Author: Joshua Lee
Date: 12:07:19 03/20/01
Go up one level in this thread
On March 19, 2001 at 04:42:43, Joshua Lee wrote: >Awhile back i said i was going to run all the test sets with fritz well i have >quite a few and some that aren't in epd format yet so i have alot to be done. >I have to remove doubles from 1 file of about 6,000 positions then convert many >that are in something pre san or epd.... over a thousand of those. >I also have another file of about 1200 that i am working on now i should be done >with that in a day or two and then i can take those unsolved and put them in the >bigger file and continue removing doubles. i wish there were a faster way then >just searching for each position manually but chessbase lite deletes quite a few >that aren't doubles!!!! if you have any ideas as to what i can do to remove >doubles faster by position i would love to hear it. > > >you are probably thinking i will never finish well i neglected to mention that i >have run quite a few of these that weren't solved in 27, 9, 5, and 3 min. > >of the large file atleast 1000 are doubles but most likely more are because i >took all epd files from dan corbits site and put them together. i Didn't run >eco_pri or sec however because it contains so many positions that start at the >starting position i would be wasting my time but i may go through those manually >and check to see what i can do with them. These and the C.A.P positions are a >pain as there are so many and i don't see the relevence of them maybe someone >could fill me in (cap not eco the eco would test how well the program does in >the opening i suspect) > >Since fritz doesn't do everything i thought i would use a more "Positional" >program for certain positions so i use gandalf 4.32h which is impressive , >Phalanx and Hiarcs i can use if those aren't solved speaking of a position both >phalanx and gandalf solve this position in under a minute but hiarcs took 32 >from the VA test set ... Chaos Vs Chess 4 > >[D] rq2k2r/3n1ppp/p2bpnb1/8/Np1N4/1B3PP1/PP2Q2P/R1BR2K1 w kq - 0 1 > >Analysis by Gandalf 4.32h: > >1.Bd2 0-0 > -+ (-1.45) Depth: 2 00:00:00 >1.Nc6 > -+ (-1.43) Depth: 2 00:00:00 >1.Nc6 Qc7 > = (-0.14) Depth: 2 00:00:00 >1.Nc6 Qc7 2.Nd4 > ³ (-0.51) Depth: 3 00:00:00 >1.Bg5 > ³ (-0.49) Depth: 3 00:00:00 >1.Bg5 Ne5 2.Rac1 > = (-0.25) Depth: 3 00:00:00 >1.Bg5 0-0 2.Nxe6 Re8 > ³ (-0.46) Depth: 4 00:00:00 >1.Bxe6 > ³ (-0.44) Depth: 4 00:00:00 >1.Bxe6 0-0 2.Nc6 Qc7 > ³ (-0.36) Depth: 4 00:00:01 >1.Bxe6 0-0 2.Nc6 Qc7 3.Qc4 > = (-0.24) Depth: 5 00:00:01 25kN >1.Bxe6 0-0 2.Nc6 Qc7 3.Qc4 Rfe8 > ³ (-0.52) Depth: 6 00:00:01 79kN >1.Bxe6 0-0 2.Bc4 Ne5 3.Bg5 Nxc4 4.Bxf6 > ³ (-0.44) Depth: 7 00:00:03 300kN >1.Nxe6 > ³ (-0.43) Depth: 7 00:00:05 500kN >1.Nxe6 fxe6 2.Qxe6+ Be7 3.Bf4 Qa7+ 4.Be3 Qc7 5.Bb6 Qe5 > = (-0.25) Depth: 7 00:00:06 500kN >1.Nxe6 fxe6 2.Qxe6+ Be7 3.Bf4 Qa7+ 4.Be3 Qb7 5.Nc5 Nxc5 > = (-0.22) Depth: 8 00:00:12 1400kN >1.Nxe6 fxe6 2.Qxe6+ Be7 3.Bf4 Qb7 4.Bd6 Nf8 5.Qe5 Bf7 > = (-0.03) Depth: 9 00:00:30 3950kN > >(Lee, N.B.I 19.03.2001) > >so untill next time .... Bad News upon finding out there were way two many doubles i have to start over i am also annoyed that chessbase doesn't give you the option to check more than 5 ply many positions which aren't solved any time soon are mistakenly listed as solved. for example there are problems like this in MATS and the BK test and others so watch out. All i can do is repair screwy epd files for you but i need to know how and who to send them to. Thanks maybe i will try this again once i remove all duplicates
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.