Komodo? Nein.... Doch!

[Spacious_Mind]

isn't there some interface that allows this? i know people have mentioned it.

You can set it with Winboard, but I sense there is lag between moves and I am not confident yet with it being accurate but I might be wrong. Perhaps a Winboard expert can explain how it might work in more detail.

regards

Nick

[Spacious_Mind]

Here are the results for the second set of 300 games.

5 PLY TEST 2

Code: Select all

    Program                          Elo    +   -   Games   Score   Av.Op.  Draws

  1 Doch_09.980_32_ja              : 2014   59  59   100    52.5 %   1997   27.0 %
  2 Komodo-10-32-ja                : 2014   55  55   100    52.5 %   1997   35.0 %
  3 Doch32-134-ja                  : 2003   59  59   100    50.5 %   1999   27.0 %
  4 Doch32-133-ja                  : 2000   56  56   100    50.0 %   2000   34.0 %
  5 Doch13-ja                      : 1994   57  58   100    49.0 %   2001   30.0 %
  6 Doch12-ja                      : 1974   58  58   100    45.5 %   2005   29.0 %

Komodo 1.0 from last in first test to equal first in second test Not bad!

Code: Select all

Individual statistics:

1 Doch_09.980_32_ja         : 2014  100 (+ 39,= 27,- 34), 52.5 %

Komodo-10-32-ja               :  20 (+  8,=  5,-  7), 52.5 %
Doch32-134-ja                 :  20 (+  6,=  5,-  9), 42.5 %
Doch12-ja                     :  20 (+ 10,=  4,-  6), 60.0 %
Doch13-ja                     :  20 (+  8,=  5,-  7), 52.5 %
Doch32-133-ja                 :  20 (+  7,=  8,-  5), 55.0 %

2 Komodo-10-32-ja           : 2014  100 (+ 35,= 35,- 30), 52.5 %

Doch_09.980_32_ja             :  20 (+  7,=  5,-  8), 47.5 %
Doch32-134-ja                 :  20 (+  8,=  6,-  6), 55.0 %
Doch12-ja                     :  20 (+ 11,=  6,-  3), 70.0 %
Doch13-ja                     :  20 (+  5,=  8,-  7), 45.0 %
Doch32-133-ja                 :  20 (+  4,= 10,-  6), 45.0 %

3 Doch32-134-ja             : 2003  100 (+ 37,= 27,- 36), 50.5 %

Doch_09.980_32_ja             :  20 (+  9,=  5,-  6), 57.5 %
Komodo-10-32-ja               :  20 (+  6,=  6,-  8), 45.0 %
Doch12-ja                     :  20 (+  4,=  7,-  9), 37.5 %
Doch13-ja                     :  20 (+ 11,=  5,-  4), 67.5 %
Doch32-133-ja                 :  20 (+  7,=  4,-  9), 45.0 %

4 Doch32-133-ja             : 2000  100 (+ 33,= 34,- 33), 50.0 %

Doch_09.980_32_ja             :  20 (+  5,=  8,-  7), 45.0 %
Komodo-10-32-ja               :  20 (+  6,= 10,-  4), 55.0 %
Doch32-134-ja                 :  20 (+  9,=  4,-  7), 55.0 %
Doch12-ja                     :  20 (+  6,=  6,-  8), 45.0 %
Doch13-ja                     :  20 (+  7,=  6,-  7), 50.0 %

5 Doch13-ja                 : 1994  100 (+ 34,= 30,- 36), 49.0 %

Doch_09.980_32_ja             :  20 (+  7,=  5,-  8), 47.5 %
Komodo-10-32-ja               :  20 (+  7,=  8,-  5), 55.0 %
Doch32-134-ja                 :  20 (+  4,=  5,- 11), 32.5 %
Doch12-ja                     :  20 (+  9,=  6,-  5), 60.0 %
Doch32-133-ja                 :  20 (+  7,=  6,-  7), 50.0 %

