Centaur vs Unassisted Engine

Discussion of computer chess matches and engine tournaments.

Moderators: hgm, Dann Corbit, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
jefk
Posts: 503
Joined: Sun Jul 25, 2010 8:07 pm
Location: the Netherlands
Contact:

Re: Centaur vs Unassisted Engine

Post by jefk » Sun Feb 07, 2021 12:07 pm

Milos,

some time ago i've downloaded this cerebelum3merged
(C3M) Just had a look, it plays in current position ,...Be7
but after ...Bb4+ there folllows indeed Nd2 (although i
still have work to do in some side analysis in Aquarium).
As for book use by my opponent, he doesn't as was agreed and
that's the point of this testgame; someone claimed that with SF
Nnue you don't need a book, and a few hrs of analysis will result
in a drawn game even against an experienced corresp. player.
Although Aquarium when using with different engines becomes
a rather cumbersome exercise (with many, many sidelines), at the
moment it's the best i have (whereby i could make a list with
improvements for the Aquarium software but that's another subject).

For the rest, well we are not going to change to conditions,
and you are not the referee. If you're not interested, than don't
follow it.

As for the result, wait and see. Yes, one or two hours is not much
indeed for mr MC who cordially and gracefully accepted the challange
as some sort of operator (but it's (comparable if you play two or three
corresp tourns simultaneously when having only one comp) ;
otherwise i don't think i would play this challenge. As for the book (C3M), i
don't think it includes Mcts analysis, but for the rest it's probably
indeed almost perfect, nevertheless after some 10 moves or so we will get
out of the book, and will i will anyway not mindlessly follow
the advise of the book. At the end indeed, like you wrote, i will try to
get an advantage for White, anyway in the endgame.

For me the motivation also is to see what kind of endgame
we get (regarding ideas for possible rule change), i don't
think stalemate will play a role, also i don't think i will get
one knight more vs a lone king, most likely is some endgame
with pieces with a pawn more (or a bishop pair advantage
if i play 5.Bd2 (instead of 5.Nd2). As for SF, it used to
underestimate the advantage of the bishop pair, but
with Nnue i'm not so sure of that.
jefk

MonteCarlo
Posts: 146
Joined: Sun Dec 25, 2016 3:59 pm

Re: Centaur vs Unassisted Engine

Post by MonteCarlo » Mon Feb 08, 2021 2:50 pm

My apologies for the slow response. There was a massive change at work over the weekend that sucked up my time, so I hadn't even checked the forums.

4...Be7


jefk
Posts: 503
Joined: Sun Jul 25, 2010 8:07 pm
Location: the Netherlands
Contact:

Re: Centaur vs Unassisted Engine

Post by jefk » Mon Feb 08, 2021 3:14 pm

no problem,
an interesting move this Be7,
hadn't expected it (...Bb4+ looked stronger to me,
and then i still would be hesitating between Nd2!? or Bd2.

We now get into a Catalan (*), it will be probably be 5.Nf3
but i'll take my time in a few hours, or otherwise (maybe)
tomorrow i'll confirm.

(*) from theory i knew that there are some tricky lines
which will lead to a draw for Black, i'll have to look
(after tomorrow or so) at my older analysis again
to avoid such lines.
jefk

jefk
Posts: 503
Joined: Sun Jul 25, 2010 8:07 pm
Location: the Netherlands
Contact:

Re: Centaur vs Unassisted Engine

Post by jefk » Mon Feb 08, 2021 6:29 pm

it's 5.Nf3 ,
indeed.

Now if it goes like i expect, i'll have to make a rather
important decision on move ten, namely Bf4 or Bg5,
both lines have been played quite often, with a lot
of draws, yet i still think White can get some advantage
jefk

MonteCarlo
Posts: 146
Joined: Sun Dec 25, 2016 3:59 pm

Re: Centaur vs Unassisted Engine

Post by MonteCarlo » Mon Feb 08, 2021 6:55 pm

5...0-0


Alayan
Posts: 527
Joined: Tue Nov 19, 2019 7:48 pm
Full name: Alayan Feh

Re: Centaur vs Unassisted Engine

Post by Alayan » Mon Feb 08, 2021 9:53 pm

Interesting game.

