"Peasant Revolt" Endgame Chess

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

Moderators: Harvey Williamson, bob, hgm

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
User avatar
Laskos
Posts: 10833
Joined: Wed Jul 26, 2006 8:21 pm
Full name: Kai Laskos

"Peasant Revolt" Endgame Chess

Post by Laskos » Sat Sep 02, 2017 4:50 am

Inspired by this thread:
http://www.talkchess.com/forum/viewtopic.php?t=65051
and keeping the name of the variant given by Jon Fredrik Åsvang, I modified the most probably decided in favor of Black opening position to this one:

[D]1nn1k1n1/4p3/8/8/8/PPPPPPPP/8/4K3 w - - 0 1

If a player stronger than me shows how to play correctly this variant and the theoretical result of it, it would be helpful. As I am more of a patzer in deciding on the decisiveness of the Black advantage here, I left Komodo 11.2.2 play self-games. I built a a 3-mover opening book containing 1000+ positions using the same Komodo, as it randomizes better, and has the "Variety" parameter. The EPD book of 1000+ positions is here:
http://s000.tinyupload.com/?file_id=009 ... 2378736001

The unbalance here is accentuated, and there are two nice properties to this Endgame Chess variant: the position is borderline Black win versus Draw (about 50% to 50%), and is very sensitive to 6-men Syzygy (from SSD). These borderline positions are desirable in order to separate in strength engines, a slight slip can lead from draw to loss and vice-versa. Using pentanomial variance here, the errors one sees in Cutechess (trinomial errors) are really 1.7-1.8 times smaller. It is important to not have any kind of adjudication, engines misevaluate the positions during the game, and might claim adjudication, even when the game should go on.

On two threads, from that 3-mover book, Komodo 11.2.2 with Syzygy 6-men default Probe Depth=2 from SSD, the results were in 100 self-games:

10''+ 0.1''
50 Black wins, 50 draws

60''+ 0.6''
51 Black wins, 49 draws

360''+ 3.6''
44 Black wins, 56 draws

Really hard to decide if there is a trend towards one outcome or the other.

Komodo 11.2.2 (Contempt=0) beats easily Stockfish dev. in 100 games at 60''+ 0.6''. Syzygy 6-men from SSD for both:

60''+ 0.6''
Score of Stockfish dev vs Komodo 11.2.2: 30 - 65 - 105 [0.412] 200
ELO difference: -61.43 +/- 33.17
Finished match
95 Black wins, 105 draws


Also, the game is extremely sensitive to Syzygy, even better than the most sensitive to Syzygy endgame suites I had:

60''+ 0.6''
Score of Komodo Syzygy vs Komodo NO TB: 71 - 28 - 101 [0.608] 200
ELO difference: 75.88 +/- 33.89
Finished match
99 Black wins, 101 draws

The real errors here using the pentanomial are about 18-19 ELO points instead of the shown by Cutechess-Cli trinomial 33-34 ELO points.

All in all, this borderline variant helps in seeing the engines dealing with accentuated endgame disbalances and benefiting from Syzygy, 6-men from SSD in my case. It has a very high sensitivity to these two issues

MikeGL
Posts: 962
Joined: Thu Sep 01, 2011 12:49 pm

Re: "Peasant Revolt" Endgame Chess

Post by MikeGL » Sat Sep 02, 2017 5:48 am

Good modification. This minor tweak is more interesting. Can't decide if black is
clearly winning this time.

But it seems there's no white wins on your self game stats.
10''+ 0.1''
50 Black wins, 50 draws

60''+ 0.6''
51 Black wins, 49 draws

360''+ 3.6''
44 Black wins, 56 draws

Really hard to decide if there is a trend towards one outcome or the other.

Komodo 11.2.2 (Contempt=0) beats easily Stockfish dev. in 100 games at 60''+ 0.6''. Syzygy 6-men from SSD for both:

60''+ 0.6''
Score of Stockfish dev vs Komodo 11.2.2: 30 - 65 - 105 [0.412] 200
ELO difference: -61.43 +/- 33.17
Finished match
95 Black wins, 105 draws


Also, the game is extremely sensitive to Syzygy, even better than the most sensitive to Syzygy endgame suites I had:

60''+ 0.6''
Score of Komodo Syzygy vs Komodo NO TB: 71 - 28 - 101 [0.608] 200
ELO difference: 75.88 +/- 33.89
Finished match
99 Black wins, 101 draws

Maybe this means it is just =+ position, where white has only drawing chances.


.

