Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: WBEC Ridderkerk edition 11 Premier Division

Author: Dann Corbit

Date: 14:16:11 10/21/05

Go up one level in this thread


On October 21, 2005 at 12:23:51, Sedat wrote:

>On October 21, 2005 at 05:40:50, Uri Blass wrote:
>
>>On October 21, 2005 at 04:45:52, Sedat wrote:
>>
>>>On October 20, 2005 at 18:17:09, Dann Corbit wrote:
>>>
>>>>On October 20, 2005 at 16:57:41, George Tsavdaris wrote:
>>>>
>>>>>On October 20, 2005 at 16:31:42, Vincent Lejeune wrote:
>>>>>
>>>>>>For those who don't go see this site regulary : http://wbec-ridderkerk.nl/ ->
>>>>>>Tournaments -> Premier Division
>>>>>
>>>>>Are there any such people? :-)
>>>>>
>>>>>>
>>>>>>Wacky race for first place !
>>>>>>
>>>>>> 1: Fruit WCCC'05         27.5 / 40
>>>>>> 2: Shredder 9-x64        25.0 / 40
>>>>>> 3: DeepSjeng X2-x64      25.0 / 40
>>>>>>
>>>>>
>>>>>All i can say is watch for The King 3.33. It's coming.........!
>>>>>Fruit and Shredder will obviously fight for the next 52 rounds with "The King
>>>>>3.33" be third.......
>>>>>
>>>>>ELOSTAT ratings:
>>>>>    Program                          Elo    +   -   Games   Score    Draws
>>>>>
>>>>>  1 Fruit WCCC'05                  : 2539   95  92    40    68.8 %   32.5 %
>>>>>  2 Shredder 9-x64                 : 2502   91  89    40    62.5 %   35.0 %
>>>>>  3 Zappa 1.1 x64                  : 2487   92  91    36    58.3 %   38.9 %
>>>>>  4 DeepSjeng X2-x64               : 2486   83  81    40    62.5 %   45.0 %
>>>>>  5 Thinker 5.0b x64               : 2465   97  96    40    57.5 %   25.0 %
>>>>>  6 SOS 5.1                        : 2461   89  88    37    58.1 %   40.5 %
>>>>>  7 Gandalf 6.01                   : 2445   78  77    40    57.5 %   50.0 %
>>>>>  8 Pro Deo 1.1                    : 2443   82  81    40    57.5 %   45.0 %
>>>>>  9 Spike 1.0a Mainz               : 2440   86  85    40    55.0 %   40.0 %
>>>>> 10 Pharaon 3.3                    : 2430   90  89    40    57.5 %   35.0 %
>>>>> 11 Naum 1.82                      : 2412   86  85    40    55.0 %   40.0 %
>>>>> 12 Ktulu 7.0a                     : 2410   89  89    40    52.5 %   35.0 %
>>>>> 13 The King 3.33                  : 2408   92  92    40    52.5 %   30.0 %
>>>>> 14 Scorpio 1.6a                   : 2404   89  89    40    50.0 %   35.0 %
>>>>> 15 Aristarch 4.50                 : 2401   83  84    40    46.2 %   42.5 %
>>>>> 16 SlowChess blitz WV2            : 2392   84  84    37    47.3 %   45.9 %
>>>>> 17 List 512                       : 2372   85  85    40    50.0 %   40.0 %
>>>>> 18 Crafty 19.20 x64BH             : 2357   92  93    40    42.5 %   30.0 %
>>>>> 19 WARP 0.65                      : 2350   98 100    40    38.8 %   22.5 %
>>>>> 20 Maestro 1.09uci                : 2337   90  91    40    46.2 %   32.5 %
>>>>> 21 SmarThink 0.18a-r165           : 2334   88  90    36    40.3 %   41.7 %
>>>>> 22 Ruffian 2.1.0                  : 2318   87  89    40    36.2 %   37.5 %
>>>>> 23 Tao 5.7 b07 uci                : 2220   93  98    40    26.2 %   32.5 %
>>>>> 24 Yace 0.99.87                   : 2193  101 107    40    21.2 %   27.5 %
>>>>>
>>>>>Too bad that Zappa 2.0 isn't participating and isn't available.......
>>>>
>>>>I am guessing that on a single CPU system, it will be somewhere in the middle of
>>>>this pack (probably top 1/3 or 1/4).  Since it is a SMP system, it will fare
>>>>much better on a multiple CPU system (like the Deep programs or Crafty or Deep
>>>>Sjeng)
>>>
>>>
>>>I think x64 bit versions have here a addvantage over 32 bit !!
>>
>>Fruit WCCC is 32 bit version
>>
>>Shredder Zappa Deep Sjeng Thinker are 64 bit versions.
>>I expect the 32 bit to win.
>>
>>Uri
>
>Hello Uri,
>
>Yes, you are right about which engines are 32 bit which 64 bit
>But as far as i know Leo uses O/S 64 bit and i'd like to mention again that on
>64 bit oparating system ,the engines which are 64 bit have addvantage than the
>32 bit versions !!

Compiling 64 bits will gain 10-20 Elo at most.  It's enough to be important, but
not enough to change most results.

When you bench 64 bit versions, they generally are about 1/3 faster at the
outside.

I guess that really good compilers will change this in the future.  Right now,
64 bit compilers are shiny-new and not as well devloped as the 32 bit
counterparts.




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.