Francesca MAD 0.28 64-bit Gauntlet for CCRL 40/15

Discussion of computer chess matches and engine tournaments.

Moderators: hgm, Dann Corbit, Harvey Williamson

User avatar
Graham Banks
Posts: 41202
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Francesca MAD 0.28 64-bit Gauntlet for CCRL 40/15

Post by Graham Banks »

gbanksnz at gmail.com
User avatar
Graham Banks
Posts: 41202
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: Francesca MAD 0.28 64-bit Gauntlet for CCRL 40/15

Post by Graham Banks »

Code: Select all

CCRL 40/15 Rating List - Custom engine selection
1139800 games played by 2648 programs, run by 23 testers
Ponder off, General books (up to 12 moves), 3-4-5 piece EGTB
Time control: Equivalent to 40 moves in 15 minutes on an Intel i7-4770k.
Computed on April 25, 2020 with Bayeselo based on 1'139'800 games
Tested by CCRL team, 2005-2020, http://ccrl.chessdom.com/ccrl/4040/

Rank                 Engine                   Elo   +    -   Score  AvOp  Games
1 Francesca MAD 0.28 64-bit               2900  +26  -26  52.7%  -19.9   460
  Francesca MAD 0.27 64-bit               2857  +28  -28  50.3%   -0.7   399
  Francesca MAD 0.26 64-bit               2836  +26  -26  52.8%  -18.8   464
  Francesca MAD 0.25 64-bit               2786  +23  -23  45.1%  +34.4   586
  Francesca MAD 0.24 64-bit               2759  +20  -20  52.8%  -17.4   870
  Francesca MAD 0.22 64-bit               2700  +19  -19  48.6%   +9.2   988
  Francesca MAD 0.21 64-bit               2671  +23  -23  55.6%  -38.7   652
  Francesca MAD 0.20 64-bit               2596  +28  -28  47.3%  +19.6   420
  Francesca MAD 0.19                      2558  +15  -15  48.6%   +9.7  1548
gbanksnz at gmail.com
User avatar
silentshark
Posts: 327
Joined: Sat Mar 27, 2010 7:15 pm

Re: Francesca MAD 0.28 64-bit Gauntlet for CCRL 40/15

Post by silentshark »

Graham Banks wrote: Mon Apr 27, 2020 2:53 am

Code: Select all

CCRL 40/15 Rating List - Custom engine selection
1139800 games played by 2648 programs, run by 23 testers
Ponder off, General books (up to 12 moves), 3-4-5 piece EGTB
Time control: Equivalent to 40 moves in 15 minutes on an Intel i7-4770k.
Computed on April 25, 2020 with Bayeselo based on 1'139'800 games
Tested by CCRL team, 2005-2020, http://ccrl.chessdom.com/ccrl/4040/

Rank                 Engine                   Elo   +    -   Score  AvOp  Games
1 Francesca MAD 0.28 64-bit               2900  +26  -26  52.7%  -19.9   460
  Francesca MAD 0.27 64-bit               2857  +28  -28  50.3%   -0.7   399
  Francesca MAD 0.26 64-bit               2836  +26  -26  52.8%  -18.8   464
  Francesca MAD 0.25 64-bit               2786  +23  -23  45.1%  +34.4   586
  Francesca MAD 0.24 64-bit               2759  +20  -20  52.8%  -17.4   870
  Francesca MAD 0.22 64-bit               2700  +19  -19  48.6%   +9.2   988
  Francesca MAD 0.21 64-bit               2671  +23  -23  55.6%  -38.7   652
  Francesca MAD 0.20 64-bit               2596  +28  -28  47.3%  +19.6   420
  Francesca MAD 0.19                      2558  +15  -15  48.6%   +9.7  1548
Well that's great news. Thanks so much for the continued testing.

When I stopped developing back in 2012 (with v0.19), I though that was about it, and I'd never be able to add much more to my old engine. That was wrong! The big change, for me, was getting a good PC, so I can do better testing. That, and understanding that small sets of results don't mean much, were the game changers for me.
User avatar
xr_a_y
Posts: 1871
Joined: Sat Nov 25, 2017 2:28 pm
Location: France

Re: Francesca MAD 0.28 64-bit Gauntlet for CCRL 40/15

Post by xr_a_y »

I'd love to see a linux version of Francesca MAD ! do you have plan for this ?
User avatar
silentshark
Posts: 327
Joined: Sat Mar 27, 2010 7:15 pm

Re: Francesca MAD 0.28 64-bit Gauntlet for CCRL 40/15

Post by silentshark »

xr_a_y wrote: Wed Apr 29, 2020 3:06 pm I'd love to see a linux version of Francesca MAD ! do you have plan for this ?
Hi Vivien,

Possibly. The only problem is, it would have to be binary.

I would love to share the source code, but it is too embarassing. Really nasty, hacky stuff :-)

All the best,
Tom