Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: One mate to solve.

Author: leonid

Date: 16:43:41 03/02/02

Go up one level in this thread


On March 02, 2002 at 19:20:49, leonid wrote:

>On March 02, 2002 at 18:31:18, Heiner Marxen wrote:
>
>>On March 02, 2002 at 14:33:53, John Merlino wrote:
>>
>>>On March 02, 2002 at 08:21:10, leonid wrote:
>>>
>>>>[D]3Q3R/Q2b2Q1/1Q1q1Q2/1qQqQq1B/Nr1q1rNB/2qkq3/1qnbnq2/Q2K2QR w - -
>>>>
>>>>Please indicate your result.
>>>>
>>>>Thanks,
>>>>Leonid.
>>>
>>>Chessmaster 8000 says Mate in 10 in 2 seconds:
>>>
>>>Time	Depth	Score	Positions	Moves
>>>0:02	1/3	Mate10	164719		1.Nxf2+ Rxf2 2.Nxb2+ Qxb2 3.Bxe2+
>>>					Qxe2+ 4.Qxe2+ Rxe2 5.Qfxd4+ Nxd4
>>>					6.Q7xd4+ Rxd4 7.Qcxd4+ Qdxd4 8.Qbxd4+
>>>					Qdxd4 9.Qaxd4+ Qxd4 10.Qgxd4#
>>>0:06	1/4	Mate10	317014		1.Nxf2+ Rxf2 2.Nxb2+ Qxb2 3.Bxe2+
>>>					Qxe2+ 4.Qxe2+ Rxe2 5.Qfxd4+ Nxd4
>>>					6.Q7xd4+ Rxd4 7.Qcxd4+ Qdxd4 8.Qbxd4+
>>>					Qdxd4 9.Qaxd4+ Qxd4 10.Qgxd4#
>>>0:27	2/5	Mate10	1265606		1.Nxf2+ Rxf2 2.Nxb2+ Qxb2 3.Bxe2+
>>>					Qxe2+ 4.Qxe2+ Rxe2 5.Qfxd4+ Nxd4
>>>					6.Q7xd4+ Rxd4 7.Qcxd4+ Qdxd4 8.Qbxd4+
>>>					Qdxd4 9.Qaxd4+ Qxd4 10.Qgxd4#
>>>2:05	2/6	Mate10	5480662		1.Nxf2+ Rxf2 2.Nxb2+ Qxb2 3.Bxe2+
>>>					Qxe2+ 4.Qxe2+ Rxe2 5.Qfxd4+ Nxd4
>>>					6.Q7xd4+ Rxd4 7.Qcxd4+ Qdxd4 8.Qbxd4+
>>>					Qdxd4 9.Qgxd4+ Qxd4 10.Q7xd4#
>>>
>>>jm
>>
>>Chest confirms this to be a shortest possible mate (congrats to CM!), and also
>>the symmetric solution found by Sjeng (no surprise from the initial nearly
>>symetric position) in 2.3 hours on K7/600 with 350 MB hash:
>>
>>PV: Nxf2+ Rxf2 Nxb2+ Qxb2 Bxe2+ Qxe2+ Qxe2+ Rxe2 Qgxd4+ Qdxd4 Qfxd4+ Qbxd4
>>Qgxd4+ Qxd4 Qcxd4+ Rxd4 Qbxd4+ Nxd4 Q7xd4#
>>PV: Nxb2+ Qxb2 Nxf2+ Rxf2 Bxe2+ Qxe2+ Qxe2+ Rxe2 Qgxd4+ Qdxd4 Qfxd4+ Qbxd4
>>Qgxd4+ Qxd4 Qcxd4+ Rxd4 Qbxd4+ Nxd4 Q7xd4#
>>
>> Nxb2+ Qxb2 Nxf2+ Rxf2  Bxe2+ Qxe2+ Qxe2+  ...+6
>>                              Rxe2  Qexd4+ ...+6
>>                  Qxf2  Bxe2+ Qxe2+ Qxe2#
>>       Rxb2 Nxf2+ Rxf2  Bxe2+ Qxe2+ Qxe2+  ...+5
>>                              Rxe2  Qexd4+ ...+5
>>                  Qxf2  Bxe2+ Qxe2+ Qxe2#
>> Nxf2+ Rxf2 Nxb2+ Qxb2  Bxe2+ Qxe2+ Qxe2+  ...+6
>>                              Rxe2  Qexd4+ ...+6
>>                  Rxb2  Bxe2+ Qxe2+ Qxe2+  ...+5
>>                              Rxe2  Qexd4+ ...+5
>>       Qxf2 Bxe2+ Qxe2+ Qxe2#
>>
>>Chest's timing info shows an EBF around 6:
>>#  3      0.03s [  3.00]        1kN [ 12.62]  0.95         93-         0
>>#  4      0.15s [  5.00]       10kN [  8.13]  1.12        308-         0
>>#  5      1.20s [  8.00]       82kN [  7.82]  1.29       2813-         0
>>#  6      7.29s [  6.08]      483kN [  5.91]  1.79      15452-         0
>>#  7     44.32s [  6.08]     3024kN [  6.26]  2.40      99438-         0
>>#  8    248.53s [  5.61]    17283kN [  5.72]  3.13     597059-         0
>>#  9   1155.17s [  4.65]    83584kN [  4.84]  4.42    2797265-       535
>># 10   8042.53s [  6.96]   526938kN [  6.30]  6.01   25077142-  16329266
>
>Thanks, Heiner!
>
>Now I see that this is mate in 10. Probably, will put later mine to search mate
>in 10 in the night. It could take hours, after my time for 9 moves. I see that
>Chess Master found better by searching shorter mate in 10. Mine found by
>selective only mate in 11 in 2 seconds. Will look if few additional versions for
>selective will take this position in 10. This position was created last Summer
>and then I had few selective versions for search.


Just tried this position with new selective version and, bingo! Mate in 10 moves
in zero seconds.

Actually, before my selective version had three "mixed" searches and now I have
5 of them. Mixed search - search start with brute force and later goes by
selective. Selective - attacking side look only promissing moves and also number
of responses is limited. In my version last Summer my three version of "mixed"
search had 1, 2 and 3 moves deep brute force before selective. Now this brute
force search goes as far as 5 moves. When I asked to search in 5, "mixed" search
response was instant. I not even tried to search for 4 moves brute force search
in mixed.

It could be that in your future selective search (if you will be interested one
day to install it) you can try something like I did.

Cheers,
Leonid.





>My brute force, for my usual conditions:
>
>Moves        Time        Branching factor         NPS.
>
>4            0.93 sec                             112k
>                         7.35
>5            6.86 sec                             110k
>                         8.23
>6            56.53 sec                            98k
>                         7.2
>7            6 min 47 sec                         106k
>                         6.62
>8            44 min 57                            109k
>                         6.03
>9            4 hours 31 min 27 sec
>
>Cheers,
>Leonid.
>
>
>>Cheers,
>>Heiner



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.