TCEC 10

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

Moderators: hgm, Rebel, chrisw

Nay Lin Tun
Posts: 708
Joined: Mon Jan 16, 2012 6:34 am

Re: TCEC 10

Post by Nay Lin Tun »

I watched stockfish 8 vs stockfish 8 on my S7 edge handphone very often and noticed that Stockfish drew a lot of games with threefold repetition. May be a little bit randomness ( let say within 0.05 evaluation) might help to improve stockfish.
User avatar
Graham Banks
Posts: 41468
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: TCEC 10

Post by Graham Banks »

Leo wrote:Stockfish is drawing the weaker engines to much. It looks like there will be a new TCEC Champion this year.
Indeed - SF missed out this time.
Komodo v Houdini in the SuperFinal.
gbanksnz at gmail.com
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: TCEC 10

Post by MikeB »

Graham Banks wrote:
Leo wrote:Stockfish is drawing the weaker engines to much. It looks like there will be a new TCEC Champion this year.
Indeed - SF missed out this time.
Komodo v Houdini in the SuperFinal.
Yes indeed.

Code: Select all

Mac-Pro:cluster.mfb michaelbyrne$ bay
version 0058, Copyright (C) 1997-2016 Remi Coulom and updated by Michael Byrne.
compiled Jul 24 2016 00:03:35.
This program comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under the terms and conditions of the GNU General Public License.
See http://www.gnu.org/copyleft/gpl.html for details.

ResultSet>rp /Users/michaelbyrne/Downloads/dl.php.pgn 
109 game(s) loaded
ResultSet>elo
ResultSet-EloRating>mm 0 1
Iteration 100: 0.0130678 
Iteration 200: 0.000372643 
Iteration 300: 1.0666e-05 
00:00:00,00
ResultSet-EloRating>confidence 0.95
0.95
ResultSet-EloRating>r
Rank Name              Rating   Δ     +    -     #     Σ    Σ%     W    L    D   W%    =%   OppR 
---------------------------------------------------------------------------------------------------------
   1 Komodo 1959.00     3158   0.0   30   30    28   18.5  66.1   11    2   15  39.3  53.6  3092 
   2 Houdini 6.02       3157   0.4   30   30    28   18.5  66.1   10    1   17  35.7  60.7  3092 
   3 Stockfish 051117   3145  11.7   31   31    27   17.0  63.0    7    0   20  25.9  74.1  3093 
   4 Fire 6.2           3120  25.6   33   33    27   15.0  55.6    6    3   18  22.2  66.7  3095 
   5 Chiron 251017      3061  58.5   32   32    27   10.5  38.9    0    6   21   0.0  77.8  3107 
   6 Ginkgo 2.01        3061   0.5   32   32    27   10.5  38.9    1    7   19   3.7  70.4  3107 
   7 Andscacs 0.921     3049  11.8   31   31    27    9.5  35.2    0    8   19   0.0  70.4  3109 
   8 Booot 6.2          3049   0.1   31   31    27    9.5  35.2    0    8   19   0.0  70.4  3109 
---------------------------------------------------------------------------------------------------------
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: TCEC 10

Post by MikeB »

MikeB wrote:
Graham Banks wrote:
Leo wrote:Stockfish is drawing the weaker engines to much. It looks like there will be a new TCEC Champion this year.
Indeed - SF missed out this time.
Komodo v Houdini in the SuperFinal.
Yes indeed.

Code: Select all

Mac-Pro:cluster.mfb michaelbyrne$ bay
version 0058, Copyright (C) 1997-2016 Remi Coulom and updated by Michael Byrne.
compiled Jul 24 2016 00:03:35.
This program comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under the terms and conditions of the GNU General Public License.
See http://www.gnu.org/copyleft/gpl.html for details.

ResultSet>rp /Users/michaelbyrne/Downloads/dl.php.pgn 
109 game(s) loaded
ResultSet>elo
ResultSet-EloRating>mm 0 1
Iteration 100: 0.0130678 
Iteration 200: 0.000372643 
Iteration 300: 1.0666e-05 
00:00:00,00
ResultSet-EloRating>confidence 0.95
0.95
ResultSet-EloRating>r
Rank Name              Rating   Δ     +    -     #     Σ    Σ%     W    L    D   W%    =%   OppR 
---------------------------------------------------------------------------------------------------------
   1 Komodo 1959.00     3158   0.0   30   30    28   18.5  66.1   11    2   15  39.3  53.6  3092 
   2 Houdini 6.02       3157   0.4   30   30    28   18.5  66.1   10    1   17  35.7  60.7  3092 
   3 Stockfish 051117   3145  11.7   31   31    27   17.0  63.0    7    0   20  25.9  74.1  3093 
   4 Fire 6.2           3120  25.6   33   33    27   15.0  55.6    6    3   18  22.2  66.7  3095 
   5 Chiron 251017      3061  58.5   32   32    27   10.5  38.9    0    6   21   0.0  77.8  3107 
   6 Ginkgo 2.01        3061   0.5   32   32    27   10.5  38.9    1    7   19   3.7  70.4  3107 
   7 Andscacs 0.921     3049  11.8   31   31    27    9.5  35.2    0    8   19   0.0  70.4  3109 
   8 Booot 6.2          3049   0.1   31   31    27    9.5  35.2    0    8   19   0.0  70.4  3109 
---------------------------------------------------------------------------------------------------------
Also, a very nice showing by Fire 6.2 in Stage 2, the only other engine to finish above 50% in Stage 2 besides the Big 3.
JJJ
Posts: 1346
Joined: Sat Apr 19, 2014 1:47 pm

Re: TCEC 10

Post by JJJ »

So I was right to hope for a Houdini / Komodo final this time. Now I would bet than Houdini is the slighty favorite here and says he will win with 52%.
I think contempt helped Komodo a lot but might be a little less strong in head to head, even if Houdini is not update. But I might be wrong if the next Komodo has won some more elo.
Dan Cooper
Posts: 184
Joined: Sun Nov 01, 2015 3:15 am

Re: TCEC 10

Post by Dan Cooper »

JJJ wrote:So I was right to hope for a Houdini / Komodo final this time. Now I would bet than Houdini is the slighty favorite here and says he will win with 52%.
I think contempt helped Komodo a lot but might be a little less strong in head to head, even if Houdini is not update. But I might be wrong if the next Komodo has won some more elo.
They announced a Houdini update in chat.
User avatar
cc2150dx
Posts: 325
Joined: Sat Nov 30, 2013 9:51 am
Location: Canada
Full name: Jason Coombs

Re: TCEC 10

Post by cc2150dx »

Dan Cooper wrote:
JJJ wrote:So I was right to hope for a Houdini / Komodo final this time. Now I would bet than Houdini is the slighty favorite here and says he will win with 52%.
I think contempt helped Komodo a lot but might be a little less strong in head to head, even if Houdini is not update. But I might be wrong if the next Komodo has won some more elo.
They announced a Houdini update in chat.
If so, I hope it will be a free upgrade for us H6 customers :)
Jouni
Posts: 3293
Joined: Wed Mar 08, 2006 8:15 pm

Re: TCEC 10

Post by Jouni »

Stockfish is first engine, which goes WHOLE season UNBEATEN!! No word on that in chessdom page? In final Houdini is clear favorite. Komodo's win in stage 2 was position, which happens once in 1000 games.
Jouni
shrapnel
Posts: 1339
Joined: Fri Nov 02, 2012 9:43 am
Location: New Delhi, India

Re: TCEC 10

Post by shrapnel »

JJJ wrote:Now I would bet than Houdini is the slighty favorite here and says he will win with 52%.
+1. More than a slight favorite. These Long Games really suit Houdini and it is inherently the more aggressive Engine, updated or not.
Komodo, no doubt good, seems to get lost sometimes.
So, I confidently predict a Houdini win, but let's see.....
i7 5960X @ 4.1 Ghz, 64 GB G.Skill RipJaws RAM, Twin Asus ROG Strix OC 11 GB Geforce 2080 Tis
User avatar
Houdini
Posts: 1471
Joined: Tue Mar 16, 2010 12:00 am

Re: TCEC 10

Post by Houdini »

cc2150dx wrote:
Dan Cooper wrote: They announced a Houdini update in chat.
If so, I hope it will be a free upgrade for us H6 customers :)
It is, and Houdini 6.03 is already available to you - you can use the same download link we sent you by e-mail, it now refers to the updated version.

Houdini 6.03 corrects a stalemate bug which was reported to me this weekend. It occurs when there are white pawn captures available, for example this position:
[d]8/3R1pk1/5p1p/5PbK/B2p2PP/3P4/1r6/8 w - - 0 1

Try this position with the previous Houdini 6, set multi-PV to 4. You will get some surprising results :).

The error was already present in Houdini 5, but apparently nobody has ever noticed it.

Robert