Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: u2600 ICC Rating List - Mar 29

Author: Dan Homan

Date: 09:02:52 03/31/99

Go up one level in this thread


On March 30, 1999 at 11:50:15, Will Singleton wrote:

>On March 30, 1999 at 07:01:59, Dan Homan wrote:
>
>>On March 29, 1999 at 20:08:26, Will Singleton wrote:
>>
>>>Under 2600 Club
>>>Mar 29 1999
>>>
>>>Program      ICC Blitz  Change   Games     Platform    Author
>>>
>>>Hossa            2613    +262     359      P/350       Steffen Jakob
>>>PostModernist    2607     +47      20      K6/300      Andrew Williams
>>>ZChess           2591     +15       1      K6/266      Franck Zibi
>>>Rajah            2544     +13     109      P/166       V. Manohararajah
>>>Rookie           2525     +11     109      P/360       Marcel van Kervinck
>>>Stobor           2519     -15       2      K6/350      Tom Kerrigan
>>>EXchess          2495     +49      10      Cel/400     Dan Homan
>>>TDChess          2491    -188     384      P/400       Jon Baxter
>>>LambChop         2453                      P/133       Pete McKenzie
>>>Grok             2390     +37       8      K6/300      Peter Kappler
>>>Amateur          2375     -75      43      Mac/300     Will Singleton
>>>InmiChess        2292                      Cyrix/233   Werner Inmann
>>>
>>>
>>> ICC account names that differ from the program name:
>>> ----------------------------------------------------------------------------
>>> Blik (Rookie), zchess1 (Zchess), Kerrigan (Stobor),  XXXX-2 (XXXXII).
>>> ----------------------------------------------------------------------------
>>>
>>>
>>>Notable
>>>
>>>We saw some large ratings variations this week, with Hossa taking the top
>>>spot away from TDchess.  In fact, two programs have broken the 2600 barrier,
>>>Hossa and PostModernist.  Congrats to Steffen and Andrew!
>>>
>>>New highs were reached by Hossa (2638), PostModernist (2607), and
>>>TDchess (2775).  TDchess lost almost 300 points from it's high on March 25,
>>>but was also the games-played leader with 384.  There does seem to be some
>>>correlation between games_played and rating swings, though some programs
>>>are more susceptible to this than others.
>>>
>>>
>>>Changes
>>>
>>>Stobor rejoins the active list, BugChess and Pedestrian go inactive.
>>>
>>>
>>>u2600 Position of the Week
>>>
>>>This comes from a 30 0 game played on ICC this week between GM Wilder and
>>>Ferret.  Wilder, playing white, has given up the forced draw to try for a
>>>win, which he apparently sensed was there.  It looks like he was right, but
>>>he did not find the move, which was 45. Rh8+!
>>>
>>>1rk5/7R/p2rb1R1/2p2p1P/4p3/1P2P1P1/P1K2P2/8 w - -
>>>
>>>Amateur 0.95, g3/300
>>>ply   score     time     nodes      n/s    pv
>>>
>>>  2    -149     0.01       831    49860
>>>  3    -105     0.02      1440    43200  Rg7 Rb5 h6
>>>  4    -117     0.06      4594    45900  Rg7 Bd7 h6 Bc6