A few months ago, when people debated whether unassisted Stockfish could lose as white from the start position (some people were really thick-headed in that thread), I offered to have a centaur vs engine match to people who thought the engine would be crushed, but with the engine having white (but less computation time than here). I thought that the centaur would have good chances but that it wouldn't be easy. However, the same people saying we are over a thousand of elo away from perfect play in the start position declined the offer.

I'm curious how this game will go. I will root for the assisted human to get a win ; it's more exciting if engines are still making errors in these conditions.

mbabigian
Posts: 189
Joined: Tue Oct 15, 2013 12:34 am
Location: US
Full name: Mike Babigian

Re: Centaur vs Unassisted Engine

Post by mbabigian » Tue Feb 09, 2021 1:47 am

I too was stunned by 4. ... Be7, but MonteCarlo's PC is very slow and there is a lot we don't know about his setup. Nor should he describe it here, as it would be an enormous advantage if described in detail. SF is not a positional player. Before NNUE it was actually laughable. Its great strength came from its massive search and pruning techniques. In the early stages of the game, its "judgement" is mostly noise without massive search.

That said, I'll provide some "reasonably" objective assessments of the position so far using Leela which has very good positional judgement, but embarrassingly bad search.

After a "directed" search of 78.3M nodes we have the statistics before any pieces are moved on line 1 below:

Code: Select all

                     Whites Edge  Win       Draw    Loss
Starting Position    53.6%        35.6%     36.1%   28.3%
After Blacks move 5  53.1%        31.2%     43.7%   25.1%
The second line gives the stats after 5. ... O-O.

The fact that the draw estimation has increased is not a direct reflection of play, as much as the issue of chess likely being a drawn game. As the game progresses without a mistake, the likelihood of a draw naturally increases. Although I also expected Bb4+, I see no major problem with Be7. Chess is a game of mistakes. Someone will need to make one to escape the draw singularity's gravitational pull. Over time I can post some additional assessments, but once there have been too many exchanges, Leela looses its luster. I have other resources to look at including my massive Aquarium trees, correspondence and other databases, etc. If I think the balance of the position has shifted significantly, I'll post something.

Mike
P.S. MonteCarlo, I would be curious about your setup, but it should only be provided via PM. Things like software version, hash size, and the stockfish bench speed using the same hash and threads you are using for the game. You'll need to run the bench command numerous times and take an average nps, as multithreaded searches can have some significant variability. BuildTester can automate that process if you are interested in passing such information in a private message.
“Censorship is telling a man he can't have a steak just because a baby can't chew it.” ― Mark Twain

Uri Blass
Posts: 8942
Joined: Wed Mar 08, 2006 11:37 pm
Location: Tel-Aviv Israel

Re: Centaur vs Unassisted Engine

Post by Uri Blass » Tue Feb 09, 2021 7:16 am

setup can be also multi-pv

with multi-pv I find that Be7 is the best move at depth 49(order when I copy from arena is from worst to best) but later Bb4+ becomes the best move

Bf8-e7 Ng1-f3 c7-c6 get a different score than c7-c6 Ng1-f3 Bf8-e7 at depth 51.

I know that stockfish is not deterministic with more than one core so analyzing the position twice may give different results.



FEN: rnbqkb1r/ppp2ppp/4pn2/3p4/2PP4/6P1/PP2PPBP/RNBQK1NR b KQkq - 1 4

Stockfish_21013116_x64_avx2:

