Dylan Sharp Vs. Harvey Williamson (G4)

Discussion of computer chess matches and engine tournaments.

Moderators: Harvey Williamson, bob, hgm

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Raphexon
Posts: 327
Joined: Sun Mar 17, 2019 11:00 am
Full name: Henk Drost

Re: Dylan Sharp Vs. Harvey Williamson (G4)

Post by Raphexon » Wed Dec 04, 2019 10:34 am

Ovyron wrote:
Mon Dec 02, 2019 11:21 pm
Uri Blass wrote:
Mon Dec 02, 2019 2:03 pm
0.00 is only for positions that you are sure that they are drawn or you think both sides have equal probability to win.
I'm sure those positions are drawn even though I can't prove it. Also, Harvey's moves up to this point have been pretty much forced, because other alternatives were 0.00. Even if I can't prove it.

I think you and other people on the forum have this idea that you have to prove something to know it, but sometimes proof is something unreachable, and you just need to reach a level of certainty to know something.
Raphexon wrote:
Mon Dec 02, 2019 6:28 pm
You don't use 7 piece TBs, do you Ovy?
I don't even have 6 piece TBs :)
That's kinda impressive then.

User avatar
Ovyron
Posts: 4399
Joined: Tue Jul 03, 2007 2:30 am

Re: Dylan Sharp Vs. Harvey Williamson (G4)

Post by Ovyron » Wed Dec 04, 2019 12:56 pm

And I'll claim that in the position after the move I'm making, 25...Be6 is the only black's move that isn't 0.00, so I know Harvey will make it, though I can't prove it.

The idea is that if I'm able to predict all of Harvey's moves, then I can draw the game, but it takes a while...

1. g4 d5 2. g5 e5 3. d4 exd4 4. Nf3 c5 5. Bg2 Ne7 6. c3 dxc3 7. Nxc3 Nbc6 8. O-O d4 9. Ne4 Ng6 10. a3 Bg4 11. Bd2 Bd6 12. Rc1 O-O 13. Nxc5 Bxc5 14. Rxc5 Qd7 15. Qb3 Rad8 16. Re1 Be6 17. Qd3 Rfe8 18. e4 b6 19. Rcc1 Bg4 20. Rxc6 Qxc6 21. Nxd4 Qc5 22. Be3 Ne5 23. Qc3 Qxc3 24. bxc3 Rc8 25. f3


User avatar
Harvey Williamson
Posts: 1874
Joined: Sun May 25, 2008 9:12 pm
Location: Newport. Shropshire, UK.
Full name: Harvey Williamson
Contact:

Re: Dylan Sharp Vs. Harvey Williamson (G4)

Post by Harvey Williamson » Wed Dec 04, 2019 1:29 pm



Predictable!

jp
Posts: 1404
Joined: Mon Apr 23, 2018 5:54 am

Re: Dylan Sharp Vs. Harvey Williamson (G4)

Post by jp » Fri Dec 13, 2019 10:27 am

Move?!

User avatar
Ovyron
Posts: 4399
Joined: Tue Jul 03, 2007 2:30 am

Re: Dylan Sharp Vs. Harvey Williamson (G4)

Post by Ovyron » Fri Dec 13, 2019 12:21 pm

No way.

I'm going ahead and admit that if this was a "one week per move" (or faster) game Harvey would have probably beaten me by force (...unless I draw and I'd have played the same moves anyway).

I guess I can give something for the audience to look at, huh... what about positions analysed for my Candidate Moves?

1. Nb5 (2872 nodes analyzed)
2. Rc1 (605 nodes analyzed)
3. Bd2 (182 nodes analyzed)
4. Ne2 (1 node analyzed - deemed lost :shock: )

So I'll probably end up playing Nb5 (lest I lose +78% of my analysis), as Rc1 would need to overcome it and then stay ahead until I analyze as many nodes for it.

But I don't think anybody would bet the house for what move I'll play :)

User avatar
Ovyron
Posts: 4399
Joined: Tue Jul 03, 2007 2:30 am

Re: Dylan Sharp Vs. Harvey Williamson (G4)

Post by Ovyron » Fri Dec 20, 2019 11:30 pm

Update (to save the thread from page 2)

1. Nb5 (4187 nodes analyzed)
2. Rc1 (605 nodes analyzed)
3. Bd2 (182 nodes analyzed)
4. Ne2 (1 node analyzed - deemed lost :shock: )

Apparently I haven't analyzed anything else but Nb5, though I've rechecked positions to look for alternatives in the other moves.

jp
Posts: 1404
Joined: Mon Apr 23, 2018 5:54 am

Re: Dylan Sharp Vs. Harvey Williamson (G4)

Post by jp » Sat Dec 21, 2019 7:26 am

Ovyron wrote:
Fri Dec 20, 2019 11:30 pm
4. Ne2 (1 node analyzed - deemed lost :shock: )
Explain this "1 node analyzed". For that to be even vaguely true, I guess you have to think about that move for a split-second and immediately judge it lost. Otherwise it would be more than 1 node... Or maybe I should step back and ask what you mean by "n nodes analyzed".

Uri Blass
Posts: 8796
Joined: Wed Mar 08, 2006 11:37 pm
Location: Tel-Aviv Israel

Re: Dylan Sharp Vs. Harvey Williamson (G4)

Post by Uri Blass » Sat Dec 21, 2019 2:25 pm

jp wrote:
Sat Dec 21, 2019 7:26 am
Ovyron wrote:
Fri Dec 20, 2019 11:30 pm
4. Ne2 (1 node analyzed - deemed lost :shock: )
Explain this "1 node analyzed". For that to be even vaguely true, I guess you have to think about that move for a split-second and immediately judge it lost. Otherwise it would be more than 1 node... Or maybe I should step back and ask what you mean by "n nodes analyzed".
I guess he meant something like after
1.Ne2 xx when xx is black's move the score of stockfish after 999999999999 nodes is more than +4 pawns for black so he considers it as a win for black.

xx is the only node in his tree.
He does not count the 999999999999 nodes of stockfish.

I do not say the number are 4 pawns and 999999999999 nodes and his numbers of course may be different.

User avatar
Ovyron
Posts: 4399
Joined: Tue Jul 03, 2007 2:30 am

Re: Dylan Sharp Vs. Harvey Williamson (G4)

Post by Ovyron » Sun Dec 22, 2019 3:49 am

Whoa, long post lost.

But anyway, the point is that, yeah, Ne2 is below some threshold (that is deemed lost) and this is reached by only examining its node (the positions where Ne2 is played) so my engine might have analyzed more than a hundred million nodes of it but I only count the time I'm on the position and ask the engine for analysis, as Uri said.

jp
Posts: 1404
Joined: Mon Apr 23, 2018 5:54 am

Re: Dylan Sharp Vs. Harvey Williamson (G4)

Post by jp » Sun Dec 22, 2019 8:04 am

Okay, but if that number is the number of nodes you "personally dwell on" (to make up a description), does that mean you personally dwelt upon 4187 positions starting from Nb5 (with SF calculating "999999999999 nodes" at each)?

Post Reply