Stockfish Testing Queue

Pending - 3 tests 9.0 hrs

26-07-16 pe fail_high1 diff
LLR: -1.76 (-2.94,2.94) [0.00,5.00]
Total: 33957 W: 5799 L: 5735 D: 22423
sprt @ 5+0.05 th 7 See if skipping fail highs at root works for helpers also
15-07-16 aj smp_thread_skip diff
LLR: -0.05 (-2.94,2.94) [0.00,5.00]
Total: 13651 W: 2070 L: 2022 D: 9559
sprt @ 5+0.05 th 23 Don't let threads deviate more than 6 PLY from main thread: STC (Profiling shows that this happens only at very high thread counts. At low threads counts this hardly makes a difference)
28-07-16 Fi twoThirdsDensity diff
LLR: 0.37 (-2.94,2.94) [0.00,5.00]
Total: 13000 W: 2226 L: 2158 D: 8616
sprt @ 5+0.05 th 7 Use two thirds density map instead of half for skipping.

Active - 57 machines 217 cores 1.53M nps (331.66M total nps) 642 games/minute

Machine Cores MNps System Version Running on Last updated
Bobo1239
1
0.00 Linux 4.6.4-1-ARCH 59 blocked_p2 seconds ago
IgorLeMasson
2
0.97 Windows 10 59 blocked_p2 seconds ago
mibere
14
1.60 Linux 3.16.0-4-amd64-2 59 blocked_p2 seconds ago
cw
3
1.11 Windows 7 59 blocked_p2 seconds ago
cw
3
1.14 Windows 7 59 blocked_p2 seconds ago
slakovv
12
1.71 Linux 4.4.0-28-generic 59 blocked_p2 seconds ago
snicolet
3
1.58 Darwin 13.4.0 59 blocked_p2 seconds ago
Thanar
3
1.58 Windows 7 59 blocked_p2 seconds ago
eastorwest
4
1.42 Linux 4.4.0-31-generic 59 blocked_p2 seconds ago
robnjr
5
2.55 Windows 7 59 blocked_p2 seconds ago
stocky
3
2.17 Linux 3.13.0-87-generic 59 blocked_p2 seconds ago
sergeballif
1
0.99 Windows 7 59 blocked_p2 seconds ago
ctoks
2
1.09 Windows 10 59 blocked_p2 seconds ago
yurikvelo
1
1.01 Windows 8 59 blocked_p2 seconds ago
lem
3
1.22 Linux 4.5.4-1-ARCH 59 blocked_p2 seconds ago
TheDarkHorde
2
1.66 Linux 4.6.4-1-ARCH 59 blocked_p2 seconds ago
hyperbolic.tom
3
1.34 Linux 3.13.0-86-generic 59 blocked_p2 seconds ago
velislav
2
1.60 Linux 3.10.0-123.8.1.el7.x86_64 59 blocked_p2 seconds ago
leszek
3
1.55 Windows 2003Server 59 blocked_p2 seconds ago
Fisherman
6
1.51 Windows 8 59 blocked_p2 seconds ago
ttruscott
3
1.65 Windows 7 59 blocked_p2 seconds ago
Thanar
3
1.52 Windows 7 59 blocked_p2 seconds ago
rpentomino186
3
1.84 Linux 4.6.0-1.slh.1-aptosid-amd64 59 blocked_p2 seconds ago
sqrt2
3
2.16 Linux 3.16.0-4-amd64 59 blocked_p2 seconds ago
psk
3
2.43 Linux 3.13.0-88-generic 59 blocked_p2 seconds ago
ako027ako
3
1.89 Windows 10 59 blocked_p2 seconds ago
mibere
7
0.86 Linux 3.16.0-4-amd64-6 59 blocked_p2 seconds ago
cw
3
1.43 Windows 7 59 blocked_p2 seconds ago
oryx
5
2.17 Windows 8 59 blocked_p2 seconds ago
mibere
5
1.60 Linux 3.16.0-4-amd64-1 59 blocked_p2 seconds ago
vdbergh
5
1.45 Linux 2.6.32-504.23.4.el6.x86_64 59 blocked_p2 seconds ago
velislav
1
2.29 Linux 4.4.0-31-generic 59 blocked_p2 seconds ago
chris
3
2.17 Windows 10 59 blocked_p2 seconds ago
anst
4
1.86 Windows 8 59 blocked_p2 seconds ago
JojoM
3
1.88 Linux 3.19.0-65-generic 59 blocked_p2 seconds ago
homyur
2
2.07 Windows 7 59 blocked_p2 seconds ago
drabel
1
1.37 Windows 8 59 blocked_p2 seconds ago
mibere
3
1.06 Linux 3.16.0-4-amd64-4 59 blocked_p2 seconds ago
joster
3
2.11 Linux 3.19.0-26-generic 59 blocked_p2 seconds ago
tvijlbrief
3
2.10 Linux 3.13.0-30-generic 59 blocked_p2 seconds ago
fastgm
4
1.41 Windows 8 59 blocked_p2 seconds ago
mibere