48/62 1:04:31 32,168,606k 8,308k -0.17 Bf8-e7 Ng1-f3 c7-c6 O-O O-O Qd1-c2 b7-b6 Nb1-d2 Bc8-b7 e2-e4 c6-c5 e4xd5 e6xd5 d4xc5 Be7xc5 Nd2-b3 Bc5-e7 c4xd5 Nb8-a6 Bc1-g5 h7-h6 Bg5xf6 Be7xf6 Ra1-d1 Na6-b4 Qc2-e4 Nb4xd5 Nf3-e5 Rf8-e8 f2-f4 Qd8-c7 Rd1xd5 Bb7xd5 Qe4xd5 Ra8-d8 Qd5-b7 Qc7xb7 Bg2xb7 Bf6xe5 f4xe5 Re8xe5 Bb7-f3 Re5-e7 Rf1-c1 g7-g6 Kg1-g2 Rd8-d3 Rc1-c2 Rd3-d7 Kg2-f1 Kg8-f8 Bf3-e2 Rd7-c7 Nb3-d4
48/64 1:04:31 32,168,606k 8,308k -0.17 c7-c6 Ng1-f3 Bf8-e7 O-O O-O Qd1-c2 b7-b6 Rf1-d1 Nb8-d7 b2-b3 Bc8-a6 Nb1-d2 c6-c5 e2-e4 d5xc4 e4-e5 Nf6-d5 b3xc4 Nd5-b4 Qc2-c3 Ba6-b7 d4xc5 Be7xc5 Nd2-b3 Qd8-c7 Nb3xc5 b6xc5 Bc1-f4 Nd7-b6 Rd1-d6 Ra8-d8 Ra1-d1 Nb6-c8 a2-a3 Nb4-c6 Rd6xd8 Rf8xd8 Rd1xd8+ Nc6xd8 Nf3-g5 Bb7xg2 Kg1xg2 Nc8-e7 Ng5-e4 Ne7-g6 Qc3-d3 Nd8-c6 Ne4xc5 Nc6xe5 Bf4xe5 Ng6xe5
48/65 1:04:31 32,168,606k 8,308k -0.12 Bf8-b4+ Bc1-d2 Bb4-e7 Ng1-f3 O-O Qd1-c2 c7-c6 O-O b7-b6 Nf3-e5 Bc8-b7 Rf1-d1 Nb8-d7 c4xd5 c6xd5 Ne5-c6 Bb7xc6 Qc2xc6 a7-a6 Rd1-c1 Ra8-c8 Qc6xc8 Qd8xc8 Rc1xc8 Rf8xc8 Nb1-c3 b6-b5 Ra1-c1 Nd7-b6 Nc3-d1 Rc8xc1 Bd2xc1 Nb6-c4 e2-e3 Nc4-a5 Bc1-d2 Na5-c6 Nd1-c3 Nf6-d7 b2-b3 Kg8-f8 Kg1-f1 Nd7-f6 h2-h3 Be7-b4 g3-g4 Kf8-e8 Kf1-e2 g7-g5 Nc3-b1 Nf6-e4
48/63 1:04:31 32,168,606k 8,308k -0.09 d5xc4 Ng1-f3 Bf8-b4+ Bc1-d2 a7-a5 Qd1-c2 c7-c5 a2-a3 Bb4xd2+ Nb1xd2 c5xd4 Nd2xc4 Nb8-c6 Qc2-d3 O-O O-O Rf8-e8 Nf3-e5 Nc6xe5 Nc4xe5 Qd8-d6 Ne5-c4 Qd6-c5 b2-b4 Qc5-e7 Nc4-b6 Ra8-b8 Qd3xd4 e6-e5 Qd4-c5 a5xb4 Qc5xb4 Qe7xb4 a3xb4 Bc8-e6 Rf1-d1 g7-g6 Nb6-a4 Be6-b3 Rd1-d6 Bb3xa4 Rd6xf6 Kg8-g7 Rf6-b6 Ba4-c6 b4-b5 Bc6xg2 Kg1xg2 e5-e4 Rb6-d6 Re8-d8 Ra1-d1 Rd8xd6 Rd1xd6
---------------------------------------------------------------------------
49/65 1:09:52 34,837,582k 8,309k -0.20 c7-c6 Ng1-f3 Bf8-e7 O-O O-O Qd1-c2 b7-b6 Nf3-e5 Bc8-b7 Rf1-d1 Nb8-d7 Bc1-f4 Nf6-h5 Bf4-d2 Nh5-f6 c4xd5 c6xd5 Ne5-c6 Bb7xc6 Qc2xc6 a7-a6 Rd1-c1 Ra8-c8 Qc6xc8 Qd8xc8 Rc1xc8 Rf8xc8 Nb1-c3 b6-b5 Ra1-c1 Nd7-b6 Nc3-d1 Rc8xc1 Bd2xc1 Nb6-c4 f2-f3 Nf6-d7 e2-e3 Nd7-b8 Bg2-f1 Nb8-c6 b2-b3 Nc4-a5 Bf1-d3 Kg8-f8 Nd1-f2 g7-g6 Bd3-f1 Be7-b4 Nf2-d3 Bb4-c3 Bc1-b2 Bc3xb2 Nd3xb2
