info depth 4 seldepth 4 multipv 1 nodes 27215 nps 528083 hashfull 688 tbhits 28 time 51 score cp 20 pv g1f3 b8c6 b1c3 c6d4
engine_info (208ms): info depth 5 seldepth 6 multipv 1 nodes 98526 nps 490433 hashfull 1000 tbhits 1628 time 200 score cp 41 pv g1f3 b8c6 f3g5 f7f5 g5f7
engine_info (3009ms): info depth 6 seldepth 8 multipv 1 nodes 1481064 nps 493259 hashfull 1000 tbhits 7032 time 3002 score cp -38 pv b1c3 e7e6 c3b5 f8c5 e2e3 c5e3
Best move: b1c3
 duration: 3170ms 
 max depth: 6
 total nodes searched: 1610557
killing child
test main ... ok

test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 3.17s

********************
 
d6 on SOURCE branch:
--------------------
�[1m�[32m   Compiling�[0m chesseng v0.6.3 (/home/runner/work/chesseng/chesseng)
�[1m�[32m    Finished�[0m `test` profile [unoptimized + debuginfo] target(s) in 1.84s
�[1m�[32m     Running�[0m tests/d6.rs (target/debug/deps/d6-fd436cfa4021ab6f)

running 1 test
[*] creating engine at version 0.6.3
[*] tt: created VL table with 8 entries.
[*] Entering UCI mode
engine_info (0ms): info string entering uci mode
[*] uci: option use_tt set to on.
engine_info (13ms): info depth 1 seldepth 0 multipv 1 nodes 60 nps 86842 hashfull 0 tbhits 0 time 0 score cp 10 pv d2d4
engine_info (13ms): info depth 2 seldepth 1 multipv 1 nodes 898 nps 294499 hashfull 750 tbhits 0 time 3 score cp 25 pv g1f3 g8f6
engine_info (22ms): info depth 3 seldepth 4 multipv 1 nodes 4701 nps 319593 hashfull 1000 tbhits 0 time 14 score cp 10 pv g1f3 g8f6 d2d4
engine_info (83ms): info depth 4 seldepth 5 multipv 1 nodes 28256 nps 351314 hashfull 1000 tbhits 1 time 80 score cp 25 pv g1f3 d7d5 d2d4 g8f6
engine_info (478ms): info depth 5 seldepth 9 multipv 1 nodes 147894 nps 317794 hashfull 1000 tbhits 1 time 465 score cp 8 pv g1f3 d7d5 d2d4 g8f6 b1c3
engine_info (2972ms): info depth 6 seldepth 10 multipv 1 nodes 937063 nps 316136 hashfull 1000 tbhits 3 time 2964 score cp 25 pv g1f3 d7d5 d2d4 b8c6 b1c3 g8f6 c3d5
Best move: g1f3
 duration: 3133ms 
 max depth: 6
 total nodes searched: 1118872
killing child
test main ... ok

test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 3.13s

github-actions · 2025-08-21T11:17:07Z

prune benchmark results:

prune on TARGET branch:
-----------------------
�[1m�[32m    Finished�[0m `test` profile [unoptimized + debuginfo] target(s) in 0.07s
�[1m�[32m     Running�[0m tests/prune.rs (target/debug/deps/prune-295a37db5b39985c)

running 1 test
[*] creating engine at version 0.6.2
[*] tt: created VL table with 8 entries.
[*] Entering UCI mode
engine_info (0ms): info string entering uci mode
engine_info (1ms): info string table resized to 1024 entries.
engine_info (1ms): info string table resized to 1024 entries.
engine_info (1ms): info string table resized to 1024 entries.
engine_info (15ms): info string table resized to 1048576 entries.
engine_info (15ms): info string table resized to 1048576 entries.
engine_info (25ms): info depth 1 seldepth 1 multipv 1 nodes 40 nps 163460 hashfull 0 tbhits 0 time 0 score cp 11 pv g1f3
engine_info (25ms): info depth 2 seldepth 2 multipv 1 nodes 860 nps 221505 hashfull 0 tbhits 0 time 3 score cp 25 pv g1f3 g8f6
engine_info (35ms): info depth 3 seldepth 3 multipv 1 nodes 2734 nps 197381 hashfull 0 tbhits 0 time 13 score cp 10 pv g1f3 g8f6 d2d4
engine_info (146ms): info depth 4 seldepth 4 multipv 1 nodes 26083 nps 212264 hashfull 1 tbhits 26 time 122 score cp 20 pv g1f3 b8c6 b1c3 c6d4
engine_info (347ms): info depth 5 seldepth 6 multipv 1 nodes 64767 nps 197353 hashfull 8 tbhits 1979 time 328 score cp 41 pv g1f3 b8c6 f3g5 f7f5 g5f7
engine_info (4973ms): info depth 6 seldepth 8 multipv 1 nodes 1041980 nps 210219 hashfull 58 tbhits 19570 time 4956 score cp -38 pv b1c3 e7e6 c3b5 f8c5 e2e3 c5e3
engine_info (9861ms): info depth 7 seldepth 10 multipv 1 nodes 1975609 nps 200662 hashfull 196 tbhits 73127 time 9845 score cp 77 pv b1c3 e7e6 g1f3 f8b4 f3e5 b4c3 d2c3
Best move: b1c3
 duration: 10022ms 
 max depth: 7
 total nodes searched: 3112073
killing child
test main ... ok

