Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Testposition - The Royal Miracle

Author: José Carlos

Date: 10:47:32 02/28/01

Go up one level in this thread


On February 28, 2001 at 07:43:23, Ed Schröder wrote:

>On February 28, 2001 at 06:06:21, Ricardo Gibert wrote:
>
>>On February 28, 2001 at 05:11:30, Sune Larsson wrote:
>>
>>>
>>>  [D]8/1p3p1p/p4P1k/P4P1p/6pP/6Pq/5P2/5K2 w - - 0 1
>>>
>>>
>>>  True story this time. The above position happened in a free game between
>>>  Dr J. Perlis and N.N., many years ago, in Germany I think. The good doctor
>>>  was a generous man and played without a queen from the start. So it's not
>>>  strange that he is missing one in the above position. And since we are
>>>  careful, scientific people we shall have a closer look at it with Fritz.
>>>
>>>  Okey, white is a queen down, is standing in check right now, Fritz gives
>>>  3 ways to go:
>>>
>>>  1.Ke2 -19.44
>>>  1.Ke1 -19.75
>>>  1.Kg1 -19.12
>>>
>>>  Guess it's time for a new game... In fact it is - because after
>>>  1.Kg1! *black* is mated in 6 moves.
>>>
>>>  Sune
>>
>>I tried this position on Gambit Tiger and got results I don't understand:
>>
>>00:00:00.2	-14.36	19	124852	Kg1
>>00:00:00.8	-13.46	20	209791	Kg1
>>00:00:00.6	-13.46	21	220376	Kg1
>>00:00:00.7	-14.36	21	275872	Kg1
>>00:00:04.8	-13.46	22	1611011	Kg1
>>00:00:04.4	-13.46	23	1624598	Kg1
>>00:00:05.8	-14.36	23	1937717	Kg1
>>00:00:53.9	-13.46	24	21645539	Kg1
>>00:00:53.3	-13.46	25	21750433	Kg1
>>00:00:54.1	-14.36	25	21949656	Kg1
>>00:01:02.7	-13.46	26	24990033	Kg1
>>00:01:03.4	-13.46	27	25264260	Kg1
>>00:01:06.7	-14.36	27	26269213	Kg1
>>
>>It likes Black so it does not seem to see the mate. At this point I forced it to
>>play Kg1:
>>
>>00:00:00.6	-Mate in 6	17	120953	b5 axb6ep a5 b7 Qxh4 gxh4 g3 b8Q gxf2+ Kxf2 a4
>>Qf8#
>>00:00:00.8	-Mate in 6	17	178642	b5 axb6ep a5 b7 Qxh4 gxh4 g3 b8Q gxf2+ Kxf2 a4
>>Qf8#
>>00:00:01.6	-Mate in 6	18	433243	b5 axb6ep a5 b7 Qxh4 gxh4 g3 b8Q gxf2+ Kxf2 a4
>>Qf8#
>>00:00:04.5	-Mate in 6	19	1180581	b5 axb6ep a5 b7 Qxh4 gxh4 g3 b8Q gxf2+ Kxf2 a4
>>Qf8#
>>00:00:12.0	-Mate in 6	20	3604732	b5 axb6ep a5 b7 Qxh4 gxh4 g3 b8Q gxf2+ Kxf2 a4
>>Qf8#
>>00:00:42.5	-Mate in 6	21	12558570	b5 axb6ep a5 b7 Qxh4 gxh4 g3 b8Q gxf2+ Kxf2 a4
>>Qf8#
>>
>>It sees the mate just fine, so now I made it take back Kg1:
>>
>>00:00:00.9	 Conversion in -10043	32	3	Kg1
>>00:00:01.9	 Conversion in -9998	32		Kg1
>>00:00:02.2	 Conversion in -9953	33		Kg1
>>00:00:02.9	 Conversion in -9908	34		Kg1
>>00:00:02.9	 Conversion in -9863	35		Kg1
>>00:00:02.2	 Conversion in -9818	36		Kg1
>>00:00:02.6	 Conversion in -9773	37		Kg1
>>00:00:03.1	 Conversion in -9728	38		Kg1
>>00:00:04.1	 Conversion in -9683	39		Kg1
>>00:00:05.7	 Conversion in -9638	40		Kg1
>>00:00:07.4	 Conversion in -9593	41		Kg1
>>00:00:08.4	 Conversion in -9548	42		Kg1
>>00:00:14.7	 Conversion in -9503	43		Kg1
>>00:00:24.2	 Conversion in -9458	44		Kg1
>>00:00:34.1	 Conversion in -9413	45		Kg1
>>00:00:45.6	 Conversion in -9368	46		Kg1
>>00:00:56.8	 Conversion in -9323	47		Kg1
>>00:01:44.0	 Conversion in -9278	48		Kg1
>>00:03:11.9	 Conversion in -9233	49		Kg1
>>
>>What's going on? What does it mean by, "Conversion in -nnnn"? Can it really get
>>to Depth 49?
>
>
>It is a bug which will be fixed in the upcoming update. In this specific
>case the mate value is not converted in the right way.
>
>Ed

  And the first search not seeing the mate looks like a null-move issue, isn't
it?

  José C.



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.