49/59 1:09:52 34,837,582k 8,309k -0.20 d5xc4 Ng1-f3 Bf8-b4+ Bc1-d2 a7-a5 Qd1-c2 c7-c5 a2-a3 Bb4xd2+ Nb1xd2 c5xd4 Nd2xc4 Nb8-c6 Qc2-d3 O-O O-O Rf8-e8 Nf3-e5 Nc6xe5 Nc4xe5 Qd8-d6 Ne5-c4 Qd6-c5 b2-b4 Qc5-e7 b4xa5 e6-e5 Qd3-b1 Bc8-e6 Nc4xe5 Ra8xa5 Ne5-f3 Ra5-d5 Rf1-d1 Re8-d8 Nf3-e1 Be6-f5 Qb1-b4 Rd5-d7 Qb4xe7 Rd7xe7 Kg1-f1 g7-g6 a3-a4 Nf6-e4 Bg2xe4 Re7xe4 f2-f3 Re4-e8 a4-a5 Kg8-g7 Kf1-f2 Rd8-c8
49/62 1:09:52 34,837,582k 8,309k -0.19 Bf8-b4+ Bc1-d2 Bb4-e7 Ng1-f3 c7-c6 Qd1-c2 Nb8-d7 O-O O-O Rf1-d1 b7-b6 Bd2-f4 Bc8-b7 Nf3-e5 Nf6-h5 Bf4-d2 Nh5-f6 c4xd5 c6xd5 Ne5-c6 Bb7xc6 Qc2xc6 a7-a6 Rd1-c1 Ra8-c8 Qc6xc8 Qd8xc8 Rc1xc8 Rf8xc8 Nb1-c3 b6-b5 Ra1-c1 Nd7-b6 Nc3-d1 Rc8xc1 Bd2xc1 Nb6-c4 f2-f3 Nf6-d7 e2-e3 Nd7-b8 Bg2-f1 Nb8-c6 b2-b3 Nc4-a5 Bf1-d3 g7-g6 Bd3-b1 Be7-b4 Nd1-f2 Bb4-c3 Nf2-d3 Nc6-b4 Bc1-b2 Bc3xb2 Nd3xb2 Na5-c6 a2-a3
49/62 1:09:52 34,837,582k 8,309k -0.16 Bf8-e7 Ng1-f3 c7-c6 O-O O-O b2-b3 b7-b6 Bc1-b2 Nb8-d7 Qd1-c2 Bc8-b7 Rf1-d1 Qd8-c7 Nb1-d2 Ra8-c8 Ra1-c1 Qc7-b8 e2-e4 d5xe4 Nd2xe4 c6-c5 Ne4xf6+ Nd7xf6 Qc2-e2 c5xd4 Rd1xd4 Qb8-a8 Nf3-h4 Rf8-d8 Rc1-d1 Rd8xd4 Rd1xd4 Bb7xg2 Nh4xg2 Rc8-d8 Rd4xd8+ Qa8xd8 Ng2-e1 h7-h6 h2-h4 Nf6-d7 Bb2-d4 Nd7-f6 Ne1-f3 Be7-d6 Nf3-e5 Bd6-c5 Bd4-c3 Qd8-a8 Qe2-f3 Qa8xf3 Ne5xf3
---------------------------------------------------------------------------
50/66 1:24:05 41,952,014k 8,314k -0.22 c7-c6 Ng1-f3 Bf8-e7 b2-b3 O-O O-O b7-b6 Bc1-b2 Nb8-d7 Qd1-c2 Bc8-b7 Rf1-d1 Ra8-c8 Nb1-c3 Qd8-c7 Ra1-c1 Qc7-b8 e2-e4 d5xe4 Nc3xe4 c6-c5 Ne4xf6+ Nd7xf6 Qc2-e2 c5xd4 Nf3-e5 Rc8-d8 Rd1xd4 Rd8xd4 Bb2xd4 Rf8-d8 Rc1-d1 Qb8-a8 Bg2xb7 Qa8xb7 Rd1-d3 Nf6-e4 Qe2-g4 Ne4-f6 Qg4-f4 h7-h6 h2-h4 h6-h5 Bd4-e3 Rd8xd3 Ne5xd3 Nf6-d7 Nd3-e5 f7-f6 Ne5xd7 Qb7xd7
50/60 1:24:05 41,952,014k 8,314k -0.22 Bf8-e7 Ng1-f3 c7-c6 O-O O-O b2-b3 b7-b6 Bc1-b2 Nb8-d7 Qd1-c2 Bc8-b7 Rf1-d1 Qd8-c7 Nb1-d2 Ra8-c8 Ra1-c1 Qc7-b8 e2-e4 d5xe4 Nd2xe4 c6-c5 Ne4xf6+ Nd7xf6 Qc2-e2 c5xd4 Nf3-e5 Rc8-d8 Rd1xd4 Rd8xd4 Bb2xd4 Rf8-d8 Rc1-d1 Qb8-a8 Bg2xb7 Qa8xb7 h2-h4 Nf6-e4 Bd4-e3 Rd8xd1+ Qe2xd1 Ne4-f6 Qd1-d3 h7-h6 Qd3-d4 Nf6-e4 Qd4-d7 Qb7xd7 Ne5xd7 Ne4-c3 a2-a4 f7-f6 c4-c5 b6xc5 Nd7xc5 Be7xc5 Be3xc5 Kg8-f7 Kg1-h2
50/63 1:24:05 41,952,014k 8,314k -0.12 Bf8-b4+ Bc1-d2 Bb4-e7 Ng1-f3 Nb8-d7 O-O c7-c6 Qd1-c2 O-O e2-e3 a7-a5 Rf1-c1 Rf8-e8 Bd2-e1 Be7-f8 b2-b3 h7-h6 Nb1-c3 b7-b6 e3-e4 Nf6xe4 Nc3xe4 d5xe4 Qc2xe4 Bc8-b7 Be1-c3 Qd8-c7 Nf3-e5 f7-f5 Qe4-f3 Nd7xe5 d4xe5 Bf8-c5 Rc1-d1 Re8-d8 Rd1xd8+ Ra8xd8 Ra1-d1 Bc5-e7 Rd1-d3 Rd8xd3 Qf3xd3 c6-c5 Bg2xb7 Qc7xb7 f2-f3 Qb7-c6 a2-a4 Be7-f8 h2-h4 Kg8-f7