test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 10.02s

***********************
 
prune on SOURCE branch:
-----------------------
�[1m�[32m   Compiling�[0m chesseng v0.6.3 (/home/runner/work/chesseng/chesseng)
�[1m�[32m    Finished�[0m `test` profile [unoptimized + debuginfo] target(s) in 0.35s
�[1m�[32m     Running�[0m tests/prune.rs (target/debug/deps/prune-1fe93b2c3120160d)

running 1 test
[*] creating engine at version 0.6.3
[*] tt: created VL table with 8 entries.
[*] Entering UCI mode
engine_info (0ms): info string entering uci mode
engine_info (1ms): info string resizing table...
engine_info (14ms): info string table resized to 1048576 entries in 0.01331128s
engine_info (24ms): info depth 1 seldepth 0 multipv 1 nodes 60 nps 109540 hashfull 0 tbhits 0 time 0 score cp 10 pv d2d4
engine_info (24ms): info depth 2 seldepth 1 multipv 1 nodes 888 nps 147980 hashfull 0 tbhits 0 time 6 score cp 25 pv g1f3 g8f6
engine_info (45ms): info depth 3 seldepth 4 multipv 1 nodes 3857 nps 128624 hashfull 0 tbhits 0 time 29 score cp 10 pv g1f3 g8f6 d2d4
engine_info (186ms): info depth 4 seldepth 4 multipv 1 nodes 22739 nps 133724 hashfull 0 tbhits 15 time 170 score cp 25 pv g1f3 g8f6 d2d4 d7d5
engine_info (881ms): info depth 5 seldepth 9 multipv 1 nodes 108663 nps 125731 hashfull 8 tbhits 2405 time 864 score cp 8 pv g1f3 g8f6 d2d4 d7d5 b1c3
engine_info (3419ms): info depth 6 seldepth 10 multipv 1 nodes 455256 nps 133735 hashfull 16 tbhits 5879 time 3404 score cp 25 pv g1f3 g8f6 d2d4 d7d5 b1c3 b8c6 c3d5
engine_info (20825ms): info depth 7 seldepth 14 multipv 1 nodes 2554082 nps 122770 hashfull 152 tbhits 56907 time 20803 score cp 6 pv g1f3 g8f6 e2e3 d7d5 f1d3 b8c6 e1g1
Best move: g1f3
 duration: 20986ms 
 max depth: 7
 total nodes searched: 3145545
killing child
test main ... ok

test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 20.99s

github-actions · 2025-08-21T11:19:28Z

1m benchmark results:

1m on TARGET branch:
--------------------
�[1m�[32m    Finished�[0m `test` profile [unoptimized + debuginfo] target(s) in 0.07s
�[1m�[32m     Running�[0m tests/1m.rs (target/debug/deps/1m-33278a205143e4b3)

running 1 test
[*] creating engine at version 0.6.2
[*] tt: created VL table with 8 entries.
[*] Entering UCI mode
engine_info (0ms): info string entering uci mode
engine_info (1ms): info string table resized to 8 entries.
[*] uci: option use_tt set to on.
engine_info (1ms): info string table resized to 1024 entries.
engine_info (1ms): info string table resized to 1024 entries.
engine_info (16ms): info depth 1 seldepth 1 multipv 1 nodes 40 nps 71219 hashfull 0 tbhits 0 time 0 score cp 11 pv g1f3
engine_info (16ms): info depth 2 seldepth 2 multipv 1 nodes 860 nps 382305 hashfull 19 tbhits 0 time 2 score cp 25 pv g1f3 g8f6
engine_info (16ms): info depth 3 seldepth 3 multipv 1 nodes 2852 nps 420765 hashfull 311 tbhits 0 time 6 score cp 10 pv g1f3 g8f6 d2d4
engine_info (64ms): info depth 4 seldepth 4 multipv 1 nodes 27035 nps 498870 hashfull 684 tbhits 22 time 54 score cp 20 pv g1f3 b8c6 b1c3 c6d4
engine_info (248ms): info depth 5 seldepth 6 multipv 1 nodes 109643 nps 462454 hashfull 1000 tbhits 1638 time 237 score cp 41 pv g1f3 b8c6 f3g5 f7f5 g5f7
engine_info (3270ms): info depth 6 seldepth 8 multipv 1 nodes 1494331 nps 458749 hashfull 1000 tbhits 7093 time 3257 score cp -38 pv b1c3 e7e6 c3b5 f8c5 e2e3 c5e3
engine_info (8139ms): info depth 7 seldepth 10 multipv 1 nodes 3857383 nps 474113 hashfull 1000 tbhits 40589 time 8135 score cp 77 pv b1c3 e7e6 g1f3 f8b4 f3e5 b4c3 d2c3
test main has been running for over 60 seconds
Best move: b1c3
 duration: 90006ms 
 max depth: 7
 total nodes searched: 5492144
killing child
test main ... ok