7
1.35 Linux 3.16.0-4-amd64-3 59 blocked_p2 seconds ago
mibere
7
1.49 Linux 3.16.0-4-amd64-8 59 blocked_p2 seconds ago
crunchy
7
1.28 Windows 7 59 blocked_p2 seconds ago
leszek
7
0.94 Windows 2003Server 59 blocked_p2 seconds ago
davar
3
1.76 Windows 8 59 blocked_p2 seconds ago
mibere
7
1.33 Linux 3.16.0-4-amd64-7 59 blocked_p2 seconds ago
Medvezhonok
3
2.02 Windows 7 59 blocked_p2 seconds ago
rkl
1
2.25 Linux 3.10.0-327.22.2.el7.x86_64 59 tune_doubled seconds ago
31m059
3
0.92 Windows 7 59 tune_doubled seconds ago
JojoM
1
1.81 Windows 10 59 tune_doubled seconds ago
mhoram
2
0.73 Linux 3.16.0-4-amd64 59 tune_doubled seconds ago
fp53fish
1
1.06 Linux 3.19.0-32-generic 59 tune_doubled seconds ago
biffhero
1
1.24 Linux 3.16.0-4-amd64 59 tune_doubled seconds ago
cw
5
0.71 Windows 7 59 tune_doubled seconds ago
horst.prack
1
1.21 Linux 4.6.0-1-amd64 59 tune_doubled seconds ago
fastgm
15
1.49 Windows 10 59 smp_thread_skip 16 minutes ago
28-07-16 El blocked_p2 diff
LLR: -2.47 (-2.94,2.94) [0.00,5.00]
Total: 20363 W: 3853 L: 3869 D: 12641
sprt @ 10+0.1 th 1 Blocked pawns in evaluate_initiative
28-07-16 jo tune_doubled diff
27385/30000 iterations
55705/60000 games played
60000 @ 20+0.2 th 1 Try to retune doubled pawn penalty per file.
27-07-16 aj smp_thread_skip diff
LLR: -0.34 (-2.94,2.94) [0.00,5.00]
Total: 472 W: 56 L: 68 D: 348
sprt @ 10+0.1 th 15 This test has run at 23 threads for two weeks! It is unlikely to finish in the near future. Run at 15 threads since there are more such machines and it can realistically complete

Finished - 12168 tests

28-07-16 Vo cft_adj diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 16101 W: 3000 L: 3101 D: 10000
sprt @ 10+0.1 th 1 Increase cft's stat weight.
28-07-16 pe fail_high3 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 8761 W: 1489 L: 1577 D: 5695
sprt @ 5+0.05 th 7 Main thread skips fails highs (passed STC & LTC), helper threads resolve fail highs only if best move became different from that of main thread
28-07-16 Vo lmr-avg diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 17374 W: 3186 L: 3237 D: 10951
sprt @ 10+0.1 th 1 Try averaging out the stats, excluding 0 values. Based off and against cft.
28-07-16 SC double_reductions diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 30567 W: 4182 L: 4197 D: 22188
sprt @ 60+0.6 th 1 Tuned values. LTC.
27-07-16 Fi pawnSpanInitiative2 diff
LLR: -2.93 (-2.94,2.94) [0.00,5.00]
Total: 19488 W: 3713 L: 3752 D: 12023
sprt @ 10+0.1 th 1 My final attempt at this. See commit notes.
27-07-16 Vo cft diff
LLR: 2.95 (-2.94,2.94) [0.00,5.00]
Total: 32759 W: 4600 L: 4370 D: 23789
sprt @ 60+0.6 th 1 Sanity check- Since patch has been rewritten and a line of code was modify which changed the bench.
27-07-16 SC double_reductions diff
LLR: 2.95 (-2.94,2.94) [0.00,5.00]
Total: 22320 W: 4378 L: 4152 D: 13790
sprt @ 10+0.1 th 1 Tuned values.
27-07-16 lu knight_dist_ks diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 14274 W: 2759 L: 2822 D: 8693
sprt @ 10+0.1 th 1 Remote knight threats in king safety (tiny penalty).
27-07-16 II delta_new diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 10473 W: 2045 L: 2125 D: 6303
sprt @ 10+0.1 th 1 Last try with formula for delta; now increasing delta with depth.
