Hard-Talkchess-2020 set, final release

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

Ras
Posts: 2557
Joined: Tue Aug 30, 2016 8:19 pm
Full name: Rasmus Althoff

Re: Hard-Talkchess-2020 set, final release

Post by Ras »

AlexChess wrote: Sun May 29, 2022 11:09 am(with all 4 CPUs it becomes too hot).
No, it doesn't, not even the fanless Macbook Air. It will throttle itself to stay at a level that Apple thinks is fine. Expect a CPU temperature around 95°C and a performance drop of around 15% (that's why the Macbook Pro has a fan) compared to the cold state, provided that Stockfish behaves similarly to rendering.
Rasmus Althoff
https://www.ct800.net
User avatar
AlexChess
Posts: 1536
Joined: Sat Feb 06, 2021 8:06 am
Full name: Alex Morales

Re: Hard-Talkchess-2020 set, final release

Post by AlexChess »

Ras wrote: Tue Jun 07, 2022 9:15 pm
AlexChess wrote: Sun May 29, 2022 11:09 am(with all 4 CPUs it becomes too hot).
No, it doesn't, not even the fanless Macbook Air. It will throttle itself to stay at a level that Apple thinks is fine. Expect a CPU temperature around 95°C and a performance drop of around 15% (that's why the Macbook Pro has a fan) compared to the cold state, provided that Stockfish behaves similarly to rendering.
Thank you Ras, your comments are always useful and I trust you. :D I'll make the test. How many seconds | minutes as time limit for my hardware that calculates 2 MN/s on starting position?
Chess engines and dedicated chess computers fan since 1981 :D Mac mini M1 8GB-256GB, Windows 11 & Ubuntu ARM64.
ProteusSF Dev Forum TROLLS KINDERGARTEN
Ras
Posts: 2557
Joined: Tue Aug 30, 2016 8:19 pm
Full name: Rasmus Althoff

Re: Hard-Talkchess-2020 set, final release

Post by Ras »

AlexChess wrote: Tue Jun 07, 2022 9:29 pmThank you Ras, your comments are always useful and I trust you. :D
Thx! :) Just for reference, that was tested in this review:
How many seconds | minutes as time limit for my hardware that calculates 2 MN/s on starting position?
Given that the whole device has to heat up to a steady state, and here is another review with 40 minutes of full-blown rendering, I'd ballbark it at around this duration.
Rasmus Althoff
https://www.ct800.net
User avatar
AlexChess
Posts: 1536
Joined: Sat Feb 06, 2021 8:06 am
Full name: Alex Morales

Re: Hard-Talkchess-2020 set, final release

Post by AlexChess »

Ras wrote: Tue Jun 07, 2022 9:44 pm
AlexChess wrote: Tue Jun 07, 2022 9:29 pmThank you Ras, your comments are always useful and I trust you. :D
Thx! :) Just for reference, that was tested in this review:
How many seconds | minutes as time limit for my hardware that calculates 2 MN/s on starting position?
Given that the whole device has to heat up to a steady state, and here is another review with 40 minutes of full-blown rendering, I'd ballbark it at around this duration.
Thank you again, scheduled the test, I will publish Hard-Talkchess 108 positions tests here:

https://banksiagui.com/forums/viewtopic.php?p=132#p132

Kind regards, Alex
Chess engines and dedicated chess computers fan since 1981 :D Mac mini M1 8GB-256GB, Windows 11 & Ubuntu ARM64.
ProteusSF Dev Forum TROLLS KINDERGARTEN
EvgeniyZh
Posts: 43
Joined: Fri Sep 19, 2014 4:54 pm
Location: Israel

Re: Hard-Talkchess-2020 set, final release

Post by EvgeniyZh »

Vinvin wrote: Sun Dec 19, 2021 10:50 pm I removed 6 positions considered as incorrect or not good enough.
So 108 positions remain.

Code: Select all

Position #038 not good : several moves are wining
Position #069 not good : several moves are wining
Position #144 is not good enough because 3 moves are clearly winning.
Position #169 is not good enough because 2 moves are clearly winning : Ng5 and Bxh7+
Position #177 is not good enough because 2 moves are clearly winning : N3h4 and Qxe1
Position #187 is not good enough because 2 moves are clearly winning : Ng4+ and Kg7
Only 6 positions removed, but that's enough to change a bit the ranking :

List sorted from best to worst :

Code: Select all

Name / #found-average / (Avg Time with penalty in seconds) / number of runs
ShashChess18.2 :         98,1  (109) on 11 runs
SugaR AI ICCF 1.90 :     98,0  (103) on 6 runs
Blue Marlin 14.4a :      97,5  (109) on 11 runs
ShashChess20.1 :         96,4  (116) on 10 runs
ShashChess20 :           95,9  (119) on 10 runs
ShashChess17.1 :         95,2  (131) on 6 runs
Blue Marlin 14.5 :       94,9  (125) on 13 runs
Crystal 3.2 :            94,0  (127) on 6 runs
SugaR AI ICCF 2.40 :     93,6  (126) on 14 runs
Honey-v14 :              93,4  (137) on 8 runs
Crystal 3.1 :            93,3  (131) on 8 runs
Stockfish_21.11.23.21 :  91,0  (156) on 5 runs
SugaR AI ICCF 2.50 :     90,3  (142) on 10 runs
Stockfish_21.08.05.16 :  88,8  (169) on 5 runs 
Sheet with the 108 positions and timings : https://www.dropbox.com/s/8dqjc2t5pvfrw ... 9.ods?dl=0
Position #95, what's the problem with Rxd2?
Vinvin
Posts: 5236
Joined: Thu Mar 09, 2006 9:40 am
Full name: Vincent Lejeune

Re: Hard-Talkchess-2020 set, final release

Post by Vinvin »

EvgeniyZh wrote: Mon Aug 01, 2022 10:02 am Position #95, what's the problem with Rxd2?
Rxd2 is draw but Kf3 is winning.
ernst
Posts: 354
Joined: Thu Mar 09, 2006 6:00 pm

Re: Hard-Talkchess-2020 set, final release

Post by ernst »

Vinvin wrote: Mon Aug 01, 2022 3:31 pm
EvgeniyZh wrote: Mon Aug 01, 2022 10:02 am Position #95, what's the problem with Rxd2?
Rxd2 is draw but Kf3 is winning.
It is a nullmove problem. SF with nullmove disabled finds it fast.
39/33 0:01 0.00 1.Rxd2 Qc1 2.Bxf4 Qc6+ 3.Kf2 Qe6 4.Kg3 Qf5 5.Rf2 Kh7 6.Rf3 Qg6+ 7.Bg5 Qe4 8.Rf7+ Kg8 9.Rf6 Qe5+ 10.Kg2 Qe4+ (59.865.528) 42010 TB:3.251
40/29 0:01 0.00 1.Rxd2 Qc1 2.Bxf4 Qc6+ 3.Kf2 Qe6 4.Kg3 Qf5 5.Rf2 Kh7 6.Rf3 Qg6+ 7.Bg5 Qe4 8.Rf7+ Kg8 9.Rf6 Qe5+ 10.Kg2 Qe4+ 11.Rf3 Qe2+ 12.Rf2 Qe4+ (66.619.631) 42137 TB:3.859
41/41 0:02 0.00 1.Rxd2 Qc1 2.Bxf4 Qc6+ 3.Kf2 Kh7 4.Bg5 Qe6 5.Kg3 Qe5+ 6.Bf4 Qf5 7.Rd6 Qc2 8.Rd7+ Kg8 9.Rd2 Qg6+ 10.Kh2 Kh7 11.Rf2 Qe4 12.h5 Qd4 13.Kg3 Qd3+ 14.Kg4 (95.813.738) 42489 TB:7.347
42/31 0:04 0.00 1.Rxd2 Qc1 2.Bxf4 Qc6+ 3.Kg3 Qc3+ 4.Kg4 Qc8+ 5.Kg5 Qc5+ 6.Kg6 Qc6+ 7.Kf5 Qc8+ 8.Kg5 (198.521.412) 42400 TB:28.852
43/29 0:05 0.00 1.Rxd2 Qc1 2.Bxf4 Qc6+ 3.Kg3 Qc3+ 4.Kg4 Qc8+ 5.Kg5 Qc5+ 6.Kg6 Qc6+ 7.Kf5 Qc8+ 8.Kf6 Qc3+ 9.Kg6 (239.948.583) 42506 TB:40.636
44/19 0:05 0.00 1.Rxd2 Qc1 2.Bxf4 Qc6+ 3.Kg3 Qc3+ 4.Kg4 Qc8+ 5.Kg5 Qc5+ 6.Kg6 Qc6+ 7.Kf5 Qc8+ 8.Kf6 Qc3+ 9.Kg6 (244.952.159) 42519 TB:41.861
45/51 0:11 0.00 1.Rxd2 Qc1 2.Bxf4 Qc6+ 3.Kg3 Qc3+ 4.Kg4 Qc8+ 5.Kg5 Qc5+ 6.Kg6 Qc6+ 7.Kf5 Qc8+ (509.886.860) 42790 TB:145.701
46/44 0:12 0.00 1.Rxd2 Qc1 2.Bxf4 Qc6+ 3.Kg3 Qg6+ 4.Bg5 Qe4 5.Rf2 Qe1 6.Bf4 Qg1+ 7.Rg2 Qc5 8.Re2 Kh7 9.Re3 Kg7 10.Bg5 Qd6+ 11.Kg2 Qd2+ 12.Kf3 Qd5+ 13.Kf2 Kg6 14.Rf3 (536.103.088) 42991 TB:151.304
47/51 0:34 +0.07++ 1.Kf3 (1.537.221.306) 44319 TB:794.525
47/51 0:34 +0.15++ 1.Kf3 (1.544.001.531) 44319 TB:797.544
47/51 0:35 +0.25++ 1.Kf3 (1.553.289.523) 44327 TB:803.016
47/51 0:35 +0.39++ 1.Kf3 (1.561.397.969) 44342 TB:810.758
47/51 0:35 +0.58++ 1.Kf3 (1.572.892.001) 44364 TB:821.026
47/51 0:35 +0.81++ 1.Kf3 (1.585.767.742) 44389 TB:829.454
47/51 0:36 +1.12++ 1.Kf3 (1.611.667.400) 44410 TB:846.130
47/51 0:36 +1.50++ 1.Kf3 (1.624.789.552) 44434 TB:852.408
47/51 0:36 +1.99++ 1.Kf3 (1.642.199.045) 44454 TB:870.939
47/51 0:37 +2.61++ 1.Kf3 (1.673.630.207) 44565 TB:882.734
47/51 0:38 +3.39++ 1.Kf3 (1.709.393.369) 44725 TB:899.854
47/51 0:39 +4.38++ 1.Kf3 (1.763.766.975) 44944 TB:926.088
47/51 0:40 +5.62++ 1.Kf3 (1.832.122.217) 45190 TB:970.968
47/51 0:41 +7.18++ 1.Kf3 (1.905.045.546) 45475 TB:1.008.772
47/51 0:43 +9.13++ 1.Kf3 (2.012.909.453) 46073 TB:1.029.060
47/51 0:44 +4.43 1.Kf3 Bc3 2.Kxf4 Kh7 3.Kf5 Bxe5 4.Kxe5 Kh6 5.Kf4 Kh5 6.Kg3 a6 7.Rc5+ Kh6 8.Rc6+ Kh5 9.Rc2 a5 10.Rc5+ Kg6 11.h5+ Kf6 12.Rc2 Kg5 13.h6 Kg6 14.Kh4 (2.036.168.597) 46188 TB:1.034.785
This post may either be cause or result of misunderstandings.
ernst
Posts: 354
Joined: Thu Mar 09, 2006 6:00 pm

Re: Hard-Talkchess-2020 set, final release

Post by ernst »

I was incorrect; it isn't a nullmove problem. Stockfish with NNUE disabled finds it fast.

Code: Select all

 37/42	 0:01 	 0.00 	1.Rxd2 f3+ 2.Kxf3 Qc1 3.Bf4 Qc3+ 4.Kg4 Qc8+ 5.Kg5 Qc5+ 6.Kg6 Qc6+ 7.Kh5 Kh7 8.Kg4 Qe6+ 9.Kg3 Qe1+ 10.Rf2 a5 11.Kg2 Qe6 12.Bg5 Qe4+ 13.Kg3 Qe1 14.Kf3 (119.529.450) 71275  TB:11.754
 38/40	 0:01 	 0.00 	1.Rxd2 f3+ 2.Kxf3 Qc1 3.Bf4 Qc6+ 4.Kg3 Qe8 5.Rf2 a5 6.Bg5 Kh7 7.Rd2 Qe1+ 8.Kg2 Qe5 9.Kf3 Qe6 10.Kf4 Kg7 11.Rf2 Qxh3 12.Rd2 Kf7 13.Rc2 Qd7 14.Rf2 (120.969.587) 71284  TB:11.972
 39/41	 0:01 	 0.00 	1.Rxd2 f3+ 2.Kxf3 Qc1 3.Bf4 Qc6+ 4.Kg3 Qe8 5.Rf2 a5 6.Bg5 Kh7 7.Rd2 Qe1+ 8.Kg2 Qe5 9.Kf3 Qe6 10.Kf4 Kg7 11.Rf2 Qxh3 12.Rd2 Kf7 13.Rc2 Qd7 14.Rf2 (122.502.276) 71305  TB:12.261
 40/32	 0:01 	 0.00 	1.Rxd2 f3+ 2.Kxf3 Qc1 3.Bf4 Qc6+ 4.Kg3 Qe8 5.Rf2 a5 6.Bg5 Kh7 7.Rd2 Qe1+ 8.Kg2 Qe5 9.Kf3 Qe6 10.Kf4 Kg7 11.Kg3 Kg6 12.Rf2 Kh5 13.Kh2 Qe1 14.Rf6 (125.887.422) 71324  TB:12.574
 41/32	 0:01 	 0.00 	1.Rxd2 f3+ 2.Kxf3 Qc1 3.Bf4 Qc6+ 4.Kg3 Qe8 5.Rf2 a5 6.Bg5 Qe4 7.h5 Qe5+ 8.Kg4 Qe6+ 9.Rf5 Kh7 10.Kf4 Qe2 11.h6 Qxa2 12.Rf7+ Kg6 13.Rg7+ Kh5 14.h7 (129.844.501) 71303  TB:13.139
 42/32	 0:02 	 0.00 	1.Rxd2 f3+ 2.Kxf3 Qc1 3.Bf4 Qc6+ 4.Kg3 Qe8 5.Rf2 a5 6.Bg5 Qe4 7.h5 Qe5+ 8.Kg4 Qe6+ 9.Rf5 Kh7 10.Kf4 Qe2 11.h6 Qxa2 12.Rf7+ Kg6 13.Rg7+ Kh5 14.h7 (167.354.822) 71856  TB:16.740
 43/41	 0:03 	 0.00 	1.Rxd2 f3+ 2.Kxf3 Qc1 3.Bf4 Qc3+ 4.Kg4 Qc8+ 5.Kg5 Qc5+ 6.Kf6 Qf8+ 7.Kg6 Qg7+ 8.Kf5 Qf7+ 9.Kg4 Qe6+ 10.Kg3 Qe1+ 11.Rf2 Kh7 12.h5 Qg1+ 13.Rg2 Qe1+ 14.Kh2 (241.890.229) 72249  TB:29.545
 44/46	 0:03 	 0.00 	1.Rxd2 f3+ 2.Kxf3 Qc1 3.Bf4 Qc3+ 4.Kg4 Qc8+ 5.Kg5 Qc5+ 6.Kf6 Qc3+ 7.Kg6 Qc6+ 8.Kf5 Kf7 9.Kg4 Qe6+ 10.Kg3 Qe1+ 11.Rf2 Kg6 12.Kg2 Qe7 13.Bg5 Qe4+ 14.Kh2 (272.270.198) 72393  TB:34.287
 45/45	 0:04 	 0.00 	1.Rxd2 f3+ 2.Kxf3 Qc1 3.Bf4 Qc6+ 4.Kg3 Qe8 5.Rf2 Kg7 6.Kh2 Qe1 7.Bg3 Qe3 8.Rc2 Kg6 9.Rg2 Kh6 10.Rb2 a5 11.Rg2 Kh7 12.Bc7 Qe4 13.Rf2 Qxh4 14.Bg3 (305.971.043) 72522  TB:42.853
 46/36	 0:04 	 0.00 	1.Rxd2 f3+ 2.Kxf3 Qc1 3.Bf4 Qc3+ 4.Kg4 Qc8+ 5.Kg5 Qc5+ 6.Kf6 Qc3+ 7.Kg6 Qc6+ 8.Kf5 Qc5+ 9.Ke4 Qe7+ 10.Kd5 Qxh4 11.Rd4 Qxh3 12.Kc5 Qh5+ 13.Kxb4 a5+ 14.Kc4 (326.945.790) 72590  TB:46.996