User avatar
Laskos
Posts: 10833
Joined: Wed Jul 26, 2006 8:21 pm
Full name: Kai Laskos

Re: "Peasant Revolt" Endgame Chess

Post by Laskos » Sat Sep 02, 2017 5:59 am

MikeGL wrote:Good modification. This minor tweak is more interesting. Can't decide if black is
clearly winning this time.

But it seems there's no white wins on your self game stats.
10''+ 0.1''
50 Black wins, 50 draws

60''+ 0.6''
51 Black wins, 49 draws

360''+ 3.6''
44 Black wins, 56 draws

Really hard to decide if there is a trend towards one outcome or the other.

Komodo 11.2.2 (Contempt=0) beats easily Stockfish dev. in 100 games at 60''+ 0.6''. Syzygy 6-men from SSD for both:

60''+ 0.6''
Score of Stockfish dev vs Komodo 11.2.2: 30 - 65 - 105 [0.412] 200
ELO difference: -61.43 +/- 33.17
Finished match
95 Black wins, 105 draws


Also, the game is extremely sensitive to Syzygy, even better than the most sensitive to Syzygy endgame suites I had:

60''+ 0.6''
Score of Komodo Syzygy vs Komodo NO TB: 71 - 28 - 101 [0.608] 200
ELO difference: 75.88 +/- 33.89
Finished match
99 Black wins, 101 draws

Maybe this means it is just =+ position, where white has only drawing chances.


.
Yes, in all these matches White had 0 wins, the position is clearly favoring Black, maybe not decisively though. But that's the idea, to have 50% Black wins, and 50% draws, and White performance of 25%. That 50% seems to be stable across time controls, hard to see any clear trend. The sensitivity to imbalance and Syzygy seems to be the highest in this 50%/50%/0% case.

User avatar
Laskos
Posts: 10833
Joined: Wed Jul 26, 2006 8:21 pm
Full name: Kai Laskos

Re: "Peasant Revolt" Endgame Chess

Post by Laskos » Sat Sep 02, 2017 6:54 am

Here is a small rating list for "Peasant Revolt" of top engines, 1440 games in 40x Round-Robin at 400ms per move:

Code: Select all

     Program                        Elo    +    -   Games   Score   Av.Op.  Draws

   1 Komodo 11.2.2 64-bit         : 3136   25   25   320    66.6 %   2983   43.1 %
   2 Stockfish 260817 64 BMI2     : 3073   23   23   320    59.1 %   2991   47.5 %
   3 Deep Shredder 13 x64         : 3044   24   24   320    55.5 %   2995   50.3 %
   4 Houdini 5.01 Pro x64-popc    : 3043   24   24   320    55.3 %   2995   53.8 %
   5 Gull 3 x64                   : 3020   23   23   320    52.5 %   2997   39.4 %
   6 Fire 5 x64                   : 2966   24   24   320    45.8 %   3004   42.8 %
   7 Fritz 15                     : 2935   23   23   320    41.9 %   3008   48.1 %
   8 Andscacs 0.91b               : 2918   23   23   320    39.8 %   3010   56.6 %
   9 Booot 6.2_x64                : 2866   24   24   320    33.6 %   3017   50.3 %

Games        : 1440 (finished)

White Wins   : 75 (5.2 %)
Black Wins   : 674 (46.8 %)
Draws        : 691 (48.0 %)
Unfinished   : 0

White Score  : 29.2 %
Black Score  : 70.8 %
Syzygy enabled engines had 6-men Syzygy on SSD.

And 10-20 minute analyses of Komodo 11.2.2 and Stockfish dev on 4 cores:
[D]1nn1k1n1/4p3/8/8/8/PPPPPPPP/8/4K3 w - -

Komodo

Code: Select all

Engine: Komodo 11.2.2 64-bit (2048 MB)
by Don Dailey, Larry Kaufman, Mark Lefler


29.01  2:38   -0.58--  1.c4 Nc6 (898.497.144) 5671 
                       TB:319.387 

29.01  3:52   -0.63    1.c4 Nf6 2.g4 Nc6 3.Kd2 e6 4.f4 e5 
                       5.Kc3 exf4 6.exf4 Nb6 7.b4 Nfd7 
                       8.Kd2 Na4 9.Ke3 Kf7 10.h4 Ne7 11.h5 Nf6 
                       12.Kf3 Nc6 13.h6 Nd4+ 14.Ke3 (1.306.093.370) 5619 
                       TB:514.426 

30.01  4:40   -0.57++  1.c4 Nf6 2.Ke2 (1.572.298.268) 5601 
                       TB:670.977 