27-07-16 II rf_new diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 8880 W: 1627 L: 1714 D: 5539
sprt @ 10+0.1 th 1 Delta/reduction combination. Also last try. Moving on after these two attempts.
27-07-16 SC double_reductions_tune diff
39869/50000 iterations
81176/100000 games played
100000 @ 10+0.1 th 1 Try to exploit double valued reductions. Reschedule as not converged.
27-07-16 pe fail_high2 diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 12893 W: 2139 L: 2211 D: 8543
sprt @ 5+0.05 th 7 Try to be more selective which fail highs to skip. The most critical part seems to be smp performance. I assume single thread will pass, as it benefits a lot from skipping.
27-07-16 pb nms_consider_hanging diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 6100 W: 1103 L: 1203 D: 3794
sprt @ 10+0.1 th 1 better calculate the static eval (tempo) in the recently added condition for null move search. Don't run this directly against master as moving EvalInfo allocation to search seems to be an impact on performance, see pull 739 (fixed bench)
24-07-16 mc master diff
ELO: 49.73 +-1.7 (95%) LOS: 100.0%
Total: 40000 W: 8102 L: 2415 D: 29483
40000 @ 60+0.6 th 1 Regression test
27-07-16 lu knight_dist diff
LLR: -2.94 (-2.94,2.94) [0.00,5.00]
Total: 8653 W: 1629 L: 1717 D: 5307
sprt @ 10+0.1 th 1 Hopefully fixed all bugs, with a few adjustments. (Low throughput)
27-07-16 sg move_count_pruning diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 6987 W: 1247 L: 1343 D: 4397
sprt @ 10+0.1 th 1 I have collected some stats. Based on them i try it now with a depth dependent formula.
27-07-16 Vo yc2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 22181 W: 4167 L: 4196 D: 13818
sprt @ 10+0.1 th 1 Yellow Combo attempt...(Fixed Bench)
27-07-16 jo fail_high diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 3081 W: 513 L: 660 D: 1908
sprt @ 10+0.1 th 1 Fully open beta already after the first fail-high. Leave alpha as is.
24-07-16 aj tune_king_safety diff
4293/30000 iterations
8867/60000 games played
60000 @ 60+0.6 th 1 Tune king safety at LTC. Run at priority -2(for an empty framework). Use much lower ck values for parameters around the lingering.
26-07-16 SC pawnSpanInitiative diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 106913 W: 20736 L: 20385 D: 65792
sprt @ 10+0.1 th 1 Very small bonus. Take 3 and call it a day.
26-07-16 pe fail_high1 diff
LLR: 2.95 (-2.94,2.94) [0.00,5.00]
Total: 17434 W: 3386 L: 3183 D: 10865
sprt @ 5+0.05 th 3 Helpers skip resolving fail highs . I want to see how this work separately with skip size 1. 7 thread test will put pri -1 for now
26-07-16 pe fail_high1 diff
LLR: -2.45 (-2.94,2.94) [0.00,5.00]
Total: 10000 W: 1670 L: 1732 D: 6598
sprt @ 5+0.05 th 7 Leave resolving fail high for helpers
26-07-16 SC double_reductions_tune diff
27530/30000 iterations
55730/100000 games played
100000 @ 10+0.1 th 1 Try to exploit double valued reductions.
26-07-16 pe fail_high1 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 5626 W: 1025 L: 1127 D: 3474
sprt @ 5+0.05 th 3 Helpers resolve fail highs . I want to see how this work separately with skip size 1
26-07-16 lb tuned diff
LLR: -2.94 (-2.94,2.94) [0.00,5.00]
Total: 13406 W: 2503 L: 2570 D: 8333
sprt @ 10+0.1 th 1 tuned
26-07-16 Vo dangerSqs2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 8485 W: 1588 L: 1677 D: 5220
sprt @ 10+0.1 th 1 One more try at this idea...
24-07-16 lb tune diff
48816/40000 iterations
98702/100000 games played
100000 @ 20+0.2 th 1 tune relative weights of cm, fm, fm2
26-07-16 II rf_new diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 13072 W: 2458 L: 2527 D: 8087
sprt @ 10+0.1 th 1 I plan to use this reduction formula in infinite analysis mode, because I'm stick to an idea that reductions should be lower on higher depths. Give it a try in the framework.
26-07-16 pe fail_high1 diff
LLR: 2.95 (-2.94,2.94) [0.00,5.00]
Total: 2031 W: 516 L: 375 D: 1140
sprt @ 5+0.05 th 1 Fast test at 5+0.05 for 1 thread to make sure that gain is still there at very short tc and to use as base for comparison with 7 thread tests
26-07-16 SC pawnSpanInitiative diff
LLR: -2.99 (-2.94,2.94) [0.00,5.00]
Total: 10836 W: 2114 L: 2194 D: 6528
sprt @ 10+0.1 th 1 Much larger bonus.
