Author: Don Dailey
Date: 04:54:07 07/26/98
Go up one level in this thread
On July 25, 1998 at 13:58:32, Howard Exner wrote:
>On July 25, 1998 at 11:01:53, Don Dailey wrote:
>
>>On July 25, 1998 at 06:47:27, Leon Stancliff wrote:
>>
>>>On July 25, 1998 at 02:00:59, Georg Langrath wrote:
>>>
>>>>I think that the time has come when commercial computers must show if they have
>>>>a chance in slow matches. The day a computer wins 8 matches with time control
>>>>40 moves i 2 hours against a GM, then we have a new era.
>>>>
>>>>Georg
>>>
>>>I think it is generally accepted that the top computers are still below the
>>>level of Kasparov, Karpov and Anand in 40/2 games. The real question is whether
>>>or not these top programs can play in the 2500-2600 Elo range. If so, they can
>>>be considered Grandmaster level.
>>> Your suggestion is valid. However, we need to find one, two or three
>>>Grandmasters who have about a 2600 Elo rating and are willing to play Rebel 10
>>>on the same hardware it used against Anand. Is there a third party sponsor out
>>>there who would financially support three eight game matches at 40/2?
>>> I have little doubt that Rebel 10 would be willing. I have considerable more
>>>doubt that there will be three GM in the 2600 range who will show an interest.
>>>
>>>Come on!! Prove me wrong!!
>>
>>This is an excellent idea. But first we should play a match with
>>some players near Rebels level before moving up. Here is my
>>suggestion:
>>
>>Bob Hyatt puts an upper bound of 2400 on the strength of the very
>>best micro's. If we play a match with some humans, then we target
>>players close to Rebel's strength to determine if there is any
>>need to "bother" the good players.
>>
>>The idea, is to get players as CLOSE as possible to the
>>range of strength we think Rebel is.
>
>Here's a small sample of A 40/2 Standard Rebel 9 tournament
>
>4.5 - 0.5 against an average field of 2312(Fide)
>
>Dominican Republic
>Rebel 9 played 5 Fide Rated opponents at 40/2 then all/1 hr. on a P-225
>Eddy De Los Santos (2190) 1-0
>Jose M. Dominguez (FM 2305) .5-.5
>Gustavo Hernandez (2320) 1-0
>Nelson Pinal (IM 2330) 1-0
>Ramon Mateo (IM 2415) 1-0
>
>TPR =Rc+400(W-L)/N
>TPR=2312+400(4)/5
>TPR=2632
>
>Note: this tournament was 11 rounds. Rebel defeated Hiarcs and drew CM5000
>in this tournament. It also defeated 4 other participants who do not have a
>fide rating, for a first place score of 10/11.
>
>Alone this tournament lacks the data pool size for the usual proclamation,
>"It's rating must therefore be 2632". However if you toss into the mix the
>24 games from Aegon, even including the 1994 Rebel 6 result on a measly 486,
>the result is still a performance rating of over 2500. Now we have 2 more
>games against a 2795 rated player. That's over the min 25 we use for ourselves.
>Deep Thought was given this 25 game limit years ago to calculate its plus 2500
>rating.
You get no argument from me. I don't claim to know the exact level
of todays top micro's but my "best guess" (for what it's worth) is
2500. I posted before that my absolute low bound is 2400, but I
feel they are very unlikely to be that weak. I guess if I take 2500
to be the midpoint I would say my UPPER bound is more like the 2600
rating you say it seems to be playing at. Part of this discussion
was spawned by the GM norm discussions however so I'm weighting
my responses accordingly. In other words all my guesses (and that's
all they are) are based on the following conditions:
1) Of course 40/2 time control, serious tourment conditions.
2) General knowledge and interest in whatever the target program
is. In other words GM's are analysing it's games and have the
general expectation of having to play against it sooner or
later.
On top of these 2 conditions, I would add that if long matches are
being played, there will be a modest lowering of the computers
expected win percentage. I don't think the effect is huge however
but I'm sure this is highly debatable so I won't go into it!
>Don, your Cilkchess was at Aegon in 1997. You must have been pleased that it
>drew its game with GM Seirawan. His comments after were also interesting. He
>expressed that he was bothered with his game because he missed a better
>continuation. To me it sounded like he took the game very seriously. My last
>sentence may be redundant - is there such a thing as a chess player not
>playing seriously to win each and every game they play? :)
I was very pleased about this but not the last round loss. In general
I am not pleased with Cilkchess and feel like there are some serious
problems. If you get a chance look at how uneven our play is in
the last two Dutch Championships, probably half the games we played
should have been lost. There are clearly some problems we have to
solve before the thing is any good.
>>You don't want Rebel
>>playing up in every game, or visa versa. It requires a larger
>>sample of games to accurately rate such matches since a single
>>draw can change your performance rating enormously.
>>
>>So I suggest a range of 2250-2450 for our opponents. This is
>>100 points on either side of 2350. I pick 2350 as center
>>because if we take 2400 as upper bound, then 2350 might be
>>considered a reasonble "best guess" estimate. Bob, if
>>you are listening, does 2350 seem like a reasonable estimate
>>of Rebel's chess strength?
>>
>>Now if Rebel does poorly in such a match, it will be an
>>indication that 2350 is too optimistic a rating for Rebel,
>>but more importantly, it will tell us we need to wait before
>>challenging a group of grandmasters, unless we just want to
>>watch Rebel get embarassed. On the other hand, if Rebel
>>does well, then we have a serious basis for making a more
>>aggressive challenge.
>>
>>The beauty of this system, is that it will be MUCH easier to
>>get a bunch of 2200+ players together for this experiment.
>>And it will tell us whether to continue with stronger players
>>or not.
>>
>>I personally take 2400 to be the lowest believable rating
>>that a program like Rebel could have, I believe it's likely
>>to be more like 2500. 2400 is MY lower bound. But since
>>Bob is more experienced in these matters I would be willing
>>to defer to him and try this experiment first, especially
>>since it would be MUCH easier to get it going. I still think
>>it's too early to go after the GM's.
>
>I fear the GM population will not be that anxious to play the computers not for
>any profound reason other than it interferes with their main focus. That is
>getting better at their game, defeating humans, winning tournaments and
>getting prize money. But from a related previous post someone said cash
>will draw them out of the woodwork to play computers - this I agree with and
>I also think it is a reasonable expectation. Hopefully there will be some
>GM's who take up the gauntlet and play match games simply for the sake of
>curiousity in discovering what these machines are capable of.
I feel the same way, GM's are simply focused on their own world of
chess and at best computers are an interesting diversion for them.
I really don't think they get that excited about what we do or
consider us any kind of threat, any more than a sprinter see's a
bicycle as a threat. And yes, they will play for money.
>Didn't Crafty have a go at Yermolinsky at a long time control? What happened
>there?
I'll let Bob answer that!
- Don
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.