Author: Jeremiah Penery
Date: 08:52:22 05/18/05
Go up one level in this thread
On May 18, 2005 at 04:51:41, Renze Steenhuisen wrote:
>On May 18, 2005 at 03:24:42, Jeremiah Penery wrote:
>
>>On May 17, 2005 at 22:31:25, Eelco de Groot wrote:
>>
>>>Pro Deo 1.1 Q5T about Nolot #3, keymove is supposed to be Nxg5, most programs
>>>eventually play Bxg5. So does the Q5T, but only with +0.36
>>>[D]r2qk2r/ppp1b1pp/2n1p3/3pP1n1/3P2b1/2PB1NN1/PP4PP/R1BQK2R w KQkq - bm Nxg5; >I've studied the Nolot positions a lot, and I'm quite sure that position #3 is
>>correct with Nxg5. White has a crushing attack, and most of black's pieces are
>>paralyzed.
>>
>>This is output from a modified Crafty 19.19, with some form of singular
>>extensions among other things:
>>
>>White(1): analyze
>>Analyze Mode: type "exit" to terminate.
>> clearing hash tables
>> time surplus 0.00 time limit 22.50 (3:30)
>> depth time score variation (1)
>> ...
>> 13-> 2:32 0.70 1. Nxg5 Bxd1(S) 2. Nxe6 Qc8(S) 3. Nxg7+
>> Kf7(S) 4. O-O+ Kxg7(S) 5. Nf5+ Qxf5
>> 6. Bxf5 Bh5 7. b4 Bg6 8. Bg4 <HT>
>>
>>The score continues to rise at higher iterations.
>>
>> 17-> 27:17 1.24 1. Nxg5 Bxd1 2. Nxe6 Qb8(S) 3. Nxg7+
>> Kf7 4. N7f5 Ba4 5. O-O Ke8 6. b3(S)
>> Bxb3 7. axb3 <HT>
>
>
>Are we talkin about the same position here? This is my log I get for Crafty 19.3
>on a PIV 3 GHz and 768 MByte Hashtable:
>
>solution 1. Nxg5
> time surplus 0.00 time limit 1200:00 (1200:00)
> depth time score variation (1)
...
> 18 548:47 0.15 1. Qe2 g6 2. h4 Nf7 3. h5 Bh4 4. Qf2
> Bxg3 5. Qxg3 gxh5 6. Bf4 Kd7 7. Kd2
> h6 8. Nh2 Rg8 9. Nxg4 Rxg4 10. Qf2
> Qg8
> 18-> 935:27 0.15 1. Qe2 g6 2. h4 Nf7 3. h5 Bh4 4. Qf2
> Bxg3 5. Qxg3 gxh5 6. Bf4 Kd7 7. Kd2
> h6 8. Nh2 Rg8 9. Nxg4 Rxg4 10. Qf2
> Qg8
> time=1200:00 cpu=99% mat=0 n=72758561382 fh=91% nps=1010k
> ext-> chk=2131479195 cap=206017646 pp=19671222 1rep=98479684
>mate=3505586
> predicted=0 nodes=72758561382 evals=1817035425
> endgame tablebase-> probes=0 hits=0
>
>Can anyone tell me what the big change was to Crafty from 19.3 to 19.19 that
>causes this difference?
Nothing. The changes that caused my version to find this move were changes that
I made. IMO, Crafty sucks at finding any kind of king attack, so I helped it
out a bit. I won't claim that it's great or anything - in fact, it's very often
wrong - but in this position it is not wrong.
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.