50/63 1:24:05 41,952,014k 8,314k -0.07 d5xc4 Ng1-f3 Bf8-b4+ Bc1-d2 a7-a5 Qd1-c2 c7-c5 a2-a3 Bb4xd2+ Nb1xd2 c5xd4 Nd2xc4 Nb8-c6 Qc2-d3 O-O O-O Rf8-e8 Nf3-e5 Nc6xe5 Nc4xe5 Qd8-d6 Ne5-c4 Qd6-c5 b2-b4 Qc5-e7 Nc4-b6 Ra8-b8 Qd3xd4 e6-e5 Qd4-c5 a5xb4 a3xb4 e5-e4 Rf1-d1 Bc8-g4 f2-f3 e4xf3 e2xf3 Bg4-e6 Rd1-d4 g7-g6 Qc5xe7 Re8xe7 Ra1-c1 Re7-e8 Bg2-f1 Rb8-d8 Rc1-d1 Rd8xd4 Rd1xd4 Re8-e7 Kg1-f2 Re7-c7 Bf1-c4 Be6xc4 Nb6xc4
---------------------------------------------------------------------------
51/76 2:24:41 56,498,408k 6,508k -0.28 Bf8-e7 Ng1-f3 c7-c6 O-O O-O Qd1-c2 b7-b6 Rf1-d1 Nb8-d7 Nb1-d2 Bc8-a6 e2-e4 Nf6xe4 Nd2xe4 d5xe4 Nf3-e5 Nd7xe5 d4xe5 Qd8-c7 Bg2xe4 h7-h6 Bc1-e3 Qc7xe5 Rd1-d7 Be7-c5 Ra1-e1 Bc5xe3 Re1xe3 Qe5-c5 b2-b3 b6-b5 Be4-h7+ Kg8-h8 c4xb5 Qc5xb5 Bh7-d3 Qb5-b6 Bd3xa6 Qb6xa6 Re3-c3 Ra8-c8 Rc3-c4 Rc8-d8 Rc4xc6 Qa6-a5 Rc6-c7 Rd8xd7 Rc7xd7 Kh8-g8 h2-h4 a7-a6 Rd7-a7 Qa5-e1+ Kg1-g2 a6-a5 Ra7-d7 Qe1-e5 Qc2-d3 Qe5-e1 a2-a3 Qe1-e5 Qd3-f3
51/67 2:24:41 56,498,408k 6,508k -0.15 d5xc4 Ng1-f3 Bf8-b4+ Bc1-d2 a7-a5 Qd1-c2 c7-c5 a2-a3 Bb4xd2+ Nb1xd2 c5xd4 Nd2xc4 Nb8-c6 Qc2-d3 O-O O-O b7-b6 Nf3xd4 Qd8xd4 Bg2xc6 Qd4xd3 e2xd3 Ra8-b8 Rf1-c1 Bc8-d7 Nc4-e5 Rb8-d8 Kg1-f1 Bd7xc6 Rc1xc6 Rd8-d5 Ne5-c4 Nf6-d7 Kf1-e2 Rf8-b8 Ra1-c1 h7-h5 a3-a4 Rd5-c5 Rc6xc5 Nd7xc5 Nc4xa5 Nc5xd3 Ke2xd3 b6xa5 Kd3-c2 h5-h4 b2-b3 g7-g6 Rc1-d1 Rb8-b4 Kc2-c3
51/67 2:24:41 56,498,408k 6,508k -0.14 c7-c6 Ng1-f3 Bf8-e7 b2-b3 b7-b6 O-O O-O Bc1-b2 Nb8-d7 Qd1-c2 Bc8-b7 Rf1-d1 Qd8-c7 Nb1-d2 Ra8-c8 Ra1-c1 Qc7-b8 e2-e4 d5xe4 Nd2xe4 c6-c5 Ne4xf6+ Nd7xf6 Qc2-e2 c5xd4 Rd1xd4 Qb8-a8 Rc1-d1 h7-h6 Nf3-e1 Bb7xg2 Ne1xg2 Rc8-d8 Ng2-e1 Rd8xd4 Rd1xd4 Qa8-c8 Rd4-d1 Rf8-d8 Rd1xd8+ Qc8xd8 Ne1-d3 Qd8-a8 h2-h4 Qa8-b7 a2-a4 Nf6-d7 Nd3-e5 Qb7-c7 Ne5xd7
51/67 2:24:41 56,498,408k 6,508k -0.14 Bf8-b4+ Bc1-d2 Bb4-e7 Ng1-f3 c7-c6 Qd1-c2 O-O O-O Nb8-d7 Rf1-c1 a7-a5 e2-e3 Rf8-e8 Bd2-e1 h7-h6 a2-a3 Be7-f8 b2-b3 b7-b6 c4xd5 c6xd5 Bg2-f1 Bc8-b7 Nb1-c3 Qd8-e7 Nc3-b5 Re8-c8 Qc2-b2 Nf6-e4 Rc1xc8 Ra8xc8 Ra1-c1 Qe7-e8 Rc1-c2 Rc8xc2 Qb2xc2 Qe8-b8 h2-h3 Qb8-c8 Qc2xc8 Bb7xc8 a3-a4 Bc8-a6 Nb5-a7 Ba6-b7 Kg1-g2 Ne4-d6 Na7-b5 Nd6xb5