test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 90.01s

********************
 
1m on SOURCE branch:
--------------------
�[1m�[32m   Compiling�[0m chesseng v0.6.3 (/home/runner/work/chesseng/chesseng)
�[1m�[32m    Finished�[0m `test` profile [unoptimized + debuginfo] target(s) in 1.93s
�[1m�[32m     Running�[0m tests/1m.rs (target/debug/deps/1m-6e6d4702253f18bd)

running 1 test
[*] creating engine at version 0.6.3
[*] tt: created VL table with 8 entries.
[*] Entering UCI mode
engine_info (0ms): info string entering uci mode
[*] uci: option use_tt set to on.
engine_info (14ms): info depth 1 seldepth 0 multipv 1 nodes 60 nps 94117 hashfull 0 tbhits 0 time 0 score cp 10 pv d2d4
engine_info (14ms): info depth 2 seldepth 1 multipv 1 nodes 896 nps 297450 hashfull 750 tbhits 0 time 3 score cp 25 pv g1f3 g8f6
engine_info (23ms): info depth 3 seldepth 4 multipv 1 nodes 4740 nps 294333 hashfull 1000 tbhits 0 time 16 score cp 10 pv g1f3 g8f6 d2d4
engine_info (93ms): info depth 4 seldepth 5 multipv 1 nodes 28279 nps 335160 hashfull 1000 tbhits 2 time 84 score cp 25 pv g1f3 d7d5 d2d4 g8f6
engine_info (520ms): info depth 5 seldepth 9 multipv 1 nodes 157222 nps 304551 hashfull 1000 tbhits 2 time 516 score cp 8 pv g1f3 d7d5 d2d4 g8f6 b1c3
engine_info (3049ms): info depth 6 seldepth 10 multipv 1 nodes 937466 nps 308682 hashfull 1000 tbhits 4 time 3036 score cp 25 pv g1f3 d7d5 d2d4 b8c6 b1c3 g8f6 c3d5
engine_info (20608ms): info depth 7 seldepth 13 multipv 1 nodes 5935802 nps 288170 hashfull 1000 tbhits 15 time 20598 score cp 6 pv g1f3 d7d5 e2e3 b8c6 f1d3 g8f6 e1g1
test main has been running for over 60 seconds
Best move: g1f3
 duration: 90009ms 
 max depth: 7
 total nodes searched: 7064465
killing child
test main ... ok

test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 90.01s

github-actions · 2025-08-21T11:25:16Z

Search benchmark results:

search_bench/id_pos_0   time:   [76.282 µs 78.316 µs 80.870 µs]
                        change: [-32.430% -11.310% +20.337%] (p = 0.46 > 0.05)
                        No change in performance detected.
Found 10 outliers among 100 measurements (10.00%)
  1 (1.00%) low mild
  4 (4.00%) high mild
  5 (5.00%) high severe
search_bench/id_pos_1   time:   [56.514 µs 60.303 µs 64.417 µs]
                        change: [-73.791% -59.284% -36.109%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 4 outliers among 100 measurements (4.00%)
  2 (2.00%) high mild
  2 (2.00%) high severe
search_bench/id_pos_2   time:   [60.105 µs 63.806 µs 67.785 µs]
                        change: [-61.951% -43.551% -12.902%] (p = 0.03 < 0.05)
                        Performance has improved.
Found 8 outliers among 100 measurements (8.00%)
  5 (5.00%) high mild
  3 (3.00%) high severe
search_bench/id_pos_3   time:   [68.759 µs 88.884 µs 111.16 µs]
                        change: [-46.552% -22.577% +12.375%] (p = 0.21 > 0.05)
                        No change in performance detected.
Found 22 outliers among 100 measurements (22.00%)
  5 (5.00%) high mild
  17 (17.00%) high severe

github-actions · 2025-08-21T11:26:32Z

Negamax benchmark results:

negamax_bench/ngm_full_depth_3
                        time:   [752.15 µs 753.51 µs 755.67 µs]
                        change: [+439.38% +441.08% +443.03%] (p = 0.00 < 0.05)
                        Performance has regressed.
Found 7 outliers among 100 measurements (7.00%)
  2 (2.00%) high mild
  5 (5.00%) high severe
negamax_bench/ngm_full_depth_4
                        time:   [444.83 µs 445.83 µs 446.88 µs]
                        change: [+159.55% +160.05% +160.60%] (p = 0.00 < 0.05)
                        Performance has regressed.
Found 6 outliers among 100 measurements (6.00%)
  3 (3.00%) high mild
  3 (3.00%) high severe
negamax_bench/ngm_full_depth_5
                        time:   [507.43 µs 508.41 µs 509.50 µs]
                        change: [+143.34% +146.18% +148.13%] (p = 0.00 < 0.05)
                        Performance has regressed.
Found 5 outliers among 100 measurements (5.00%)
  2 (2.00%) high mild
  3 (3.00%) high severe
negamax_bench/ngm_full_depth_6
                        time:   [411.05 µs 411.63 µs 412.28 µs]
                        change: [+71.865% +72.238% +72.704%] (p = 0.00 < 0.05)
                        Performance has regressed.
Found 6 outliers among 100 measurements (6.00%)
  4 (4.00%) high mild
  2 (2.00%) high severe

github-actions · 2025-08-21T11:27:14Z

Evaluation benchmark results:

eval_bench/eval_full    time:   [452.04 µs 452.33 µs 452.61 µs]
                        change: [-0.6653% -0.1759% +0.2176%] (p = 0.49 > 0.05)
                        No change in performance detected.
Found 2 outliers among 100 measurements (2.00%)
  2 (2.00%) high severe

github-actions · 2025-08-21T11:28:12Z

MoveGen benchmark results:

move_gen_bench/unordered
                        time:   [790.37 µs 791.05 µs 791.87 µs]
                        change: [-0.5092% -0.2654% -0.0799%] (p = 0.01 < 0.05)
                        Change within noise threshold.
Found 8 outliers among 100 measurements (8.00%)
  4 (4.00%) high mild
  4 (4.00%) high severe
move_gen_bench/ordered  time:   [916.53 µs 917.45 µs 918.65 µs]
                        change: [+1.0325% +1.3720% +2.0448%] (p = 0.00 < 0.05)
                        Change within noise threshold.
Found 3 outliers among 100 measurements (3.00%)
  3 (3.00%) high severe

andtsa force-pushed the cleaning branch from a78944b to 05b950d Compare April 27, 2025 11:57

andtsa changed the base branch from cleaning to main April 27, 2025 11:57

andtsa changed the base branch from main to cleaning April 27, 2025 11:57

This comment was marked as outdated.

Sign in to view

andtsa mentioned this pull request Apr 28, 2025

spring cleaning & move ordering #38

Closed

This comment was marked as outdated.

Sign in to view

andtsa changed the base branch from cleaning to main April 28, 2025 11:55

This comment was marked as outdated.

Sign in to view

andtsa added 9 commits August 21, 2025 14:11

spring cleaning

b9f363d

mvv-lva

3ef4ed7

assert for total order

0c137f6

better threefold prevention

d210b25

remove mvv-lva

6c8d86a

stronger threefold repetition prevention

a56e6cc

(unoptimised) quiescence search

f1246f6

reduce contention on Opts lock

37aea09

uci changes

b9dec81

Signed-off-by: Andreas Tsatsanis <[email protected]>

andtsa force-pushed the quiescence branch from 23d8db9 to b9dec81 Compare August 21, 2025 11:15

Quiescence search #39

Are you sure you want to change the base?

Quiescence search #39

Uh oh!

Conversation

andtsa commented Apr 27, 2025

Uh oh!

This comment was marked as outdated.

This comment was marked as outdated.

This comment was marked as outdated.

This comment was marked as outdated.

This comment was marked as outdated.

This comment was marked as outdated.

This comment was marked as outdated.

This comment was marked as outdated.

This comment was marked as outdated.

This comment was marked as outdated.

This comment was marked as outdated.

This comment was marked as outdated.

This comment was marked as outdated.

This comment was marked as outdated.

github-actions bot commented Aug 21, 2025

Uh oh!

github-actions bot commented Aug 21, 2025

Uh oh!

github-actions bot commented Aug 21, 2025

Uh oh!

github-actions bot commented Aug 21, 2025

Uh oh!

github-actions bot commented Aug 21, 2025

Uh oh!

github-actions bot commented Aug 21, 2025

Uh oh!

github-actions bot commented Aug 21, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants