Balanced opening lines or advantageous opening lines?

[Graham Banks]

If you could back up your choice with reasons, that would be helpful too.

[Dann Corbit]

It depends on the goals.
Balanced openings are the most fair, and the most boring.
Advantageous lines are the least fair, and the most exciting.

I think TCEC has it right for a watched contest. Play polar openings, but force both engines to play as white and black to even it out.
For an unsupervised contest, I don't think it matters.

[Rebel]

It depends on the goals.
Balanced openings are the most fair, and the most boring.
Advantageous lines are the least fair, and the most exciting.

I think TCEC has it right for a watched contest. Play polar openings, but force both engines to play as white and black to even it out.
For an unsupervised contest, I don't think it matters.

It does matter, advantageous lines favor the stronger engines, the stronger engines usually have a better search.

[Graham Banks]

I think that TCEC is all about entertainment value now, which is fine for those who enjoy it.
It also demonstrates how to convert an advantageous opening into a win, which some might find useful.

However, the results have no substantial meaning if you're wanting to know the reality of engine comparisons.
Imagine giving advantageous opening lines to Carlsen and the other top players and stipulating that they must be played for entertainment value.

[Hai]

If you could back up your choice with reasons, that would be helpful too.

Play "The 20 Possible Moves" tournament and you will have both, because it's the best to have both.

[Graham Banks]

If you could back up your choice with reasons, that would be helpful too.

Play "The 20 Possible Moves" tournament and you will have both, because it's the best to have both.

I have a FirstMove book (2 ply) and a SecondMove book (4 ply) as well.
I used the former for a number of recent tournaments.

However, I approach my testing from a strong chess player's point of view.
I like to see opening lines that are reasonably balanced, so that I can see how engines either achieve equality, maintain equality or create a winning advantage.

Having said that, it is perfectly fine for people to test the way that they prefer, and for viewers to enjoy watching whatever they prefer.
It's not a contest.

[carldaman]

Definitely both, since that's the only way to come close to covering the whole gamut of chess openings.

The only openings I'd exclude are forced draws and total busts.

[Uri Blass]

There is a place for both but I prefer balanced because there is already too many non balanced games.
It is obvious after lc0 beat berserk with 1.g4 in TCEC that balanced opening lines or even no opening book is not always a draw even between top engines.

I test now stockfish15 at x nodes per move against berserk13 at 2x nodes per move(with contempt 0,10,20,30,40,50 when I started with x=1000 and double x after every match.

it is still not obvious if contempt of berserk change the playing strenght of it and it is still not clear if the berserk team is better or worse than stockfish15 in these conditions.

a win with black from the tournament.
White used 1024000 nodes per move.

[pgn][Event "Computer event"]
[Site "Somewhere on Earth"]
[Date "2024.04.03"]
[Time "06:36:28"]
[Round "2"]
[Board "2"]
[White "Stockfish 15"]
[Black "Berserk 13 contempt 0"]
[Result "0-1"]
[Termination "mate"]
[ECO "E10"]
[Opening "Queen's pawn game"]
[TimeControl "nodes: 10000;nodes: 30000000"]
[PlyCount "140"]

