Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: How high would the Rating of Belle realy be

Author: Robert Hyatt

Date: 20:52:03 09/23/01

Go up one level in this thread


On September 23, 2001 at 17:14:20, Marc van Hal wrote:

>On September 23, 2001 at 10:28:26, Robert Hyatt wrote:
>
>>On September 23, 2001 at 10:09:45, Tony Werten wrote:
>>
>>>On September 23, 2001 at 09:41:01, Robert Hyatt wrote:
>>>
>>>>On September 23, 2001 at 05:38:29, Tony Werten wrote:
>>>>
>>>>>On September 22, 2001 at 13:06:54, Marc van Hal wrote:
>>>>>
>>>>>>Acording to the programmer it would have a rating of 2345
>>>>>>After which Donner told them they realy did do everything to mislead the people
>>>>>>
>>>>>>To my point of vieuw he was right
>>>>>>
>>>>>>Marc van Hal
>>>>>
>>>>>It's rating was considerably lower than that of Donner's niece when she was 4.
>>>>>
>>>>>Tony
>>>>
>>>>
>>>>Who is that?  A female IM/GM at that age???
>>>
>>>Donner was not very impressed by Belle. He played a match against it and claimed
>>>afterwards:" It was terrible, my 4 year old niece plays better chess"
>>>
>>>Donner was famous in Holland for his rudeness (well, also for his chess of
>>>course). If you had never been insulted by Donner, you didn't count.
>>>
>>>His nickname was "Iron Hein". Hein except for Donners first name, also being the
>>>name for Death in dutch ( Skinny Hein).
>>>
>>>Tony
>>
>>
>>Did he win?  Or did he get beat by a machine that could beat his 4-year old
>>niece?  :)
>
>Ehm he said that his Nice coul;d play better chess then the machine
>
>Donner won
>
>The programmer was Ken Thompson
>
>The game was:



Actually, Ken Thompson and Joe Condon _built_ the thing.  Belle was not
software at all, except for the first 2 plies.  The rest was a huge PLA-based
hardware black box that did the search and evaluation at a speed that made
us all drool in 1980.

It really was veryt strong.  Probably around 2350 USCF or so.  Not sure how
that would translate to FIDE in the 1983-1984 time frame.




>
>Donner,Belle Delft/Muray Hill 1982
>1.d4,d5
>2.c4,c6
>3.Nf3,c6
>4.Nc3,e6
>5.Bg5,Nbd7
>6.e3,Qa5
>7.Nd2,Bb4
>8.Qc2,0-0
>9.Be2,dxc4
>10.Bxf6,Nxf6
>11.Nxc4,Qc7
>12.0-0,Rd8
>13.a3,Be7
>14.b4,b6
>15.Rfc1,Bb7
>16.Bf3,a6
>17.Rab1,Nd7
>18.a4,Rab8
>19.g3?,Kh8? (Ken Thompson said that the program did see the move 19..Bxb4 after
>for example 20.Rxb4,c5 21.dxc5,Bxf3 22.cxb6,Nxb6 23Rxb6,Qxb6 with a plus for
>Black)
> but for some reason gave it a lower evaluation then the mainline
>20.Bg2,f5??
>21.Ne2,Re8
>22.Nf4,Bg5
>23.Nf4,Bg5
>24.Nd3,Bf6
>25.f4,Rec8
>26.Nd2,Ba8
>27.Nb3,Qa7
>28.Qe2,Rc7
>29.Nbc5,Nf8
>30.Rc2,Re8
>31.Rbc1,Kg8
>32.Bf3,Rd8
>33.g4,Rdc8
>34.Rc3,fxg4
>35.Bxg4,Re7
>36.Qa2,Rce8
>37.f5,Rd8
>38.Nf4,Bxd4
>39.exd4,Rxd4
>40.f6,Rxf4
>41.fxe7,Rg4+
>42Rg3,Rxg3
>43.hxg3,Qxe7
>44.Qd2,h6
>45.Rd1,Qe8
>46.Qd6,e5
>47.Qc7,Qh5
>48.Rf1,Qe8
>49.Qd6,e4
>50.Rf4,e3
>51.Re4,Qc8
>52.Rxe3,h5
>53.Re7,Qf5
>54.Qf4,Qg6
>55.Ra7,Qe8
>56.Qd4,Qg6
>57. Here the game was stopped there was decided no longer to use the trans
>atlantic conection
>and Olafson Vice president of the Fide decided and concluded that Donner had won
>
>Ofcourse there must be serious doubt his Nice could make it this long



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.