The Bryntse Gambit

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

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Dann Corbit
Posts: 10189
Joined: Wed Mar 08, 2006 7:57 pm
Location: Redmond, WA USA
Contact:

Re: The Bryntse Gambit

Post by Dann Corbit » Tue May 21, 2019 10:26 pm

bm is what is supposed to be the best move.
pm means the predicted move.
I do something really naughty and use pm in a way not defined by the standard.
This is what the EPD standard says:
"The "pm" opcode is used to provide a single predicted move for the indicated position. It has exactly one operand, a move playable from the position. This move is judged by the EPD writer to represent the best move available to the active player.

If a non-empty "pv" (predicted variation) line of play is also present in the same EPD record, the first move of the predicted variation is the same as the predicted move.

The "pm" opcode is intended for use as a general "display hint" mechanism."

However, that's downright silly. If the pm is just the first node of the pv, then it is utterly useless, since we already have it in the pv.
So I use the pm as follows:
If I do not have any statistics for the position, then I use the standard's definition of the pm, the first node of the pv.
But if I have data for the position, I provide all likely moves and the number of times that they occurred in actual games.
I also use the pm to hold all possible mating positions (if I have them).
For instance, Chest319 by Heiner Marxen will generate ALL mating moves of the minimum distance. So, if I have a position that is a direct mate in 6 moves and if I have 5 moves that all will lead to checkmate in 6, I put all 5 of them into the pm field.
Yes, pv is the principle variation. I just call it "the current plan" because that is what it is.

acd 42; bm exd5; c3 "Nf3"; cce -30; ce 3; pm exd5 {1567} Nc3 {495} e5 {229} d3 {152} Nf3 {49} Bb5+ {21}; white_wins 808; black_wins 971; draws 655; Opening ECO:B21i; Sicilian: Grand Prix, Tal Defence; 1.e4 c5 2.f4 d5 *;
The field acd of 42 means that a computer searched to a depth of 42 plies.
The field c3 of Nf3 means that a recent game I was analyzing chose that move, or it was proposed as a best move by someone.
The field cce is my invention, and it approximates the ce of the engine search but it is calculated as a function of the wins, losses and draws for the position.
The field ce of 3 means that the computer thought the side to move (white in this case) was 3/100ths of a pawn ahead (basically a dranw position).
The field pm breaks down this way
exd5 was the most common move, occurring 1567 times in my database of high quality filtered games.
Nc3 was the next most common move, occurring 495 in my database.
etc. down to Bb5+ which was played on 21 times.

cce is invented by me.
pm is abused by me.
But that is what this stuff means.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.

jp
Posts: 839
Joined: Mon Apr 23, 2018 5:54 am

Re: The Bryntse Gambit

Post by jp » Thu May 23, 2019 1:50 pm

Nordlandia wrote:
Thu May 16, 2019 7:50 am
There is a very similar line.
Yes, this looks like it might have some chance of appearing OTB.
SF, depth=27, evaluates this as (-1.99) with PV 9...Bg7 10. Nc3 h6, etc.

Nordlandia wrote:
Thu May 16, 2019 2:15 pm
Or by playing the scandinavian:
This may not, though. e.g. after 3.Ng5, SF prefers all of 3... Bf5, 3...e5, 3...Nc6 and 3...e6, with advantage for Black.

If the line is followed to 5... Kd7, then White can get an advantage without sacrificing by playing 6.f3 exf3 7.gxf3 Bf5 8.Bb3. SF, depth=40, evaluates this as (+0.95).

Dann Corbit
Posts: 10189
Joined: Wed Mar 08, 2006 7:57 pm
Location: Redmond, WA USA
Contact:

Re: The Bryntse Gambit

Post by Dann Corbit » Thu May 23, 2019 9:06 pm

jp wrote:
Thu May 23, 2019 1:50 pm
This may not, though. e.g. after 3.Ng5, SF prefers all of 3... Bf5, 3...e5, 3...Nc6 and 3...e6, with advantage for Black.

