Page 1 of 2

Openings for Season 7

PostPosted: Thu May 22, 2014 9:57 am
by Psyck
Cato agreed in principle to consider openings suggested by the TCEC community for inclusion into the Season 7 games (for Stage 1 to Stage 4). While the modalities for the same has not been officially formalized, I thought it would be a good idea for the users to consolidate the list of opening suggestions here.

Re: Openings for Season 7

PostPosted: Thu May 22, 2014 3:12 pm
by nemo engine fan
How many moves? Again 8 moves or is this also part of the suggestions?

Re: Openings for Season 7

PostPosted: Sat May 24, 2014 3:09 pm
by dennis1382
I would love to see the games just start with some mainline tabyas...

Like 1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6 5.0-0 Be7 6.Re1 b5 7.Bb3 d6 8.c3 0-0 9.h3
or 1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6
or 1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.Nf3 0-0 6.Be2 e5 7.0-0
etcetera

Re: Openings for Season 7

PostPosted: Sun May 25, 2014 4:48 pm
by anonymous coward
I appreciate principles, strategy, and statistical analysis done to prepare the Season 6 opening book http://open-chess.org/viewtopic.php?f=41&t=2639. It sounds like a lot of work was done and the goals were clearly thought out and well reasoned. The only problem I see is that the analysis was done with engines released a couple months prior to the tournament. Of course, the engines have progressed since then - and these newer versions are what actually participated in Season 6 - thereby raising into question whether some (if not most?) of the analysis has become obsolete ... especially by the time K & S played the Superfinal ... arguably, the most important time to have proper analysis on the opening book.

To improve this for Season 7, perhaps a shorter opening book would be in order? In other words, keep the principles that were used to choose the Season 6 openings, but shorten the book to just the first 3 moves (6 plies). In this fashion, the appearance of humans and/or older engines muddling the opening book is lessened, and the engines in the tournament would choose which moves to make from an earlier point in the game (rather than what some previously released version of an engine may have analysed months ago). This should also greatly reduce the amount of work needed to generate the Season 7 opening book ... so a win for the tournament directors as well as the fans!

Re: Openings for Season 7

PostPosted: Tue May 27, 2014 4:37 pm
by nemo engine fan
For engines 1. e4 e5 2. Nf3 Nc6 3. Bc4 Nf6 could be interesting; in my preperation I saw engines suggesting some crazy lines like after 4. Ng5 d5 5. exd5 Na5 6. Bb5+ c6 7. dxc6 bxc6 8. Bd3 the interesting move Ng4 (after which engines could play it out). It followed a second pawn sacrafice and huge attack by black but engines showed it is equal (something like 0.2).

Re: Openings for Season 7

PostPosted: Wed May 28, 2014 9:50 pm
by Psyck
Sorry, user suggested community openings will not be used in Season 7.

Quoting from TCEC chat:

Cato the Younger (VIP)
...there have been a number of lessons learned from this Season and I am planning to make adjustments for next Season...biggest change is in the Superfinal, where I think I will be getting professional assistance for the final selections...but for the other Stages I think I pretty well using my own methods and, assuming we have the same or a very similar structure next Season, all I would need to do really is tweak Stage 2 so there were fewer draws in that single round-robin format.

Re: Openings for Season 7

PostPosted: Thu May 29, 2014 1:15 am
by Martin Thoresen
Please, nothing is decided yet. I will monitor these forums for input on what people want.

Re: Openings for Season 7

PostPosted: Thu Jul 03, 2014 1:22 pm
by fauzi
anonymous coward wrote:I appreciate principles, strategy, and statistical analysis done to prepare the Season 6 opening book http://open-chess.org/viewtopic.php?f=41&t=2639. It sounds like a lot of work was done and the goals were clearly thought out and well reasoned. The only problem I see is that the analysis was done with engines released a couple months prior to the tournament. Of course, the engines have progressed since then - and these newer versions are what actually participated in Season 6 - thereby raising into question whether some (if not most?) of the analysis has become obsolete ... especially by the time K & S played the Superfinal ... arguably, the most important time to have proper analysis on the opening book.

To improve this for Season 7, perhaps a shorter opening book would be in order? In other words, keep the principles that were used to choose the Season 6 openings, but shorten the book to just the first 3 moves (6 plies). In this fashion, the appearance of humans and/or older engines muddling the opening book is lessened, and the engines in the tournament would choose which moves to make from an earlier point in the game (rather than what some previously released version of an engine may have analysed months ago). This should also greatly reduce the amount of work needed to generate the Season 7 opening book ... so a win for the tournament directors as well as the fans!



I totally agree with this quote, and specially for the decreament of the numbers of ply used in the openings, I think its much better for many reasons, to have a 3 moves (6 ply) openings, or maximum a 4 moves (8 ply) for season 7

Re: Openings for Season 7

PostPosted: Thu Jul 03, 2014 1:24 pm
by fauzi
I agree that the number of moves used in the opening book should be lesser than what we seen in season 6 to give more free to engines and alot of other reasons

Re: Openings for Season 7

PostPosted: Thu Jul 31, 2014 2:06 pm
by nemo engine fan
fauzi wrote:I agree that the number of moves used in the opening book should be lesser than what we seen in season 6 to give more free to engines and alot of other reasons

I am not sure about that. With less moves we probably will repeat openings anywhen and also some of the long lines were interesting.
I don't know if it's possible but what about different opening book move number? In some openings a deep line, in some only a short line. Since both engines have the same time on the clock and we don't have something like additional time after 40 moves it wouldn't be unfair.