Author: Igor Gorelikov
Date: 06:50:56 10/24/03
Go up one level in this thread
On October 24, 2003 at 09:12:45, Drexel,Michael wrote: >On October 24, 2003 at 07:58:40, Igor Gorelikov wrote: > >>On October 24, 2003 at 07:41:26, Jonas Bylund wrote: >> >>>On October 24, 2003 at 07:32:52, Igor Gorelikov wrote: >>> >>>>On October 24, 2003 at 06:58:32, Jonas Bylund wrote: >>>> >>>>>On October 24, 2003 at 06:44:22, Igor Gorelikov wrote: >>>>> >>>>>>On October 24, 2003 at 06:16:59, Jonas Bylund wrote: >>>>>> >>>>>>>On October 24, 2003 at 05:47:18, Igor Gorelikov wrote: >>>>>>> >>>>>>>>I ran two final rounds of the tournament. >>>>>>>> >>>>>>>>Time control: 15'+3" >>>>>>>>Celeron 1.7GHz 128MB; hash settings is up to 38MB in sum for each engine; >>>>>>>>resign is set to 6; main 5-men Nalimov's TBs. >>>>>>>>Galis WBTM 0.5.0 is used to arrange the event. >>>>>>>> >>>>>>>>Tough Guys >>>>>>>>Celeron 1.7GHz 128MB, 2003.10.23 - 2003.10.24 >>>>>>>> Score 1 2 3 4 >>>>>>>>----------------------------------------------------- >>>>>>>> 1: SOS.3 for Arena 9.0 / 12 XXXX 110= =110 1111 >>>>>>>> 2: Crafty 16.19 7.0 / 12 001= XXXX =101 =11= >>>>>>>> 3: Ruffian 1.0.1 6.5 / 12 =001 =010 XXXX 111= >>>>>>>> 4: Chinito 3.1 1.5 / 12 0000 =00= 000= XXXX >>>>>>>>----------------------------------------------------- >>>>>>>>24 games: +12 =6 -6 >>>>>>>> >>>>>>>>Here is the link to download all the games: >>>>>>>>http://www.digichess.gr/infiniteloop/games/S0001r.zip >>>>>>>> >>>>>>>>Best regards, >>>>>>>>Igor Gorelikov >>>>>>> >>>>>>>There is obviously something wrong with your setup, try rerunning this >>>>>>>tournament using Arena. >>>>> >>>>>I will, although i can't get Crafty 16.19 up and running in Arena, could you >>>>>post the crafty.rc that you use with 16.19? >>>>> >>>>>I have some engagements this weekend, so i will run the same tournament setup >>>>>using a different (either crafty or another engine at par) engine ver. >>>>> >>>>>>There is obviously something wrong with your words. :-( >>>>>>You've better try to run a similar event with the same participants instead of >>>>>>that. >>>>> >>>>>I see nothing wrong with my choice of words, i have run several games between >>>>>El_Chinito 3.1, Ruffian etc and see so far that chinito is at par with Ruffian, >>>>>who usually beats up on SOS and SOS usually beats up on Crafty :) >>>>> >>>>>The point is, Chinito 3.1 only scoring 3 draws in that company is VERY unlikely, >>>>>when i say i suspect something is wrong with your setup, i don't mean it as >>>>>something peronal, just we are only humans and make mistakes once in a while and >>>>>that might be the case here. >>>>> >>>>>Regards >>>>>Jonas >>>> >>>>OK, now I know what you mean. ;-) >>>>I think any thorough testing needs a variaty of people and hardware. >>>> >>>>Best regards, >>>>Igor Gorelikov >>> >>>I have started my own test with the same setup as you, i just replaced Crafty >>>16.19 with Yace Paderborn. If needed though, i will run it again with Crafty >>>16.19 given i can make it run ubder Arena. >>> >>>0 of 72 games played >>>Tournament start: 2003.10.24, 12:44:16 >>>Latest update: 2003.10.24, 12:46:41 >>>Site/ Country: Aarhus, Denmark >>>Level: Blitz 15/3 >>>Hardware: AMD Athlon(tm) Processor 1210 MHz (1.5 Gb ram) >>>Operating system: Microsoft Windows XP Build 2600 >>> >>>Participants: >>> >>>ElChinito_31 >>>Ruffian 1.0.1 >>>Yace Paderborn >>>SOS.3 for Arena >>> >>>Regards >>>Jonas >> >>It will be better to include Crafty 16.19 for proper comparison. >> >>IMPO, Yace Paderborn is weaker than Crafty 16.19 (about 100 ELO point) on fast >>hardware. >> >>regards, >>Igor > >Yes certainly, according to your personal opinion Crafty 16.19 is roughly as >strong as Junior 7 or Hiarcs 8. > >I suggest you play tournament time control matches between Crafty 16.19 and >those engines and watch Crafty 16.19 to score round about 33%. > >Those '100 point stronger claims/opinions' seem to accumulate recently. >Not long ago I read a posting on the occasion of the Ruffian 2.0 announcement >from Quisinsky (Arena Webmaster): > >"(I believe Ruffian is one of TOP 3 chess programs in the World). The Leiden >version 10.10.03 is more as 100 ELO stronger as Ruffian 1.05". > >Michael Sorry, but I'm testing only free available engines. My POOR opinion is based on my tests, in particular on the following ratings which are still far from perfection. They can be found at my site: http://www.digichess.gr/infiniteloop/home.php At time control 80m+3s Change Place Program Elo + - Games Score Av.Op. Draws in Places 0 4 Ruffian 1.0.1 : 2733 75 128 48 76.0 % 2532 18.8 % 0 5 Crafty 16.19 : 2673 117 120 30 61.7 % 2590 23.3 % -3 20 Yace 0.99.56 : 2505 132 92 29 53.4 % 2481 37.9 % At time control 15m+3s Change Place Program Elo + - Games Score Av.Op. Draws in Places -2 5 Yace 0.99.56 : 2551 58 68 105 61.0 % 2474 17.1 % +10 6 Yace Paderborn : 2513 87 88 54 58.3 % 2455 20.4 % At time control 5m+3s Change Place Program Elo + - Games Score Av.Op. Draws in Places -2 5 Yace 0.99.56 : 2541 58 68 105 61.0 % 2463 17.1 % +4 16 Yace Paderborn : 2460 105 90 45 52.2 % 2444 20.0 % If you have opposite results, feel free to publish them and dispose my poor and humble tests. Best regards, Igor Gorelikov
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.