1.d4 {+0.4/20 1262 1024766} Nf6 {+0.2/22 1718 1944292 54/816/130} 2.c4 {+0.3/22 1242 1024532} e6 {+0.3/24 1914 1813632 41/809/150}
3.Nf3 {+0.4/22 1385 1024088} d5 {+0.3/23 1855 1500778 40/824/136; E10: Queen's pawn game} 4.g3 {+0.3/21 1354 1025398} Bb4+ {+0.2/24 1805 1936476 41/841/118}
5.Bd2 {+0.4/21 1348 1024970} Be7 {+0.1/23 2240 1484700 46/862/92} 6.Bg2 {+0.3/22 1564 1024490} O-O {+0.3/23 2008 1371053 27/842/131}
7.O-O {+0.4/21 1269 1024451} c6 {+0.1/23 1773 1794480 39/880/81} 8.Qb3 {+0.3/20 1515 1024935} Nbd7 {+0.1/20 2035 1378218 41/892/67}
9.Bf4 {+0.2/18 1360 1024613} a5 {+0.0/21 1944 2021584 49/903/48} 10.Rd1 {+0.1/19 1188 1024151} Ne4 {+0.2/21 1780 1855633 23/891/86}
11.Nc3 {+0.2/19 1270 1024548} g5 {+0.0/21 1927 1519237 36/917/47} 12.Bc1 {+0.1/18 1307 1024092} f5 {+0.2/22 1928 1978045 20/909/71}
13.Qc2 {+0.3/20 1713 1025315} Bf6 {+0.2/23 3138 1901290 17/912/71} 14.Nd2 {+0.0/18 1872 1024784} Nxd2 {+0.1/22 2127 2017499 23/933/44}
15.Rxd2 {-0.1/19 1441 1025028} Rf7 {+0.0/21 1927 1642051 30/941/29} 16.Rb1 {-0.1/19 1256 1024624} h5 {+0.0/20 1905 1493884 26/946/28}
17.Rd1 {+0.0/19 1288 1025454} Nf8 {+0.1/22 1955 1965344 14/943/43} 18.a3 {-0.2/19 1329 1025025} h4 {+0.0/23 1948 1648608 21/956/23}
19.b4 {-0.1/21 1312 1025463} hxg3 {+0.1/23 1933 1756406 16/959/25} 20.hxg3 {+0.1/22 1247 1025206} axb4 {+0.1/22 1982 1668614 12/960/28}
21.axb4 {+0.1/23 1826 1024775} Bd7 {+0.0/24 2310 1885272 16/967/17} 22.b5 {+0.1/23 1272 1024026} Be8 {+0.0/25 2121 1679372 15/969/16}
23.Qb3 {+0.1/23 1643 1024996} Rd7 {+0.0/27 2315 1685840 13/972/15} 24.e3 {+0.1/20 1479 1024700} Bg6 {+0.0/27 2168 1867617 12/975/13}
25.Rb2 {-0.1/18 1500 1024865} Rc8 {+0.0/27 2085 1694044 11/977/12} 26.Ra2 {-0.0/18 1434 1024357} Bh5 {+0.0/20 2165 778444 10/978/12}
27.Ne2 {-0.1/19 1341 1025096} Nh7 {+0.0/24 2094 1921966 9/981/10} 28.Qd3 {+0.1/20 1837 1025312} Bxe2 {-0.2/22 2256 1534789 27/970/3}
29.Qxe2 {-0.3/20 1481 1024379} g4 {-0.1/25 2207 1309290 14/981/5} 30.Ba3 {-0.2/22 1504 1025118} Ng5 {-0.2/24 2195 1728150 22/975/3}
31.Bc5 {-0.1/23 1659 1024272} Kf7 {-0.2/24 2347 1067813 22/976/2} 32.Ra7 {-0.5/19 1772 1024670} Kg6 {-0.3/24 1906 777060 35/964/1}
33.Rda1 {-0.6/18 1446 1024683} Rh7 {-0.5/24 2239 1457838 64/935/1} 34.bxc6 {-1.1/18 1522 1025702} Qh8 {-1.2/24 2357 1842772 756/244/0}
35.f3 {-1.1/19 1584 1025272} gxf3 {-1.3/21 2416 1084979 836/164/0} 36.Bxf3 {-1.6/18 1570 1024230} Rh2 {-1.1/22 2717 2048889 572/428/0}
37.Bg2 {-1.7/19 1968 1024676} bxc6 {-1.2/22 2348 1889499 732/268/0} 38.Bd6 {-1.6/18 1713 1024857} dxc4 {-1.1/22 2511 1498722 660/340/0}
39.g4 {-1.8/19 1767 1024905} Rh4 {-1.1/23 2636 2031593 694/306/0} 40.gxf5+ {-2.1/17 1762 1024069} exf5 {-1.4/23 2500 1931905 889/111/0}
41.R1a6 {-2.6/16 1621 1024514} Qd8 {-1.6/22 2320 1736542 949/51/0} 42.Bh2 {-2.2/17 1462 1024513} c3 {-2.0/20 2172 1345482 995/5/0}
43.Rb7 {-3.1/16 1569 1024420} Qe8 {-3.0/21 2343 1978570 1000/0/0} 44.Ra2 {-3.9/16 1538 1024844} Qh8 {-3.4/22 2330 1976771 1000/0/0}
