TCEC superfinal openings 9-17, edited

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

Moderators: hgm, Rebel, chrisw

User avatar
pohl4711
Posts: 2435
Joined: Sat Sep 03, 2011 7:25 am
Location: Berlin, Germany
Full name: Stefan Pohl

TCEC superfinal openings 9-17, edited

Post by pohl4711 »

I added the openings of the TCEC Superfinal season 17 to my edited TCEC-openings of J.Noomen (no double endpositions, all endpositons analyzed by Komodo, Season 9-17). This is, what I did:
1) deleted all lines 4 plies or shorter
2) deleted all lines with non-unique endpositions
3) checked all endpositions with Komodo 13.02: 10'/move, Quadcore, eval-interval [-0.50;+0.60]

293 out of 450 lines remaining. All endpositions are unique and not too bad for white or black.

293 lines: TCEC_9_17_edited.pgn (and EPD)
253 lines (enough for 500 games head-to-head of 2 engines) and no line is longer than 10 moves: TCEC_9_17_10mvs.pgn (and EPD)

Forthcoming Superfinal seasons openings will be added in the future.

Download on my website: https://www.sp-cc.de/files/jn_openings.zip

Have fun!
Paul Bedrey
Posts: 1146
Joined: Thu Mar 09, 2006 11:46 am
Location: Saratoga Springs New York

Re: TCEC superfinal openings 9-17, edited

Post by Paul Bedrey »

Hi Stefan, thanks for the testsets. What exactly is the too_bad_TCEC testsuite? It's 96 positions. Are these games in which at least one game was won?
User avatar
pohl4711
Posts: 2435
Joined: Sat Sep 03, 2011 7:25 am
Location: Berlin, Germany
Full name: Stefan Pohl

Re: TCEC superfinal openings 9-17, edited

Post by pohl4711 »

These are the lines in which the endposition-eval of Komodo is outside the eval-interval of [-0.50;+0.60].