30.01  5:04   -0.60    1.c4 Nf6 2.Ke2 Nc6 3.g4 e6 4.Kd2 Ke7 
                       5.f4 e5 6.Kc3 exf4 7.exf4 Nb6 8.f5 Kf7 
                       9.a4 Kg7 10.a5 Nxa5 11.Kb4 Nb7 
                       12.Kb5 Nc8 13.Kc6 Na5+ 14.Kc7 (1.702.855.602) 5590 
                       TB:747.603 

31.01  5:55   -0.54++  1.c4 Nf6 2.Ke2 (1.980.836.829) 5575 
                       TB:929.792 

31.01  6:38   -0.61--  1.c4 Nc6 (2.227.303.936) 5596 
                       TB:1.122.899 

31.01  8:22   -0.62    1.c4 Nf6 2.Ke2 Nc6 3.g4 e6 4.Kd2 Ke7 
                       5.h4 Kd7 6.b4 Ne5 7.Ke2 Ne7 8.a4 N7g6 
                       9.g5 Nh5 10.f4 Nf7 11.c5 e5 12.fxe5 Nxh4 
                       13.d4 Nxg5 14.b5 (2.836.689.267) 5647 
                       TB:1.604.093 

32.01  9:32   -0.56++  1.c4 Nf6 2.Kd2 (3.234.704.654) 5654 
                       TB:1.932.058 
Stockfish

Code: Select all

Engine: Stockfish 260817 64 BMI2 (2048 MB)
by T. Romstad, M. Costalba, J. Kiiski, G.


36/54  11:31  -1.00    1.c4 Nf6 2.g4 Nc6 3.h4 e6 4.Ke2 Kf7 
                       5.Kd2 Ke7 6.Ke2 e5 7.b4 Ke6 8.h5 Nd6 
                       9.Kd2 e4 10.fxe4 Ne5 11.c5 Nde8 
                       12.g5 Nxh5 13.b5 Ng3 14.a4 (5.245.535.166) 7581 
                       TB:4.127.524 

37/40  11:38  -1.07--  1.c4 Nf6 (5.301.485.761) 7585 
                       TB:4.172.900 

37/40  11:46  -1.15--  1.c4 Nf6 (5.358.712.368) 7587 
                       TB:4.213.284 

37/52  12:09  -1.07++  1.c4 (5.543.430.083) 7596 
                       TB:4.390.894 

37/58  14:51  -0.92++  1.c4 (6.818.986.281) 7644 
                       TB:5.799.077 

37/58  16:13  -1.26--  1.c4 Nf6 (7.461.387.273) 7665 
                       TB:6.554.305 

37/61  16:21  -0.98++  1.c4 (7.527.161.527) 7666 
                       TB:6.608.850 

37/61  16:28  -1.07    1.c4 Nf6 2.g4 Nc6 3.Kd2 Kd7 4.b4 Ne5 
                       5.Ke2 Nb6 6.h4 Ng6 7.h5 Ne5 8.e4 Na4 
                       9.c5 Nf7 10.d4 Nb2 11.b5 e5 12.dxe5 Nxe5 
                       13.c6+ Kc7 14.Ke3 (7.577.040.735) 7668 
                       TB:6.727.261 

38/58  16:36  -1.15--  1.c4 Nf6 (7.645.071.233) 7669 
                       TB:6.778.267 

38/58  17:10  -1.22--  1.c4 Nf6 (7.909.307.571) 7678 
                       TB:7.143.521 

38/58  18:33  -1.15++  1.c4 (8.567.219.756) 7695 
                       TB:8.082.549 

Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 10:41 am

Re: "Peasant Revolt" Endgame Chess

Post by Lyudmil Tsvetkov » Sat Sep 02, 2017 12:17 pm

Laskos wrote:Inspired by this thread:
http://www.talkchess.com/forum/viewtopic.php?t=65051
and keeping the name of the variant given by Jon Fredrik Åsvang, I modified the most probably decided in favor of Black opening position to this one:

[D]1nn1k1n1/4p3/8/8/8/PPPPPPPP/8/4K3 w - - 0 1

If a player stronger than me shows how to play correctly this variant and the theoretical result of it, it would be helpful. As I am more of a patzer in deciding on the decisiveness of the Black advantage here, I left Komodo 11.2.2 play self-games. I built a a 3-mover opening book containing 1000+ positions using the same Komodo, as it randomizes better, and has the "Variety" parameter. The EPD book of 1000+ positions is here:
http://s000.tinyupload.com/?file_id=009 ... 2378736001

The unbalance here is accentuated, and there are two nice properties to this Endgame Chess variant: the position is borderline Black win versus Draw (about 50% to 50%), and is very sensitive to 6-men Syzygy (from SSD). These borderline positions are desirable in order to separate in strength engines, a slight slip can lead from draw to loss and vice-versa. Using pentanomial variance here, the errors one sees in Cutechess (trinomial errors) are really 1.7-1.8 times smaller. It is important to not have any kind of adjudication, engines misevaluate the positions during the game, and might claim adjudication, even when the game should go on.

