Author: Jim Monaghan
Date: 23:35:51 06/27/04
Go up one level in this thread
>>When I was seperating out the positions by ply depth, I had inadvertently left >>in those 6. Obviously, if a position is solvable at ply 5, it's still solvable >>at ply 7. Second duplicates are eliminated and positions re-numbered. >> >>Thanks, >>Jim >> >> >>Chess IQ Test, Version 3 >>by Jim Monaghan >> >> Tactical IQ3 Test for Chess Programs >> >>This test is based on a test suite from the master section of Livshits' book >>"Test Your Chess IQ". It has been debugged and improved. (Refer to the >>previous IQ page for detailed explanations.) The positions are carefully >>balanced with medium to hard examples. The recent changes: >> >> I dropped the "easy" positions solvable in 4 ply or less > >Are you sure? Hi Uri, I used Yace 0.99.56 (or 0.99.58 I think) and Shredder 6.0 when stepping the test. The 4 ply or less positions were solved "instantly" by almost any reasonably strong program. So I thought they were just taking up space and using time. Am I sure? No ... >Movei solved position number 2 at depth 3 and it is not only the latest private >version but also the public version that everybody can try. > >4r1k1/1pq2p1p/2p2npb/2N5/1PPQn1P1/7P/6B1/B2R2K1 b - - bm Ng3 > > >> I dropped the "hard" positions requiring 12 ply or more > >Are you sure? With regards to the 12 ply or more positions, they are very complicated and take a long time to resolve, say a half to one hour per position for a top commercial program. I'm not even sure if the solution(s) are correct. I didn't see the point of including them when each position gets only 10 seconds. That extra ply or two takes "forever" ... Am I sure? Again, no ... >latest movei needs more than 12 plies to solve the last position in your test >and the same is for the public version. > >I think also that it is better to use times and not plies and 12 plies in a >simple position may be harder to find than 11 plies in a complicated position. > >using a specific problem to decide what is hard and what is easy is also wrong >because what is hard for A may be not hard for B. You're right, the way the test is stepped is biased towards Yace and Shredder. Generally, though the positions go from easier to harder ... Cheers, Jim >Uri
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.