45.Be5 {-4.5/16 1525 1024487} Bxe5 {-4.9/21 2297 1661824 1000/0/0} 46.dxe5 {-5.3/18 1840 1025708} Qxe5 {-5.4/19 2390 1189773 1000/0/0}
47.Qe1 {-5.9/16 1551 1024574} Rg4 {-6.4/20 2015 1220260 1000/0/0} 48.Kf1 {-7.3/16 1393 1025107} Ne4 {-7.0/21 1949 1153916 1000/0/0}
49.Rb1 {-8.3/16 1433 1024130} Rd8 {-7.3/24 2115 1254749 1000/0/0} 50.Rd1 {-9.5/16 1362 1024873} Rb8 {-7.7/25 2058 1407382 1000/0/0}
51.Rc1 {-10.9/16 1390 1025081} Qh2 {-8.2/25 1860 1599293 1000/0/0} 52.Qd1 {-12.1/17 1167 1026207} Ng3+ {-8.7/24 1478 1536568 1000/0/0}
53.Kf2 {-12.7/18 1027 1025539} Ne2 {-8.7/26 1459 1713159 1000/0/0} 54.Qh1 {-13.0/19 983 1024947} Rxg2+ {-8.9/25 1353 669787 1000/0/0}
55.Qxg2+ {-13.8/19 867 1024369} Qxg2+ {-9.5/23 1417 1471679 1000/0/0} 56.Kxg2 {-14.5/17 1026 1024482} Nxc1 {-9.6/23 1404 1851608 1000/0/0}
57.Ra6 {-15.9/17 1002 1024511} Rc8 {-10.0/22 1287 1613906 1000/0/0} 58.Ra4 {-16.5/17 969 1025610} Nd3 {-10.9/22 1435 1384580 1000/0/0}
59.Kf3 {-19.7/16 1257 1026144} c2 {-11.7/24 1267 1732405 1000/0/0} 60.Ra1 {M-20/27 1063 1025082} c1=Q {M+13/32 1614 2045665 1000/0/0}
61.Rxc1 {M-13/30 1126 1024785} Nxc1 {M+11/33 1482 1795938 1000/0/0} 62.e4 {M-11/29 1184 1025411} Re8 {M+9/32 1200 2035888 1000/0/0}
63.Ke3 {M-9/27 1246 1025140} Rxe4+ {M+8/33 1613 1851428 1000/0/0} 64.Kf3 {M-7/30 1192 1025027} Nd3 {M+7/34 1108 1795095 1000/0/0}
65.Kg3 {M-6/34 959 1024259} Re3+ {M+6/38 1018 1836834 1000/0/0} 66.Kg2 {M-5/38 664 1024126} Kg5 {M+5/46 788 1965720 1000/0/0}
67.Kf1 {M-4/85 398 1024282} Re1+ {M+4/134 791 2032534 1000/0/0} 68.Kg2 {M-3/245 44 225399} Kg4 {M+3/200 69 110694 1000/0/0}
69.Kh2 {M-2/245 32 7460} Kf3 {M+2/200 43 6159 1000/0/0} 70.Kh3 {M-1/245 31 492} Rh1# {M+1/200 43 5742 1000/0/0}
0-1

[/pgn]

[bastiball]

I think balanced openings because if a strong engine finds a way to win it out, then it is a game breaking

[Whiskers]

I like the idea behind balanced lines, and I think they're actually better for weak engines, unfortunately we are getting to the point where tons and tons of games are draws not just at the highest level, but at the levels below it, where engines such as Stormphrax, Willow, etc. are.

The most frustrating part about it is that many of these draws don't come because of good defensive effort, but because the position was so dead out of book that an engine had zero chance of making any progress against a similarly strong opponent. (It may not actually be completely dead, as evidenced by weaker engines' inability to hold, but it sure looks dead to the spectator watching the game!) In unbalanced positions, the disadvantaged side generally has to work to hold the draw, and there's tension to if it will be able to save the draw or not; a 0.95 engine score for 30 moves has audiences much more invested than a game that immediately flatlines to 0, whether or not the engine actually manages to draw.