 47/49	 0:04 	 0.00 	1.Rxd2 f3+ 2.Kxf3 Qc1 3.Bf4 Qc6+ 4.Kg3 Qe8 5.Rf2 Kg7 6.Kh2 Qe1 7.Bg3 Qe3 8.Rc2 Kg6 9.Rg2 Kh6 10.Rb2 a5 11.Rg2 Kh7 12.Bc7 Qc3 13.Bd6 Qd4 14.Bb8 (357.982.556) 72760  TB:52.464
 48/49	 0:05 	 0.00 	1.Rxd2 f3+ 2.Kxf3 Qc1 3.Bf4 Qc6+ 4.Kg3 Qe8 5.Rf2 Kg7 6.Kh2 Qe1 7.Bg3 Qe3 8.Rc2 Kg6 9.Rg2 Kh6 10.Rb2 a5 11.Rg2 Kh7 12.Bc7 Qc3 13.Bd6 Qd4 14.Bb8 (393.037.350) 72771  TB:62.764
 49/50	 0:08 	 0.00 	1.Rxd2 Qc1 2.Bxf4 Qc5 3.Rf2 Kh7 4.Bg3 Kg7 5.Re2 Qh5 6.Rd2 Qe8 7.Kh2 Qe7 8.Bf4 Qxh4 9.Rg2+ Kh7 10.Bb8 Qd8 11.Be5 Qd3 12.Rf2 Qe3 13.Bg3 Kg6 14.Rc2 (594.605.567) 73380  TB:129.657
 50/51	 0:08 	 0.00 	1.Rxd2 Qc1 2.Bxf4 Qc5 3.Rf2 Kh7 4.Bg3 Kg7 5.Re2 Qh5 6.Rd2 Qe8 7.Kh2 Qe7 8.Rf2 Qe3 9.Rc2 a5 10.Rb2 Qe4 11.Rf2 Kh6 12.Bf4+ Kh5 13.Bc7 Qxh4 14.Re2 (595.398.090) 73379  TB:129.885
 51/51	 0:10 	 0.00 	1.Rxd2 Qc1 2.Bxf4 Qc5 3.Rf2 Kh7 4.Bg3 Kg7 5.Re2 Qh5 6.Rd2 Qe8 7.Kh2 Qe7 8.Rf2 Qe3 9.Rc2 a5 10.Rb2 Qe4 11.Rf2 Kh6 12.Bf4+ Kh5 13.Bc7 Qxh4 14.Re2 (755.232.318) 73566  TB:178.451
 52/53	 0:10 	 0.00 	1.Rxd2 Qc1 2.Bxf4 Qc5 3.Rf2 Kh7 4.Bg3 Kg7 5.Re2 Qh5 6.Rd2 Qe8 7.Kh2 Qe7 8.Rf2 Qe3 9.Rc2 a5 10.Rb2 Qe4 11.Rf2 Qe6 12.Bf4 Qe1 13.Rg2+ Kh7 14.Bb8 (756.248.331) 73565  TB:178.732
 53/56	 0:13 	+0.07++	1.Kf3 (999.991.778) 73572  TB:273.421
 53/56	 0:14 	+0.15++	1.Kf3 (1.042.977.177) 73537  TB:293.061
 53/56	 0:14 	+0.25++	1.Kf3 (1.044.990.863) 73539  TB:293.534
 53/56	 0:14 	+0.39++	1.Kf3 (1.045.796.002) 73544  TB:293.905
 53/56	 0:14 	+0.58++	1.Kf3 (1.054.931.418) 73565  TB:298.629
 53/56	 0:14 	+0.81++	1.Kf3 (1.068.818.562) 73584  TB:308.330
 53/56	 0:14 	+1.12++	1.Kf3 (1.078.825.060) 73614  TB:314.081
 53/56	 0:14 	+1.50++	1.Kf3 (1.089.150.156) 73621  TB:318.222
 53/56	 0:14 	+1.99++	1.Kf3 (1.097.500.351) 73633  TB:322.243
 53/56	 0:15 	+2.61++	1.Kf3 (1.124.634.644) 73741  TB:336.867
 53/56	 0:15 	+3.39++	1.Kf3 (1.133.921.149) 73770  TB:340.075