6 Doch12-ja                 : 1974  100 (+ 31,= 29,- 40), 45.5 %

Doch_09.980_32_ja             :  20 (+  6,=  4,- 10), 40.0 %
Komodo-10-32-ja               :  20 (+  3,=  6,- 11), 30.0 %
Doch32-134-ja                 :  20 (+  9,=  7,-  4), 62.5 %
Doch13-ja                     :  20 (+  5,=  6,-  9), 40.0 %
Doch32-133-ja                 :  20 (+  8,=  6,-  6), 55.0 %

Now more tests are needed. I will run this a couple of more times for a total of 1200 games.

best regards

Nick

[Spacious_Mind]

Here is the third test.

5 PLY TEST 3

Code: Select all

    Program                          Elo    +   -   Games   Score   Av.Op.  Draws

  1 Doch32-133-ja                  : 2029   58  58   100    55.0 %   1994   30.0 %
  2 Doch12-ja                      : 2009   55  55   100    51.5 %   1998   35.0 %
  3 Doch13-ja                      : 2009   55  55   100    51.5 %   1998   35.0 %
  4 Komodo-10-32-ja                : 1991   56  56   100    48.5 %   2002   33.0 %
  5 Doch32-134-ja                  : 1983   56  56   100    47.0 %   2003   34.0 %
  6 Doch_09.980_32_ja              : 1980   55  56   100    46.5 %   2004   35.0 %

Code: Select all

Individual statistics:

1 Doch32-133-ja             : 2029  100 (+ 40,= 30,- 30), 55.0 %

Doch_09.980_32_ja             :  20 (+  8,=  6,-  6), 55.0 %
Komodo-10-32-ja               :  20 (+ 10,=  3,-  7), 57.5 %
Doch32-134-ja                 :  20 (+  5,=  9,-  6), 47.5 %
Doch12-ja                     :  20 (+ 10,=  6,-  4), 65.0 %
Doch13-ja                     :  20 (+  7,=  6,-  7), 50.0 %

2 Doch12-ja                 : 2009  100 (+ 34,= 35,- 31), 51.5 %

Doch_09.980_32_ja             :  20 (+  5,=  8,-  7), 45.0 %
Komodo-10-32-ja               :  20 (+  6,=  9,-  5), 52.5 %
Doch32-134-ja                 :  20 (+ 10,=  6,-  4), 65.0 %
Doch13-ja                     :  20 (+  9,=  6,-  5), 60.0 %
Doch32-133-ja                 :  20 (+  4,=  6,- 10), 35.0 %

3 Doch13-ja                 : 2009  100 (+ 34,= 35,- 31), 51.5 %

Doch_09.980_32_ja             :  20 (+  9,=  8,-  3), 65.0 %
Komodo-10-32-ja               :  20 (+  7,=  8,-  5), 55.0 %
Doch32-134-ja                 :  20 (+  6,=  7,-  7), 47.5 %
Doch12-ja                     :  20 (+  5,=  6,-  9), 40.0 %
Doch32-133-ja                 :  20 (+  7,=  6,-  7), 50.0 %

4 Komodo-10-32-ja           : 1991  100 (+ 32,= 33,- 35), 48.5 %

Doch_09.980_32_ja             :  20 (+  6,=  7,-  7), 47.5 %
Doch32-134-ja                 :  20 (+  9,=  6,-  5), 60.0 %
Doch12-ja                     :  20 (+  5,=  9,-  6), 47.5 %
Doch13-ja                     :  20 (+  5,=  8,-  7), 45.0 %
Doch32-133-ja                 :  20 (+  7,=  3,- 10), 42.5 %

5 Doch32-134-ja             : 1983  100 (+ 30,= 34,- 36), 47.0 %

Doch_09.980_32_ja             :  20 (+  8,=  6,-  6), 55.0 %
Komodo-10-32-ja               :  20 (+  5,=  6,-  9), 40.0 %
Doch12-ja                     :  20 (+  4,=  6,- 10), 35.0 %
Doch13-ja                     :  20 (+  7,=  7,-  6), 52.5 %
Doch32-133-ja                 :  20 (+  6,=  9,-  5), 52.5 %

