Author: José Carlos
Date: 05:30:34 08/16/00
Go up one level in this thread
On August 16, 2000 at 07:34:11, Tony Werten wrote: >On August 15, 2000 at 17:31:01, José Carlos wrote: > >>On August 14, 2000 at 16:05:34, Will Singleton wrote: >> >>> **************************************************** >>> ** ** >>> ** The Green List ** >>> ** ** >>> ** Developmental Chess Programs on ICC ** >>> ** ** >>> **************************************************** >>> >>> >>>8/14/2000 16:03 EDT >>>------------------------------------------------------------------------- >>> Active Accounts -- Blitz >>>------------------------------------------------------------------------- >>>ACCOUNT BLITZ R(mhz) AVG CHANGE WIN% GP/WK TGP ACTIVE >>> >>>JZInsomniac 2786 2805 | 2795 -3 | 62% 28 412 100 >>>PostModernist 2755 2785 | 2664 +30 | 67% 222 13013 410 >>>counterplay 2735 2776 | 2729 +1 | 73% 66 410 43 >>>Falke 2711 2696 | 2664 +15 | 74% 6 130 7 >>>MissSilicon 2662 2692 | 2700 -13 | 61% 35 2082 405 >>>QueenMaxima 2623 2653 | 2533 +29 | 79% 341 9422 157 >>>Hossa 2619 2652 | 2507 +37 | 57% 262 15378 214 >>>Amy 2616 2635 | 2598 +5 | 63% 27 196 7 >>>ArasanX 2631 2609 | 2598 +11 | 63% 31 1838 405 >>>Grok 2548 2578 | 2541 +2 | 57% 3 179 35 >>>lambchop 2498 2548 | 2525 -9 | 53% 3 140 0 >>>inmichess 2441 2471 | 2423 +6 | 39% 4 262 21 >>>Tinker 2478 2458 | 2492 -5 | 69% 112 3447 35 >>>Dorky 2422 2452 | 2353 +22 | 67% 269 548 14 >>>kerrigan 2422 2393 | 2451 -10 | 40% 0 30 35 >>>stormx 2205 2205 | 2188 +5 | 75% 20 598 7 >>>LoadedGun 2200 2171 | 2292 -31 | 78% 51 475 64 >> >> I have a bug I cannot understand. When compiling Averno in Debug mode, it >>works slow, but correctly in every computer. When compiling it in Release mode, >>with speed optimization, it works fine on PII's but makes strange moves on >>Athlon's. As Roger's account is running an Athlon-800, my program is making >>horrible moves, losing game after game. >> Any idea about what is happening? Please... help me! > >Sounds to me like an uninitialized variable. In debug mode they are set to zero, >in release mode they just get any value. Take a testposition and look if you >need the same number of nodes to solve it in debug as you need in release. > >Tony Thanks for the idea. I'll check it as soon as I get home (I'm at work right now), but if I remember right, the number of nodes was different. But, what most confuses me is that that behaviour does not happen on the PII. José C.
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.