On two threads, from that 3-mover book, Komodo 11.2.2 with Syzygy 6-men default Probe Depth=2 from SSD, the results were in 100 self-games:

10''+ 0.1''
50 Black wins, 50 draws

60''+ 0.6''
51 Black wins, 49 draws

360''+ 3.6''
44 Black wins, 56 draws

Really hard to decide if there is a trend towards one outcome or the other.

Komodo 11.2.2 (Contempt=0) beats easily Stockfish dev. in 100 games at 60''+ 0.6''. Syzygy 6-men from SSD for both:

60''+ 0.6''
Score of Stockfish dev vs Komodo 11.2.2: 30 - 65 - 105 [0.412] 200
ELO difference: -61.43 +/- 33.17
Finished match
95 Black wins, 105 draws


Also, the game is extremely sensitive to Syzygy, even better than the most sensitive to Syzygy endgame suites I had:

60''+ 0.6''
Score of Komodo Syzygy vs Komodo NO TB: 71 - 28 - 101 [0.608] 200
ELO difference: 75.88 +/- 33.89
Finished match
99 Black wins, 101 draws

The real errors here using the pentanomial are about 18-19 ELO points instead of the shown by Cutechess-Cli trinomial 33-34 ELO points.

All in all, this borderline variant helps in seeing the engines dealing with accentuated endgame disbalances and benefiting from Syzygy, 6-men from SSD in my case. It has a very high sensitivity to these two issues
this is extremely specific, as it is important not to sac one knight for a couple of pawns and then maybe win, maybe not, depending on the rank and file the pawn for the weaker side is placed on.

if you don't have tbs, then you simply don't know when a sac works and when not, and most positions would simply boil down to this.

I am not certain if SF and Komodo have 2Ns vs P endgame knowledge, probably not, as too detailed, and even if they do have only knowledge that 2Ns vs bare king is draw.

but even if they have the second one, that does not help in any way, as there is a surplus pawn.

User avatar
Laskos
Posts: 10833
Joined: Wed Jul 26, 2006 8:21 pm
Full name: Kai Laskos

Re: "Peasant Revolt" Endgame Chess

Post by Laskos » Sun Sep 03, 2017 5:16 am

Lyudmil Tsvetkov wrote:
Laskos wrote:

All in all, this borderline variant helps in seeing the engines dealing with accentuated endgame disbalances and benefiting from Syzygy, 6-men from SSD in my case. It has a very high sensitivity to these two issues
this is extremely specific, as it is important not to sac one knight for a couple of pawns and then maybe win, maybe not, depending on the rank and file the pawn for the weaker side is placed on.

if you don't have tbs, then you simply don't know when a sac works and when not, and most positions would simply boil down to this.

I am not certain if SF and Komodo have 2Ns vs P endgame knowledge, probably not, as too detailed, and even if they do have only knowledge that 2Ns vs bare king is draw.

but even if they have the second one, that does not help in any way, as there is a surplus pawn.
I think your insight is good. This is regarding the sensitivity to Syzygy (6-men from SSD in my case) versus no TB. The difference seems huge, at least in Komodo and Stockfish cases. Indeed, this is very peculiar to position, and engines make that sac decision pretty late in the game, so openings here don't matter much, therefore built opening book doesn't distort the result. There is a third property of the positions (and the built opening book). At 2160''+21.6'' (6 times the previous 360''+ 3.6''), the result between two identical Komodo 11.2.2 with 6-men Syzygy from SSD in 100 games was 48 Black wins, 52 draws:

100 games matches:

1x time control:
10''+ 0.1''
50 Black wins, 50 draws, 0 White wins

6x time control:
60''+ 0.6''
51 Black wins, 49 draws, 0 White wins

6x6 = 36x time control:
360''+ 3.6''
44 Black wins, 56 draws, 0 White wins

6x6x6 = 216x time control:
2160''+ 21.6''
48 Black wins, 52 draws, 0 White wins


No any scaling with time control. This is again very peculiar, usually outcomes White/Draw/Black scale with time control control, whether balanced or unbalanced openings.

