Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: TestPosition

Author: Michel Langeveld

Date: 14:32:59 11/13/01

Go up one level in this thread


On November 13, 2001 at 15:18:12, Slater Wold wrote:

>On November 13, 2001 at 13:18:55, Robert Hyatt wrote:
>
>>On November 13, 2001 at 12:00:47, Michel Langeveld wrote:
>>
>>>How long does your program need to see that Nxf1 is deadly lost:
>>>
>>>[D] 2r3k1/1b3p2/p4q2/2bp1NpQ/Pr6/3BP2P/3n1PPB/1R3RK1 b - - 0 30
>>
>>
>>Should take crafty this long:
>>
>>                9    13.25  -1.53   1. ... Nxf1 2. Rxb4 Bxb4 3. Be5 Ng3
>>                                    4. Nh6+ Kf8 5. Bxg3 Rc1+ <HT>
>>                9->  14.45  -1.53   1. ... Nxf1 2. Rxb4 Bxb4 3. Be5 Ng3
>>                                    4. Nh6+ Kf8 5. Bxg3 Rc1+ <HT>
>>               10    16.55     --   1. ... Nxf1
>>               10    37.19  -0.47   1. ... Nxf1 2. Be5 Rxb1 3. Bxb1 Ng3
>>                                    4. fxg3 Bxe3+ 5. Kh2 Bg1+ 6. Kxg1 Qxe5
>>                                    7. Ne7+ Kg7 8. Nxc8 Bxc8
>>               10->  43.73  -0.47   1. ... Nxf1 2. Be5 Rxb1 3. Bxb1 Ng3
>>                                    4. fxg3 Bxe3+ 5. Kh2 Bg1+ 6. Kxg1 Qxe5
>>                                    7. Ne7+ Kg7 8. Nxc8 Bxc8
>>               11    44.67     --   1. ... Nxf1
>>               11    57.23   3.07   1. ... Nxf1 2. Be5 Rxb1 3. Bxb1 Ng3
>>                                    4. fxg3 Bxe3+ 5. Kh2 Rc1 6. Ne7+ Kf8
>>                                    7. Ng6+ fxg6 8. Qh6+ Ke7 9. Bxf6+ Kxf6
>>                                    10. Bxg6
>>
>>
>>IE it fails low at 44 seconds, at 57 seconds it sees white is winning...
>>
>>quad 700
>
>I am guessing this game was played on an 800mhz computer.  (Michel's computer.)
>Which means it would take about 4 minutes for it to get this correct, on his
>machine.
>
>What does Crafty do when its time limit is up, and it's currently failing low on
>a move?  I am guessing it extends the time, but what if the move time is
>exhausted and it's still failing low?
>
>The reason I ask, is because if I am correct, and this was a game that was
>played on Michel's computer, it was more than likely failing low when it made
>the move.

Yes this was played on my computer... I was in a match against another strong
engine. Note that use dual 800Mhz for these games and not a single 800 system. I
thought Kf8 was better. But according to Gian-Carlo's post it seems to be that
also Kf8 is already lost. So the error of this game must be before this move.

Other engines in comparison to Crafty see very quick that this position is lost.
In that way they don't get this on board with black.

Michel






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.