6 Doch_09.980_32_ja         : 1980  100 (+ 29,= 35,- 36), 46.5 %

Komodo-10-32-ja               :  20 (+  7,=  7,-  6), 52.5 %
Doch32-134-ja                 :  20 (+  6,=  6,-  8), 45.0 %
Doch12-ja                     :  20 (+  7,=  8,-  5), 55.0 %
Doch13-ja                     :  20 (+  3,=  8,-  9), 35.0 %
Doch32-133-ja                 :  20 (+  6,=  6,-  8), 45.0 %

Best regards

Nick

[Spacious_Mind]

Here is the 4th Test:

5 PLY TEST 4

Code: Select all

    Program                          Elo    +   -   Games   Score   Av.Op.  Draws

  1 Doch13-ja                      : 2032   59  59   100    55.5 %   1994   27.0 %
  2 Doch32-134-ja                  : 2029   60  59   100    55.0 %   1994   26.0 %
  3 Doch_09.980_32_ja              : 2003   56  56   100    50.5 %   1999   33.0 %
  4 Doch12-ja                      : 1991   54  54   100    48.5 %   2002   39.0 %
  5 Doch32-133-ja                  : 1988   55  55   100    48.0 %   2002   36.0 %
  6 Komodo-10-32-ja                : 1956   61  61   100    42.5 %   2009   23.0 %

Code: Select all

Individual statistics:

1 Doch13-ja                 : 2032  100 (+ 42,= 27,- 31), 55.5 %

Doch_09.980_32_ja             :  20 (+  8,=  6,-  6), 55.0 %
Komodo-10-32-ja               :  20 (+ 12,=  1,-  7), 62.5 %
Doch32-134-ja                 :  20 (+  7,=  4,-  9), 45.0 %
Doch12-ja                     :  20 (+  7,=  7,-  6), 52.5 %
Doch32-133-ja                 :  20 (+  8,=  9,-  3), 62.5 %

2 Doch32-134-ja             : 2029  100 (+ 42,= 26,- 32), 55.0 %

Doch_09.980_32_ja             :  20 (+  8,=  7,-  5), 57.5 %
Komodo-10-32-ja               :  20 (+ 11,=  2,-  7), 60.0 %
Doch12-ja                     :  20 (+  6,=  8,-  6), 50.0 %
Doch13-ja                     :  20 (+  9,=  4,-  7), 55.0 %
Doch32-133-ja                 :  20 (+  8,=  5,-  7), 52.5 %

3 Doch_09.980_32_ja         : 2003  100 (+ 34,= 33,- 33), 50.5 %

Komodo-10-32-ja               :  20 (+ 11,=  5,-  4), 67.5 %
Doch32-134-ja                 :  20 (+  5,=  7,-  8), 42.5 %
Doch12-ja                     :  20 (+  7,=  8,-  5), 55.0 %
Doch13-ja                     :  20 (+  6,=  6,-  8), 45.0 %
Doch32-133-ja                 :  20 (+  5,=  7,-  8), 42.5 %

4 Doch12-ja                 : 1991  100 (+ 29,= 39,- 32), 48.5 %

Doch_09.980_32_ja             :  20 (+  5,=  8,-  7), 45.0 %
Komodo-10-32-ja               :  20 (+  5,=  8,-  7), 45.0 %
Doch32-134-ja                 :  20 (+  6,=  8,-  6), 50.0 %
Doch13-ja                     :  20 (+  6,=  7,-  7), 47.5 %
Doch32-133-ja                 :  20 (+  7,=  8,-  5), 55.0 %

5 Doch32-133-ja             : 1988  100 (+ 30,= 36,- 34), 48.0 %

Doch_09.980_32_ja             :  20 (+  8,=  7,-  5), 57.5 %
Komodo-10-32-ja               :  20 (+  7,=  7,-  6), 52.5 %
Doch32-134-ja                 :  20 (+  7,=  5,-  8), 47.5 %
Doch12-ja                     :  20 (+  5,=  8,-  7), 45.0 %
Doch13-ja                     :  20 (+  3,=  9,-  8), 37.5 %

6 Komodo-10-32-ja           : 1956  100 (+ 31,= 23,- 46), 42.5 %

Doch_09.980_32_ja             :  20 (+  4,=  5,- 11), 32.5 %
Doch32-134-ja                 :  20 (+  7,=  2,- 11), 40.0 %
Doch12-ja                     :  20 (+  7,=  8,-  5), 55.0 %
Doch13-ja                     :  20 (+  7,=  1,- 12), 37.5 %
Doch32-133-ja                 :  20 (+  6,=  7,-  7), 47.5 %

One more test to go. Changed my mind again Have decided that I want 100 games for each engine against each engine therefore 1500 games in total.

Best regards

Nick

[Spacious_Mind]

Here is the 5th and last 5 Ply Test for Komodo and Doch Engines.

5 PLY TEST 5

Code: Select all

    Program                          Elo    +   -   Games   Score   Av.Op.  Draws

  1 Doch_09.980_32_ja              : 2038   57  56   100    56.5 %   1992   33.0 %
  2 Doch13-ja                      : 2017   56  56   100    53.0 %   1997   34.0 %
  3 Doch32-134-ja                  : 2009   59  59   100    51.5 %   1998   27.0 %
  4 Doch12-ja                      : 2000   60  60   100    50.0 %   2000   24.0 %
  5 Komodo-10-32-ja                : 1977   58  59   100    46.0 %   2005   28.0 %
  6 Doch32-133-ja                  : 1959   59  60   100    43.0 %   2008   26.0 %

Code: Select all

Individual statistics:

1 Doch_09.980_32_ja         : 2038  100 (+ 40,= 33,- 27), 56.5 %

Komodo-10-32-ja               :  20 (+ 10,=  5,-  5), 62.5 %
Doch32-134-ja                 :  20 (+  5,=  8,-  7), 45.0 %
Doch12-ja                     :  20 (+ 10,=  5,-  5), 62.5 %
Doch13-ja                     :  20 (+  5,= 10,-  5), 50.0 %
Doch32-133-ja                 :  20 (+ 10,=  5,-  5), 62.5 %

2 Doch13-ja                 : 2017  100 (+ 36,= 34,- 30), 53.0 %

Doch_09.980_32_ja             :  20 (+  5,= 10,-  5), 50.0 %
Komodo-10-32-ja               :  20 (+  5,=  8,-  7), 45.0 %
Doch32-134-ja                 :  20 (+  8,=  4,-  8), 50.0 %
Doch12-ja                     :  20 (+  9,=  6,-  5), 60.0 %
Doch32-133-ja                 :  20 (+  9,=  6,-  5), 60.0 %

3 Doch32-134-ja             : 2009  100 (+ 38,= 27,- 35), 51.5 %

Doch_09.980_32_ja             :  20 (+  7,=  8,-  5), 55.0 %
Komodo-10-32-ja               :  20 (+  9,=  4,-  7), 55.0 %
Doch12-ja                     :  20 (+  5,=  5,- 10), 37.5 %
Doch13-ja                     :  20 (+  8,=  4,-  8), 50.0 %
Doch32-133-ja                 :  20 (+  9,=  6,-  5), 60.0 %

4 Doch12-ja                 : 2000  100 (+ 38,= 24,- 38), 50.0 %

Doch_09.980_32_ja             :  20 (+  5,=  5,- 10), 37.5 %
Komodo-10-32-ja               :  20 (+  7,=  5,-  8), 47.5 %
Doch32-134-ja                 :  20 (+ 10,=  5,-  5), 62.5 %
Doch13-ja                     :  20 (+  5,=  6,-  9), 40.0 %
Doch32-133-ja                 :  20 (+ 11,=  3,-  6), 62.5 %

5 Komodo-10-32-ja           : 1977  100 (+ 32,= 28,- 40), 46.0 %

Doch_09.980_32_ja             :  20 (+  5,=  5,- 10), 37.5 %
Doch32-134-ja                 :  20 (+  7,=  4,-  9), 45.0 %
Doch12-ja                     :  20 (+  8,=  5,-  7), 52.5 %
Doch13-ja                     :  20 (+  7,=  8,-  5), 55.0 %
Doch32-133-ja                 :  20 (+  5,=  6,-  9), 40.0 %

6 Doch32-133-ja             : 1959  100 (+ 30,= 26,- 44), 43.0 %

Doch_09.980_32_ja             :  20 (+  5,=  5,- 10), 37.5 %
Komodo-10-32-ja               :  20 (+  9,=  6,-  5), 60.0 %
Doch32-134-ja                 :  20 (+  5,=  6,-  9), 40.0 %
Doch12-ja                     :  20 (+  6,=  3,- 11), 37.5 %
Doch13-ja                     :  20 (+  5,=  6,-  9), 40.0 %

Best Regards

Nick

[Spacious_Mind]

Here is the Final Table after 1500 games:

FINAL TABLE

Code: Select all

    Program                          Elo    +   -   Games   Score   Av.Op.  Draws

  1 Doch13-ja                      : 2015   25  25   500    52.6 %   1997   31.6 %
  2 Doch_09.980_32_ja              : 2004   26  26   500    50.7 %   1999   29.8 %
  3 Doch32-133-ja                  : 2002   25  25   500    50.4 %   2000   31.2 %
  4 Doch32-134-ja                  : 1999   26  26   500    49.9 %   2000   29.0 %
  5 Doch12-ja                      : 1999   26  26   500    49.9 %   2000   29.8 %
  6 Komodo-10-32-ja                : 1980   26  26   500    46.5 %   2004   29.0 %

In the end not much in it between the different DOCH versions. Doch13-ja might be slighltly better than the rest but it is hard to tell. It seems that Komodo-10-32-ja is only slightly weaker at 5 Ply but not by much.

Here are the games if anyone is interested for download.

http://spacious-mind.com/forum_reports/ ... komodo.zip

The final individual performances were as follows;

Code: Select all

Individual statistics:

1 Doch13-ja                 : 2015  500 (+184,=158,-158), 52.6 %

Doch_09.980_32_ja             : 100 (+ 37,= 32,- 31), 53.0 %
Komodo-10-32-ja               : 100 (+ 43,= 28,- 29), 57.0 %
Doch32-134-ja                 : 100 (+ 34,= 29,- 37), 48.5 %
Doch12-ja                     : 100 (+ 37,= 32,- 31), 53.0 %
Doch32-133-ja                 : 100 (+ 33,= 37,- 30), 51.5 %

2 Doch_09.980_32_ja         : 2004  500 (+179,=149,-172), 50.7 %

Komodo-10-32-ja               : 100 (+ 46,= 27,- 27), 59.5 %
Doch32-134-ja                 : 100 (+ 26,= 34,- 40), 43.0 %
Doch12-ja                     : 100 (+ 43,= 25,- 32), 55.5 %
Doch13-ja                     : 100 (+ 31,= 32,- 37), 47.0 %
Doch32-133-ja                 : 100 (+ 33,= 31,- 36), 48.5 %

3 Doch32-133-ja             : 2002  500 (+174,=156,-170), 50.4 %

Doch_09.980_32_ja             : 100 (+ 36,= 31,- 33), 51.5 %
Komodo-10-32-ja               : 100 (+ 39,= 30,- 31), 54.0 %
Doch32-134-ja                 : 100 (+ 35,= 30,- 35), 50.0 %
Doch12-ja                     : 100 (+ 34,= 28,- 38), 48.0 %
Doch13-ja                     : 100 (+ 30,= 37,- 33), 48.5 %

4 Doch32-134-ja             : 1999  500 (+177,=145,-178), 49.9 %

Doch_09.980_32_ja             : 100 (+ 40,= 34,- 26), 57.0 %
Komodo-10-32-ja               : 100 (+ 41,= 24,- 35), 53.0 %
Doch12-ja                     : 100 (+ 24,= 28,- 48), 38.0 %
Doch13-ja                     : 100 (+ 37,= 29,- 34), 51.5 %
Doch32-133-ja                 : 100 (+ 35,= 30,- 35), 50.0 %

5 Doch12-ja                 : 1999  500 (+175,=149,-176), 49.9 %

Doch_09.980_32_ja             : 100 (+ 32,= 25,- 43), 44.5 %
Komodo-10-32-ja               : 100 (+ 26,= 36,- 38), 44.0 %
Doch32-134-ja                 : 100 (+ 48,= 28,- 24), 62.0 %
Doch13-ja                     : 100 (+ 31,= 32,- 37), 47.0 %
Doch32-133-ja                 : 100 (+ 38,= 28,- 34), 52.0 %

6 Komodo-10-32-ja           : 1980  500 (+160,=145,-195), 46.5 %

Doch_09.980_32_ja             : 100 (+ 27,= 27,- 46), 40.5 %
Doch32-134-ja                 : 100 (+ 35,= 24,- 41), 47.0 %
Doch12-ja                     : 100 (+ 38,= 36,- 26), 56.0 %
Doch13-ja                     : 100 (+ 29,= 28,- 43), 43.0 %
Doch32-133-ja                 : 100 (+ 31,= 30,- 39), 46.0 %

Best Regards

Nick

[Richard Allbert]

Nice tournaments.

In the results, the error bars overlap between first and last place... so conclusion is all the engines are equal?

After about 1600 games the error drops to +/- 16, so maybe a few more games are needed for definite conclusions.

It's intersting to see the results, though. Fixed depth vs fast TC is something I can never decide on.

One gives lots of games very fast, but doesn't take into account engine speed, the other not enough games..

Richard

[Spacious_Mind]

Nice tournaments.

In the results, the error bars overlap between first and last place... so conclusion is all the engines are equal?

After about 1600 games the error drops to +/- 16, so maybe a few more games are needed for definite conclusions.

It's intersting to see the results, though. Fixed depth vs fast TC is something I can never decide on.

One gives lots of games very fast, but doesn't take into account engine speed, the other not enough games..

Richard

Thanks for your observations. Will play another 600 games overnight and see if this helps for the results to be more conclusive.

But it is possible that all the DOCH are in fact the same at 5 ply, or changes are so minimal between versions that it is hardly noticable therefore the results are just random acts. With Komodo the only exception, since it seems to be that it's search was tweaked for speed as per comments made earlier by Larry Kaufmann.

I have not tried to make one version play back games of another version yet. So I am speculating aloud.

What I might also do is to play the same set again at 1 second per move and compare the individual versions performance results. This would give me perhaps an idea of what effect 5 ply has compared to normal set time per move games.

Also none of this really means much on how they would fare against other engines of similar strength. It may well be that another version or even Komodo does better then.

For me it is just a curiosity test

Best regards

Nick

[Spacious_Mind]

I have now played 2100 games and the only thing that remains clear is Doch13-ja and Doch_09.980_32_ja are better than Komodo 1.0 at 5 ply.

