Computer Chess Club Archives




Subject: Re: Twin miniatures [by Werner Speckmann,1955] Mate in 17 and Mate in 13

Author: Paul

Date: 06:15:46 01/27/01

Go up one level in this thread

On January 27, 2001 at 06:28:25, Heiner Marxen wrote:

>On January 26, 2001 at 13:14:27, Paul wrote:
>>On January 26, 2001 at 06:55:42, José Antônio Fabiano Mendes wrote:
>>>        Werner Speckmann,1955  Mate in 13
>>>        [D]8/8/4p3/2p5/p3Q3/K7/2r5/1k6 w
>>01:15 WM13 1. Qh1+ Rc1 2. Qh7+ Rc2 3. Qe4 e5 4. Qe1+ Rc1 5. Qd2 Rc2 6. Qd1+ Rc1
>>7. Qd3+ Rc2 8. Qe4 c4 9. Qe1+ Rc1 10. Qd2 Rc2 11. Qb4+ Ka1 12. Qe1+ Rc1 13.
>>Is this one oké, José? These solutions (correct or not) sure are nice!
>Hi Paul,
>Chest confirms the above PV in 21.5 min (K7/600, 350 MB hash).
>Currently it is chewing the other twin.  I expect it to need 8 hours for
>the complete mate in 17.  Aren't transposition tables great?
>The effective branching factor is reduced to below 2.


Yeah, TT's are marvellous! Just wish I had 350MB like you! No, I'm not envious
... well, maybe just a little :) Must see if I can plug in some more memory. A
better way would be to improve my algorithms of course ... yours sure seem to be
working ok!

I just stopped Pretz after it had been thinking for 17 hours on the Mate in 20
follow up from José Antônio, had quite a sleepless night coz of the noise right
next to my bed. But when someone asks so nicely, well ... it's hard to refuse
the challenge :)

It was stuck at ply 17, after a massive bf explosion (96MB hash) :

10 00:00:10
11 00:00:18
12 00:00:49
13 00:02:03
14 00:03:48
15 00:08:52
16 00:21:18
17 16:59:56

Think I'll stop for the moment trying to solve these difficult ones, have to
find my bug(s) first; can't debug/compile/link when Pretz is thinking. But why
are these d@#$rned little critters sooo hard to find, sigh ... ;)


This page took 0.02 seconds to execute

Last modified: Thu, 07 Jul 11 08:48:38 -0700

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