>>>  5    -130     0.14     10939    46860  Rg7 Rd7 h6 Rb7 Rh8 Kc7
>>>  6    -132     0.36     30172    50280  Rg7 Rd7 h6 c4 Pxc4 Rxg7 Pxg7 Bxc4
>>>  6    -118     1.07     60819    54420  Re7 Rb6 Rg7 Rd7 Re8 Rd8 Re7
>>>  7     -62     3.06    161330    52020  Re7 Rb6 h6 c4 Pxc4
>>>  8     -68     6.12    345462    55680  Re7 Rb6 h6 Bxb3 Pxb3 Rxg6 h7 Kd8 Re8
>>>Kxe8 Qh8 Kd7
>>>  9     -40    19.02   1106827    58140  Re7 Rb6 h6 Bxb3 Pxb3 Rxg6 h7 Rh6 Re8
>>>Kc7 Qh8 Rxh8 Rxh8
>>> 10     -40    63.47   3761689    58920  Re7 Rb6 h6 Bxb3 Pxb3 Rxg6 h7 Rh6 Re8
>>>Kc7 Qh8 Rxh8 Rxh8
>>> 11     -35   271.35  16472404    60600  Re7 Rb6 h6 Bxb3 Pxb3 Rxg6 h7 Rh6 Re8
>>>Kd7 Qh8
>>> 11      72   493.49  31288461    63360  Rh8 Kc7 Rxb8 Kxb8 Kc1 c4 h6 Pxb3 h7 Rc6
>>>Kd1 Rd6 Ke1
>>
>>EXchess wants to play this move after just a couple of seconds....
>>
>>  6.  -0.31     1    83157   1. Rgg7 Bd7 2. h6 Rb5 3. Kb1 Rd2
>>  6.  -0.31     1    94108   1. Rgg7 Bd7 2. h6 Rb5 3. Kb1 Rd2
>>  7.  -0.16     2   131487   1. Rgg7 Bd7 2. h6 Kc7 3. Kc1 Rd3 4. Kb1
>>  7.   ++       2   176580   1. Rh8+
>>  7.   --
>>  7.   0.00     3   236701   1. Rh8+ Kb7 2. Rg7+ Ka8 3. Rhh7 Rbb6 4. Rh8+ Rb8
>>  7.   0.00     3   268556   1. Rh8+ Kb7 2. Rg7+ Ka8 3. Rhh7 Rbb6 4. Rh8+ Rb8
>>  8.   0.00     4   450392   1. Rh8+ Kb7 2. Rg7+ Ka8 3. Rhh7 Rbb6 4. Rh8+ Rb8
>>  8.   0.00     6   605807   1. Rh8+ Kb7 2. Rg7+ Ka8 3. Rhh7 Rbb6 4. Rh8+ Rb8
>>  9.   ++       8   795537   1. Rh8+
>>  9.   --
>>  9.   0.30    12  1329524   1. Rh8+ Kb7 2. Rg7+ Ka8 3. Rxb8+ Kxb8 4. h6 Rd8
>>5. h7 c4 6. Rg6 cxb3+ 7. axb3 Bc8 8. Rc6
>>  9.   0.30    15  1680651   1. Rh8+ Kb7 2. Rg7+ Ka8 3. Rxb8+ Kxb8 4. h6 Rd8
>>5. h7 c4 6. Rg6 cxb3+ 7. axb3 Bc8 8. Rc6
>> 10.   ++      26  2974960   1. Rh8+
>> 10.   --
>> 10.   0.82    37  4257394   1. Rh8+ Kc7 2. Rxb8 Kxb8 3. Kb2 Rd2+ 4. Kc1 Rxf2
>>5. Rxe6 Rxa2 6. h6 Ra1+ 7. Kc2 Ra2+ 8. Kd1
>> 10.   0.82    41  4698644   1. Rh8+ Kc7 2. Rxb8 Kxb8 3. Kb2 Rd2+ 4. Kc1 Rxf2
>>5. Rxe6 Rxa2 6. h6 Ra1+ 7. Kc2 Ra2+ 8. Kd1
>> 11.   ++      46  5333035   1. Rh8+
>> 11.   1.19    70  8207374   1. Rh8+ Kc7 2. Rxb8 Kxb8 3. Kc1 Kc7 4. h6 Bg8
>>5. Rxg8 Rxh6 6. Rg5 Rf6 7. a3 Kb8
>
>
>That's pretty cool, Dan.  Like the others, it takes 10-11 ply to see the proper
>continuation, but you like it earlier, for positional reasons.
>
>What do you attribute that to?

EXchess assumes a draw if a position is repeated twice in the search.
I believe this happens in the line at ply 7.  So at this point
EXchess thinks it can force a draw.  Later on, EXchess sees that it
can get more than a draw....  How much more increases with search depth
to ply 11.  I think that EXchess' confidence that it can force a
draw at ply 7 is due to the fact that EXchess values the passed pawn
(which is located on the opposite wing from the king) rather strongly.

 - Dan

>
>Will



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.