jefk
Posts: 503
Joined: Sun Jul 25, 2010 8:07 pm
Location: the Netherlands
Contact:

Re: Centaur vs Unassisted Engine

Post by jefk » Tue Feb 09, 2021 12:46 pm

MC you wrote :
change at work that sucked up my time, hadn't even checked the forums.
like i wrote, no problem, and you can talkchess notify you via email
when there's a response (although such responses also sometimes
come from others, like now :)

NB first thought i would simply play 6.0-0, but now it's not so clear.
May take a few days to decide if 6.Nc3!? might be slightly better
(played less often, but may give chances in a game like this one)
jefk

jefk
Posts: 503
Joined: Sun Jul 25, 2010 8:07 pm
Location: the Netherlands
Contact:

Re: Centaur vs Unassisted Engine

Post by jefk » Tue Feb 09, 2021 6:05 pm

ok it's 6.0-0

Not so suprising, although the cerebellum merge book
gives 6.Nc3 (probably because it couldn't find significant
advantage lateron in the ''main' Catalan lines in which
we (first) seem be entering.
Although SF can quickly go wrong, in this Chinese database
http://chessdb.cn/queryc_en/
there's eg. 6...c6 besides 6...dxc4 (whereby only the latter
makes a chance of a draw imho. But then in that chessdb.cn
most analysis is still with SF without Nnue, i suspect.
jefk

Post Reply