Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Gamescore between Fritz 7 vs The King 40/2

Author: Mike Byrne

Date: 19:13:33 11/26/02

Go up one level in this thread



>
>Depending on hardware and time control, The King MIGHT choose Ra2 here, but it
>appears to be much more likely to choose Kf8. This is the default personality on
>a P3-733.

It does depend on hardware and tc - at game 40 moves/2 hours on my machine even
the defualt personality will play Ra2 on my machine - it's just one of those
things.  Even on your  machine it would probably play that if it did not ponder
correctly - and it did't. (between 1:19 and 4:36 on your machine )



>
>Time	Depth	Score	Positions	Moves
>0:00	1/4	-0.56	23396		57...Kf8 58.Rxd5 Qxe6 59.Rd8+ Kf7
>					60.Bb5
>0:00	1/5	0.03	36594		57...Kf8 58.Rxd5 Qxe6 59.Rd6 f4
>					60.Rxe6 fxg3+ 61.Kxg3
>0:00	1/5	-0.43	55108		57...Kh7 58.Rxd5 Qxe6 59.Qc7+ Kh6
>					60.Rd7
>0:01	1/6	2.61	92893		57...Kh7 58.Rxd5 Kh8 59.Rd7 Qxe6
>					60.Rxa7 Rd2
>0:01	1/6	0.68	124356		57...Ra2 58.Bxf5 Kh8 59.Bxg6 Qxe6
>					60.Qc3+ Kg8 61.Bxh5
>0:02	1/6	-0.49	178483		57...Kf8 58.Rxd5 Rb2 59.Bb5 Qxe6
>					60.Rd8+ Kf7 61.Rd7+ Kg8
>0:03	1/7	-0.14	265434		57...Kf8 58.Rxd5 Rb2 59.Qg5 Bb8+
>					60.Kh3 Qxg5 61.hxg5 Ke7
>0:07	1/8	-0.03	589183		57...Kf8 58.Rxd5 Rb2 59.Kh3 Bf2
>					60.Qd6 Bxh4 61.Qxe7+ Bxe7 62.Rd7
>0:15	1/9	0.00	1363008		57...Kf8 58.Rxd5 Rb2 59.Kh3 Bf2
>					60.Qf4 Qxe6 61.Qh6+ Kf7 62.Qh7+
>					Kf6 63.Qh8+ Kf7 64.Qh7+
>0:39	2/10	0.77	3646767		57...Kf8 58.Rxd5 f4 59.Qg5 Qxg5
>					60.Rxg5 f3 61.Bxg6 Ke7 62.Bf7 Re2
>					63.Rg7
>1:19	2/10	0.72	7196698		57...Ra2 58.Bxf5 Kf8 59.Bxg6 Ra4
>					60.Qf3+ Kg7 61.Bf5 Kh6 62.Qxd5
>					Bb8+ 63.Kg1 Ba7+ 64.Kf1 Qxh4
>3:31	3/11	0.90	19024804	57...Ra2 58.Bxf5 Kf8 59.Bxg6 Ra4
>					60.Qf3+ Kg7 61.Bf5 Kh6 62.g3 Ra2+
>					63.Kh3 Bg1 64.Qf4+ Kg7
>4:36	3/11	0.60	25140498	57...Kf8 58.Rxd5 f4 59.Qg5 Qxg5
>					60.Rxg5 Ke7 61.Rxg6 Bd4 62.Kh3
>					Bf6 63.Bf5 Rd2
>
>While it is certainly possible that The King can lose to Fritz in any given
>game, it does seem likely that The King's setup in your tournament is faulty
>somehow. Scoring only 1.5 out of 7 seems like more than bad luck (at least to
>me, but I'm biased).

I have never seen The King do this poorly - but then again I have not played
40/2 too often.  General observation - it does not take enough time at this time
control.  Junior and Gandalf are very aggressive use almost all their time.

I am tempted after this tourney to run the gaunlet with  "The King" with 64MB
hash.  Utzinger has told me that 128MB hash does not woek well with "The King"
-by these results I have to agree.

It is also true that with a limited number of games - you can get results that
are outside of the expected results.  I think this is happening/.  I will
continue with current settings and then use Utzinger prescribed settings
afterwards.

Based on what is known about "The King" and my own experciences, I believe the
results are either an aberation or there is something about the 128MB hash that
makes it play weaker than expected. (or a combination of both).  It is
interesting that this is happening.

Utzinger also noticed that some of my parameters were out of normal sequence.
Perhaps that is also influence here.

I don't want to change settings in midstream - I rather finish what I started
with faulty settings perhaps and all - and then run the gaunlet with the more
correct "Utzinger" settings - and see what happens.

I may have to *asterik* the "The King" to show that these were not defualt
settings




>
>jm



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.