Code: Select all

    Program                          Elo    +   -   Games   Score   Av.Op.  Draws

  1 Doch13-ja                      : 2016   22  22   700    52.8 %   1997   30.4 %
  2 Doch_09.980_32_ja              : 2012   22  22   700    52.0 %   1998   28.3 %
  3 Doch32-133-ja                  : 1998   22  22   700    49.6 %   2000   29.4 %
  4 Doch32-134-ja                  : 1996   22  22   700    49.4 %   2001   28.4 %
  5 Doch12-ja                      : 1995   22  22   700    49.1 %   2001   29.4 %
  6 Komodo-10-32-ja                : 1983   22  22   700    47.1 %   2003   28.6 %

Code: Select all

Individual statistics:

1 Doch13-ja                 : 2016  700 (+263,=213,-224), 52.8 %

Doch_09.980_32_ja             : 140 (+ 51,= 43,- 46), 51.8 %
Komodo-10-32-ja               : 140 (+ 60,= 42,- 38), 57.9 %
Doch32-134-ja                 : 140 (+ 55,= 36,- 49), 52.1 %
Doch12-ja                     : 140 (+ 51,= 43,- 46), 51.8 %
Doch32-133-ja                 : 140 (+ 46,= 49,- 45), 50.4 %

2 Doch_09.980_32_ja         : 2012  700 (+265,=198,-237), 52.0 %

Komodo-10-32-ja               : 140 (+ 65,= 32,- 43), 57.9 %
Doch32-134-ja                 : 140 (+ 43,= 46,- 51), 47.1 %
Doch12-ja                     : 140 (+ 61,= 36,- 43), 56.4 %
Doch13-ja                     : 140 (+ 46,= 43,- 51), 48.2 %
Doch32-133-ja                 : 140 (+ 50,= 41,- 49), 50.4 %

3 Doch32-133-ja             : 1998  700 (+244,=206,-250), 49.6 %

Doch_09.980_32_ja             : 140 (+ 49,= 41,- 50), 49.6 %
Komodo-10-32-ja               : 140 (+ 58,= 40,- 42), 55.7 %
Doch32-134-ja                 : 140 (+ 47,= 39,- 54), 47.5 %
Doch12-ja                     : 140 (+ 45,= 37,- 58), 45.4 %
Doch13-ja                     : 140 (+ 45,= 49,- 46), 49.6 %

4 Doch32-134-ja             : 1996  700 (+246,=199,-255), 49.4 %

Doch_09.980_32_ja             : 140 (+ 51,= 46,- 43), 52.9 %
Komodo-10-32-ja               : 140 (+ 51,= 37,- 52), 49.6 %
Doch12-ja                     : 140 (+ 41,= 41,- 58), 43.9 %
Doch13-ja                     : 140 (+ 49,= 36,- 55), 47.9 %
Doch32-133-ja                 : 140 (+ 54,= 39,- 47), 52.5 %

5 Doch12-ja                 : 1995  700 (+241,=206,-253), 49.1 %

Doch_09.980_32_ja             : 140 (+ 43,= 36,- 61), 43.6 %
Komodo-10-32-ja               : 140 (+ 36,= 49,- 55), 43.2 %
Doch32-134-ja                 : 140 (+ 58,= 41,- 41), 56.1 %
Doch13-ja                     : 140 (+ 46,= 43,- 51), 48.2 %
Doch32-133-ja                 : 140 (+ 58,= 37,- 45), 54.6 %

6 Komodo-10-32-ja           : 1983  700 (+230,=200,-270), 47.1 %

Doch_09.980_32_ja             : 140 (+ 43,= 32,- 65), 42.1 %
Doch32-134-ja                 : 140 (+ 52,= 37,- 51), 50.4 %
Doch12-ja                     : 140 (+ 55,= 49,- 36), 56.8 %
Doch13-ja                     : 140 (+ 38,= 42,- 60), 42.1 %
Doch32-133-ja                 : 140 (+ 42,= 40,- 58), 44.3 %

Interesting are the results between Komodo-10-32-ja and Doch32-134-ja, which seems to show Komodo was a result of the last Doch version Doch32-134-ja, which I guess is only natural .

Going to play some 1 second/move games next for comparison.

Best regards

Nick

[lkaufman]

Doch 13, Doch 133, and Doch 134 should all be identical unless you are using a personality file. The earlier versions had various bugs when using that file, doch 134 fixed those bugs.