Author: Dann Corbit
Date: 16:38:18 07/12/04
Go up one level in this thread
On July 12, 2004 at 18:35:28, Kurt Utzinger wrote: >On July 12, 2004 at 17:41:52, John Merlino wrote: > >>On July 12, 2004 at 17:27:03, Dann Corbit wrote: >> >>>On July 12, 2004 at 16:41:44, Kurt Utzinger wrote: >>> >>>>Detailed report and games at http://www.utzingerk.com/rk_2004_english.htm >>>> >>>> >>>>Final standing of the original tourney >>>>Endstand des Originalturniers >>>> Program Score % Av.Op. Elo + - >>>> >>>> 1 Shredder 8 : 78.5/140 56.1 2495 2537 54 43 >>>> 2 Junior 8 : 76.5/140 54.6 2496 2528 55 44 >>>> 3 Hiarcs 9 : 75.5/140 53.9 2497 2524 56 43 >>>> 4 Fritz 8 : 74.5/140 53.2 2497 2520 56 37 >>>> 5 CM9 T06 : 71.0/140 50.7 2500 2504 59 40 >>>> 6 Ruffian 2.1.0 : 67.5/140 48.2 2502 2489 40 58 >>>> 7 List 513 : 63.0/140 45.0 2504 2470 36 55 >>>> 8 Aristarch 4.41 : 53.5/140 38.2 2511 2427 50 50 >>>> >>>> (Generated with EloStat of Dr. Frank Schubert, start Elo 2500) >>>> >>>> >>>>Final standing with CM9 T05 settings >>>>Endstand Alternativ-Turnier mit CM9 T05 Setting >>>> Program Score % Av.Op. Elo + - >>>> >>>> 1 CM9 T05 : 88.5/140 63.2 2488 2582 49 47 >>>> 2 Shredder 8 : 78.0/140 55.7 2495 2535 54 42 >>>> 3 Junior 8 : 76.5/140 54.6 2496 2528 55 45 >>>> 4 Hiarcs 9 : 73.5/140 52.5 2498 2515 57 42 >>>> 5 Fritz 8 : 71.0/140 50.7 2499 2504 59 37 >>>> 6 Ruffian 2.1.0 : 62.0/140 44.3 2505 2465 42 54 >>>> 7 List 513 : 60.5/140 43.2 2506 2459 36 53 >>>> 8 Aristarch 4.41 : 50.0/140 35.7 2513 2411 52 48 >>> >>>You should definitely call Johan de Koning's attention to your CM9_05 settings. >>>This is a very amazing result, and with enough data to show a strong >>>correlation. >>> >>> Program Elo + - Games Score Av.Op. Draws >>>1 CM9 T05 : 2688 49 47 140 63.2 % 2594 36.4 % >>>2 Shredder 8 : 2636 38 30 280 55.9 % 2595 37.5 % >>>3 Junior 8 : 2628 39 31 280 54.6 % 2596 32.9 % >>>4 Hiarcs 9 : 2620 40 30 280 53.2 % 2597 35.0 % >>>5 Fritz 8 : 2612 40 26 280 52.0 % 2598 45.4 % >>>6 CM9 T06 : 2599 59 40 140 50.7 % 2594 37.1 % >>>7 Ruffian 2.1.0 : 2577 29 39 280 46.2 % 2603 38.9 % >>>8 List 513 : 2564 25 38 280 44.1 % 2605 48.9 % >>>9 Aristarch 4.41: 2519 36 34 280 37.0 % 2612 31.1 % >> >>Just noticed that the T05 settings are practically identical to the SKR >>settings. The one and only difference is that Attack/Defend value is -1 in SKR >>and 0 in T05 (in a range of -100 to +100, where negative numbers favor attack). >> >>jm > > > Hi John > You are right and we are indeed wondering why such > a small difference can have such a great influence. It could be an aberration, of course. However, with such a large nuber of games, I am expecting an I/O bug of some kind when reading that parameter.
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.