26-07-16 tv mcpruning diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 3022 W: 527 L: 641 D: 1854
sprt @ 10+0.1 th 1 No Movecount pruning when no ttHit
24-07-16 SC double_reductions diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 80669 W: 15286 L: 15058 D: 50325
sprt @ 10+0.1 th 1 Take 2: rounding instead of converting.
25-07-16 SC pawnSpanInitiative diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 120031 W: 23335 L: 22925 D: 73771
sprt @ 10+0.1 th 1 Pawn span initiative only for strong side.
26-07-16 lu knight_dist diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 7213 W: 1343 L: 1438 D: 4432
sprt @ 10+0.1 th 1 Enemy pieces at n knight steps: I wish to test the idea before refining...
26-07-16 pe fail_high1 diff
LLR: 2.96 (-2.94,2.94) [0.00,5.00]
Total: 8631 W: 1277 L: 1128 D: 6226
sprt @ 60+0.6 th 1 LTC. Skip resolving fail highs at root and partially compensate for saved time.
26-07-16 pe fail_high1 diff
LLR: 2.96 (-2.94,2.94) [0.00,5.00]
Total: 6659 W: 1360 L: 1203 D: 4096
sprt @ 10+0.1 th 1 Wondering how bad it will fail if I completely remove resolving fail highs at root. Here I partially compensate for saved time.
24-07-16 SC double_reductions diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 115588 W: 22437 L: 22047 D: 71104
sprt @ 10+0.1 th 1 And take 3: adjust history contribution to have a reasonable bench. I'll wait the results of these three before deciding whether to stop or go on.
25-07-16 SC pawnSpanInitiative diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 22860 W: 4378 L: 4403 D: 14079
sprt @ 10+0.1 th 1 Or, alternatively, the difference of pawn spans.
25-07-16 pe fail_high diff
LLR: 2.96 (-2.94,2.94) [0.00,5.00]
Total: 11063 W: 2258 L: 2080 D: 6725
sprt @ 10+0.1 th 1 Ignore fail high at root if not failed low
25-07-16 sg move_count_pruning diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 11446 W: 2206 L: 2282 D: 6958
sprt @ 10+0.1 th 1 Take 2 seems promising, so try even greater margin of 3 * pawns. Take 3
25-07-16 Vo capOrder diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 12880 W: 2439 L: 2509 D: 7932
sprt @ 10+0.1 th 1 Use from square to try improve capture order.
25-07-16 jo check_for_draw diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 22427 W: 4268 L: 4295 D: 13864
sprt @ 10+0.1 th 1 Take 2a, bugfix. (Include qsearch.)
25-07-16 sg move_count_pruning diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 41606 W: 7912 L: 7855 D: 25839
sprt @ 10+0.1 th 1 Do move count pruning only if static eval <= beta + 2 * pawn value. Take 2
25-07-16 jo check_for_draw diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 24108 W: 4530 L: 4550 D: 15028
sprt @ 10+0.1 th 1 The side to move gives up the tempo bonus. Take 2.
25-07-16 SC double_reductions diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 11893 W: 2253 L: 2327 D: 7313
sprt @ 10+0.1 th 1 My experiment seems quite positive, so make some more attempts. Take 4: use same rounding strategy in shallow depth pruning.
25-07-16 jo check_for_draw diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 22995 W: 4345 L: 4370 D: 14280
sprt @ 10+0.1 th 1 Check for a draw separately. In all other pruning methods we have some conditions whether to take the cutoff or not. Not so when checking for a draw. Start with the obvious 50-move counter. Take 1. (I know, a draw is a draw. But maybe this gives some inspiration to others.)
25-07-16 SC double_reductions diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 12033 W: 2233 L: 2307 D: 7493
sprt @ 10+0.1 th 1 And (on top) cut node reductions 1.5. Take 5.
25-07-16 sg move_count_pruning diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 19074 W: 3701 L: 3742 D: 11631
sprt @ 10+0.1 th 1 Do move count pruning only if static eval <= beta + pawn.
25-07-16 aj king_safety_tuned3 diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 19402 W: 3644 L: 3734 D: 12024
sprt @ 10+0.1 th 1 Try only the Check parameters that were tuned with SPSA: STC
21-07-16 pb lazy_high_density diff
ELO: -0.00 +-4.5 (95%) LOS: 50.0%
Total: 5431 W: 649 L: 649 D: 4133
10000 @ 20+0.2 th 15 As queue is almost empty, trying high_density at LTC with 15 threads to see if we have the same effect as we had on 7 threads. (Last try if the outcome is negative)