The peculiarities of this opening (and the built book) are the following:
  • 1/ This borderline variant helps in seeing the engines dealing with accentuated endgame disbalances.
    2/ Benefits heavily from Syzygy, 6-men from SSD in my case.
    3/ Always stays borderline with close in strength engines, close to 50% Black wins, 50% draws, independent of time control. This is the sweet point of sensitivity with unbalanced openings using pentanomial variance.
Also, to note that the games are pretty long (I observed this when checking for total time used when playing with increment). The histogram of game lengths for this borderline opening:

[D]1nn1k1n1/4p3/8/8/8/PPPPPPPP/8/4K3 w - - 0 1

Image

Median game: 55.5 moves
Commonest game: 49 moves

Observe very many 100-150 moves games. Compare that to another variant, although not very related, of "Theban Chess". It also has only Knights and pawns, but many more pawns. "Theban Chess" is very balanced White/Black performance (but not the position itself), with very few draws (10-15%), and the play should be very aggressive from the start.

[D]1p6/2p3kn/3p2pp/4pppp/5ppp/8/PPPPPPPP/PPPPPPKN w - - 0 0

The game length distribution is the following:

Image

Median game: 33 moves
Commonest game: 28 moves

All in all, I think I will use "Peasant Revolt Chess" to check engines in endgame imbalances and Syzygy use, the position seems very peculiar, sensitive and still undecided if Black can win or White can hold.

User avatar
Laskos
Posts: 10833
Joined: Wed Jul 26, 2006 8:21 pm
Full name: Kai Laskos

Re: "Peasant Revolt" Endgame Chess

Post by Laskos » Sun Sep 03, 2017 7:02 am

A small rating list of top 10 engines at Blitz 360''+ 3.6''. The error margins as shown in Ordo are really 1.7-1.8 times smaller using pentanomial variance. Syzygy 6-men in use for engines supporting them.

Code: Select all

     Program                        Elo    +    -   Games   Score   Av.Op.  Draws

   1 Komodo 11.2.2 64-bit         : 3156   40   40    90    67.8 %   2983   53.3 %
   2 Deep Shredder 13 x64         : 3091   42   42    90    60.6 %   2990   54.4 %
   3 Stockfish 260817 64 BMI2     : 3082   40   40    90    59.4 %   2991   47.8 %
   4 Gull 3 x64                   : 3057   40   40    90    56.7 %   2994   44.4 %
   5 Houdini 5.01 Pro x64-popc    : 3014   42   42    90    51.7 %   2998   52.2 %
   6 Fire 5 x64                   : 2971   42   42    90    46.7 %   3003   46.7 %
   7 Fritz 15                     : 2952   40   40    90    44.4 %   3005   62.2 %
   8 Andscacs 0.91b               : 2918   44   44    90    40.6 %   3009   56.7 %
   9 Fizbo 1.9                    : 2909   42   42    90    39.4 %   3010   47.8 %
  10 Booot 6.2_x64                : 2849   42   42    90    32.8 %   3017   56.7 %


Games        : 450 (finished)

White Wins   : 6 (1.3 %)
Black Wins   : 209 (46.4 %)
Draws        : 235 (52.2 %)
Unfinished   : 0

White Score  : 27.4 %
Black Score  : 72.6 %
Observe the underperformance of Stockfish in this disbalance, although it used Syzygy 6-men from SSD.

User avatar
fern
Posts: 8755
Joined: Sun Feb 26, 2006 3:07 pm

Re: "Peasant Revolt" Endgame Chess

Post by fern » Tue Sep 05, 2017 7:57 pm

The book is on the way since 10 days ago or so....
Let me know on arrival.

Misterious transactions here regards
Fern

User avatar
Laskos
Posts: 10833
Joined: Wed Jul 26, 2006 8:21 pm
Full name: Kai Laskos

Re: "Peasant Revolt" Endgame Chess

Post by Laskos » Wed Sep 06, 2017 6:21 am

fern wrote:The book is on the way since 10 days ago or so....
Let me know on arrival.

Misterious transactions here regards
Fern
Nothing yet here, checked this morning. My personal messaging on talkchess was messed up, the box was full or something like that, I still have a message to you in the outbox not received by you. Thanks, and if we manage personal messaging, I will send you the photos of two coins I bought when on that trip, you will choose an emperor you like more (sorry, no republican ones bought there). Thanks again!

Mafia never dies!

User avatar
fern
Posts: 8755
Joined: Sun Feb 26, 2006 3:07 pm

Re: "Peasant Revolt" Endgame Chess

Post by fern » Wed Sep 06, 2017 4:37 pm

according to Chilean mail service, the package is already in Rumania, so it is a matter of local delivery now

Fern

Post Reply