Hard talkchess 2016 set in PGN

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

Moderator: Ras

peter
Posts: 3393
Joined: Sat Feb 16, 2008 7:38 am
Full name: Peter Martan

Re: Hard talkchess 2016 set in PGN

Post by peter »

Hi Vincent!
Thanks for the collection!
I took the liberty to let latest SF run a Backward of your 139.pgn of the position in discussion and kept most of the evals to show the realtions between them.
Eelco's 11...h5 with some moves from Forward- Backward- Forward with SF became main variant but doesn't change the outcome probably, which is not fully clear to be won anyhow.
Yet I think 1.Nf5 a clear best move:

Peter.
Top
peter
Posts: 3393
Joined: Sat Feb 16, 2008 7:38 am
Full name: Peter Martan

Re: Hard talkchess 2016 set in PGN

Post by peter »

Too late to edit: forgot to set the commentaries too, at least 1.Nf5 should get ! and 1...h4 and 1...g5 ?! as for my pov
Peter.
Top
dadij
Posts: 40
Joined: Wed Jul 05, 2006 8:25 pm

Re: Hard talkchess 2016 set in PGN

Post by dadij »

Vinvin wrote: For the second time : check the line !
For a long and accurate line, IDeA has about 0% chance to find the correct continuation.
Carl with his IDeA analysis is actually correct, 1.Nf5 doesn't give white much. Your analysis is flawed. It didn't take me long to see this. I used IDeA too, admittedly taking advantage of a new feature in Aquarium 2016. Hint: Have a look at 10...Kf7 instead of 10...Re8.
Top
Vinvin
Posts: 5287
Joined: Thu Mar 09, 2006 9:40 am
Full name: Vincent Lejeune

Re: Hard talkchess 2016 set in PGN

Post by Vinvin »

dadij wrote:
Vinvin wrote: For the second time : check the line !
For a long and accurate line, IDeA has about 0% chance to find the correct continuation.
Carl with his IDeA analysis is actually correct, 1.Nf5 doesn't give white much. Your analysis is flawed. It didn't take me long to see this. I used IDeA too, admittedly taking advantage of a new feature in Aquarium 2016. Hint: Have a look at 10...Kf7 instead of 10...Re8.
Give your best lines and eval so we can judge ...
Top
yanquis1972
Posts: 1766
Joined: Wed Jun 03, 2009 12:14 am

Re: Hard talkchess 2016 set in PGN

Post by yanquis1972 »

either way were talking a difference of a few centipawns between the two moves...i think its an interesting position but a good test move should surely be starkly best?
Top
User avatar
Eelco de Groot
Posts: 4660
Joined: Sun Mar 12, 2006 2:40 am
Full name:   Eelco de Groot

Re: Hard talkchess 2016 set in PGN

Post by Eelco de Groot »

dadij wrote:
Vinvin wrote: For the second time : check the line !
For a long and accurate line, IDeA has about 0% chance to find the correct continuation.
Carl with his IDeA analysis is actually correct, 1.Nf5 doesn't give white much. Your analysis is flawed. It didn't take me long to see this. I used IDeA too, admittedly taking advantage of a new feature in Aquarium 2016. Hint: Have a look at 10...Kf7 instead of 10...Re8.
By coincidence, that is where I got stuck as well, I am leaning towards a draw but Serpent's eval is still 1.25 here after 16...Rc6. Probably 16. Rg5 Qe1 is the better defence instead of 16. Rg5 Rc6. And I don't have the Syzygy tablebases, let alone Lomonosov that you could probably use to good advantage so maybe someone with tablebases can give an easy answer?

New line using Dadi's Kf7 and 16...Qe1 :



8 
7 
6 
5 
4 
3 
2 
1 
abcdefgh

2r5/4k3/4n3/7R/8/P1P5/KP5P/8 b - -
Debugging is twice as hard as writing the code in the first
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan
Top
User avatar
Eelco de Groot
Posts: 4660
Joined: Sun Mar 12, 2006 2:40 am
Full name:   Eelco de Groot

Re: Hard talkchess 2016 set in PGN

Post by Eelco de Groot »

Hi Peter, you can just put the pgn between [ pgn] [/pgn] without leading spaces and then the custom written board software does the rest for you.

Regards, Eelco
Debugging is twice as hard as writing the code in the first
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan
Top
peter
Posts: 3393
Joined: Sat Feb 16, 2008 7:38 am
Full name: Peter Martan

Re: Hard talkchess 2016 set in PGN

Post by peter »

Eelco de Groot wrote:Hi Peter, you can just put the pgn between [ pgn] [/pgn] without leading spaces and then the custom written board software does the rest for you.
Thanks, Eelco, I thought, that was just what I did, except that you have a space between [ and pgn in your kind of typing [ pgn].
Was that deliberately done so and is this the point indeed?

BTW did you have a look at my .pgn with 11...h5 already?
Peter.
Top
User avatar
Eelco de Groot
Posts: 4660
Joined: Sun Mar 12, 2006 2:40 am
Full name:   Eelco de Groot

Re: Hard talkchess 2016 set in PGN

Post by Eelco de Groot »

peter wrote:
Eelco de Groot wrote:Hi Peter, you can just put the pgn between [ pgn] [/pgn] without leading spaces and then the custom written board software does the rest for you.
Thanks, Eelco, I thought, that was just what I did, except that you have a space between [ and pgn in your kind of typing [ pgn].
Was that deliberately done so and is this the point indeed?

BTW did you have a look at my .pgn with 11...h5 already?
Hi Peter, I think you used the backslash \ and the it does not work... :) I had to put in a space, otherwise what you see is just
and that is probably not too helpful.

Other trick to see what other people used is just temporarily open their message quoted, and you see the message tags they used.

I have not really looked to close at the pgn, but I think it is about the same eval that Vincent got? But if Dadi's variation is a draw...

I deleted some of the variations already after 11...h5, I had to switch some Serpent instances for looking at 10...Kf7 My Serpent's endresult I think was

41/86 666:01 +1.10 11...h5 12.Qe3 Qf7 13.Qc3 Rd8 14.Qe5 Rd4
15.Qb8+ Rd8 16.Qxb5 Rd1+ 17.Rxd1 Qxg6
18.Rd6 Qf7 19.a4 gxf6 20.Qf5 Nc7
21.Rd8+ Ne8 22.Qe4 Kf8 23.Rxe8+ Qxe8
24.Qxh7 Qe1+ (283.179.090.954) 7086

Vincent got a higher eval for White that is probably because I used an older Stockfish eval than the Development Stockfish. Or, it still is a draw
:P

Regards, Eelco
Debugging is twice as hard as writing the code in the first
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan
Top
dadij
Posts: 40
Joined: Wed Jul 05, 2006 8:25 pm

Re: Hard talkchess 2016 set in PGN

Post by dadij »

Eelco de Groot wrote: By coincidence, that is where I got stuck as well, I am leaning towards a draw but Serpent's eval is still 1.25 here after 16...Rc6. Probably 16. Rg5 Qe1 is the better defence instead of 16. Rg5 Rc6. And I don't have the Syzygy tablebases, let alone Lomonosov that you could probably use to good advantage so maybe someone with tablebases can give an easy answer?

New line using Dadi's Kf7 and 16...Qe1 :
The line you give is correct. My IDeA score after 1.Nf5 is +0.07. Your line can continue: 20...Rc4 21.Kb3 Rf4 and now either 22.a4 or 22.c4. The latest Stockfish should give something close to 0.00 here with 6-piece tablebases.
Top

Return to “Computer Chess Club: General Topics”