 53/56	 0:15 	+4.38++	1.Kf3 (1.149.177.380) 73816  TB:342.536
 53/56	 0:15 	+5.62++	1.Kf3 (1.181.841.231) 73943  TB:352.209
 53/60	 0:16 	+7.18++	1.Kf3 (1.252.576.688) 74152  TB:371.476
 53/67	 0:18 	+9.13++	1.Kf3 (1.381.695.729) 74488  TB:439.662
 53/70	 0:23 	+11.59++	1.Kf3 (1.741.936.142) 75372  TB:596.828
 53/70	 0:54 	+14.66++	1.Kf3 (4.460.208.342) 82363  TB:1.127.040
 53/70	 1:25 	+18.50++	1.Kf3 (7.141.634.015) 83394  TB:1.543.230
 53/70	 2:14 	+23.32++	1.Kf3 (10.937.569.138) 81519  TB:3.088.291
 53/70	 3:04 	+29.35++	1.Kf3 (14.785.564.504) 80218  TB:5.292.383
 53/70	 3:12 	+36.89++	1.Kf3 (15.439.375.610) 80122  TB:5.774.584
 53/70	 3:28 	+46.33++	1.Kf3 (16.728.039.448) 80228  TB:6.607.945
 53/70	 3:35 	+50.44	1.Kf3 Bc3 2.Kxf4 Kh7 3.Kf5 Kh6 4.Bf4+ Kh5 5.Bc1 Qxc1 6.Rxc1 Kxh4 7.a5 Kg3 8.Rd1 Kf2 9.h4 Kg3 10.h5 Kg2 11.Rd7 Kf3 12.Rxa7 Bh8 13.Ra6 Ke2 14.Re6+ (17.316.073.391) 80413  TB:7.408.291
 54/70	 5:03 	+50.44	1.Kf3 Bc3 2.Kxf4 Kh7 3.Kf5 Kh6 4.Bf4+ Kh5 5.Bc1 Qxc1 6.Rxc1 Kxh4 7.a5 Kg3 8.Rd1 Kf2 9.h4 Kg3 10.h5 Kg2 11.Rd7 Kf3 12.Rxa7 Bh8 13.Ra6 Ke2 14.Re6+ (24.950.437.793) 82279  TB:19.697.905
best move: Kg2-f3 time: 5:04.610 min  n/s: 82.279.235  nodes: 24.950.437.793 TB: 19.697.905 
This post may either be cause or result of misunderstandings.
amchess
Posts: 347
Joined: Tue Dec 05, 2017 2:42 pm

Re: Hard-Talkchess-2020 set, final release

Post by amchess »

I created a new Hard Positions 2022 to test engines.
Every position is unsolved by at least a top engine and they are classified based on Shashin theory.
Every suggestion/advise is welcome.
https://github.com/amchess/ShashChess/b ... ns2022.epd
https://github.com/amchess/ShashChess/b ... s2022.xlsx
Paloma
Posts: 1169
Joined: Thu Dec 25, 2008 9:07 pm
Full name: Herbert L

Re: Hard-Talkchess-2020 set, final release

Post by Paloma »

What was the Time you spend Stockfish on this .xlsx file ?
15 sec. pro Position or less ?