If the line is followed to 5... Kd7, then White can get an advantage without sacrificing by playing 6.f3 exf3 7.gxf3 Bf5 8.Bb3. SF, depth=40, evaluates this as (+0.95).
acd 36; bm e5; c3 "Nf6"; cce 187; ce 110; pm Bf5 {7} e5 {6} Nf6 {5} Qd5 {1}; pv e5; white_wins 4; black_wins 12; draws 3;

I have Nf6 as a plausible alternative that happens in real games.
Bf5 and e5 are safer responses in this position, but gambit players like risk.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.

Dann Corbit
Posts: 10189
Joined: Wed Mar 08, 2006 7:57 pm
Location: Redmond, WA USA
Contact:

Re: The Bryntse Gambit

Post by Dann Corbit » Thu May 23, 2019 9:18 pm

Here are all the correspondence chess games I have involving the Bryntse Gambit:

Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.

Dann Corbit
Posts: 10189
Joined: Wed Mar 08, 2006 7:57 pm
Location: Redmond, WA USA
Contact:

Re: The Bryntse Gambit

Post by Dann Corbit » Thu May 23, 2019 10:10 pm

I currently have 2146 positions for this opening in my main EPD database.
Might grow, with some more investigation.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.

jp
Posts: 839
Joined: Mon Apr 23, 2018 5:54 am

Re: The Bryntse Gambit

Post by jp » Fri May 24, 2019 12:50 am

Dann Corbit wrote:
Thu May 23, 2019 9:06 pm
rnbqkbnr/ppp1pppp/8/6N1/4p3/8/PPPP1PPP/RNBQKB1R b KQkq - acd 36; bm e5; c3 "Nf6"; cce 187; ce 110; pm Bf5 {7} e5 {6} Nf6 {5} Qd5 {1}; pv e5; white_wins 4; black_wins 12; draws 3;

I have Nf6 as a plausible alternative that happens in real games.
Bf5 and e5 are safer responses in this position, but gambit players like risk.
Gambit players like risk, but Black is not the gambit player!

White is the gambit player, but I don't know it makes sense to refuse to play 6.f3 when you get a pawn advantage, and Black's King on d7, for free.

jp
Posts: 839
Joined: Mon Apr 23, 2018 5:54 am

Re: The Bryntse Gambit

Post by jp » Fri May 24, 2019 1:48 am

Dann Corbit wrote:
Thu May 23, 2019 9:06 pm
jp wrote:
Thu May 23, 2019 1:50 pm
e.g. after 3.Ng5, SF prefers all of 3... Bf5, 3...e5, 3...Nc6 and 3...e6, with advantage for Black.
pm Bf5 {7} e5 {6} Nf6 {5} Qd5 {1};

I have Nf6 as a plausible alternative that happens in real games.
I guess Black players who don't know or care what SF thinks may play 3...Nf6, but how many Black players will play 4...Bg4 here?



SF prefers 4...e6, (-0.35), depth=31, but Black doesn't need SF to see that f6 is a move that makes Bg4 dubious.

Dann Corbit
Posts: 10189
Joined: Wed Mar 08, 2006 7:57 pm
Location: Redmond, WA USA
Contact:

Re: The Bryntse Gambit

Post by Dann Corbit » Fri May 24, 2019 3:27 am

If we only play optimally, we only get to 1.c4
So I guess the assumption is that we get to the gambit position, by hook or by crook.
It's all the way out here:
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.

jp
Posts: 839
Joined: Mon Apr 23, 2018 5:54 am

Re: The Bryntse Gambit

Post by jp » Fri May 24, 2019 3:43 am

I don't disagree that the assumption is we get there. It's just that some hooks and crooks look much more likely than others. The first two routes look more likely than the Scandinavian one, which requires some dubious moves by both sides (apart from whether the sacrifice itself is dubious).

jp
Posts: 839
Joined: Mon Apr 23, 2018 5:54 am

Re: The Bryntse Gambit

Post by jp » Fri May 24, 2019 8:00 am



Here's a game with 1... d5, 2...dxe4, 3...Nf6, 4...e6 that White drew with a GM:


Post Reply