Author: Christophe Theron
Date: 09:14:12 09/11/98
Go up one level in this thread
On September 10, 1998 at 16:48:34, Hans Christian Lykke wrote:
>The GREAT NunnTest has ended.
>
>Those of you who want the games in PGN, please send me a E-mail, and
>I will send them to you. I think it´s too many to publish here.
>
>Facts: in 3 NunnTests Fritz 5 - Hiarcs 6 1 - 2
>Hiarcs 6 best at av.67. sec. and av.600 sec. and
>Fritz 5 best at av.200. sec.
>
>Totally: Fritz 5 - Hiarcs 6 31,5 - 28,5
>
>Conclusion: I think it is very difficult to draw a clear conclusion, because
>Hiarcs 6 is best at the fastest and the slowest level, and
>Fritz 5 best at the in between.
>
>The meeting Fritz 5 - Hiarcs 6/DOS at the SSDF 40/2 was
>20,5 - 21,5.
>
>Nunn - test P200MMX 64MB engine/engine, no book, no PB
>Both programs have 23MB hash - In the first line Fritz5 has white then black.
>
> av 67 seconds av 200 seconds av 600 seconds
> Fritz 5-Hiarcs 6 Fritz 5-Hiarcs 6 Fritz 5-Hiarcs 6
> res. subtot move res. subtot move res. subtot move
>1- B33 1-0 1-0 48 1/2 0,5-0,5 72 1-0 1-0 40
>1- B33 1-0 2-0 66 1-0 1,5-0,5 50 1-0 2-0 38
>2- B89 0-1 2-1 76 1-0 2,5-0,5 68 0-1 2-1 57
>2- B89 0-1 2-2 43 1-0 3,5-0,5 65 0-1 2-2 86
>3- C19 0-1 2-3 79 1-0 4,5-0,5 67 0-1 2-3 60
>3- C19 0-1 2-4 78 1/2 5-1 60 1/2 2,5-3,5 56
>4- C97 1-0 3-4 78 1-0 6-1 89 1/2 3-4 68
>4- C97 0-1 3-5 62 1/2 6,5-1,5 146 0-1 3-5 50
>5- D36 1/2 3,5-5,5 85 1-0 7,5-1,5 50 1-0 4-5 53
>5- D36 1/2 4-6 144 1-0 8,5-1,5 39 1/2 4,5-5,5 68
>6- D85 1/2 4,5-6,5 35 1-0 9,5-1,5 37 1/2 5-6 41
>6- D85 0-1 4,5-7,5 69 0-1 9,5-2,5 106 0-1 5-7 68
>7- E15 1-0 5,5-7,5 76 1-0 10,5-2,571 1-0 6-7 62
>7- E15 1/2 6-8 39 0-1 10,5-3,587 1-0 7-7 54
>8- E98 1/2 6,5-8,5 63 1/2 11-4 41 1-0 8-7 91
>8- E98 1/2 7-9 115 1/2 11,5-4,534 0-1 8-8 36
>9- A25 1-0 8-9 49 0-1 11,5-5,558 1-0 9-8 72
>9- A25 1/2 8,5-9,5 53 1/2 12-6 36 0-1 9-9 54
>10-B97 0-1 8,5-10,556 1-0 13-6 59 0-1 9-10 70
>10-B97 1/2 9-11 71 0-1 13-7 62 1/2 9,5-10,572
>Total 9-11 13-7 9,5-10,5
>
>
>Venlig hilsen
>Hans Chr.
The results only show that you cannot say that one opponent is better than the
other one.
You should not draw any conclusion. None of the 4 results (67, 200, 600 or
overall) allow this. Even the 13-7 result is, IMO, not relevant when compared to
the error margin you can expect in a 10 games match between equal opponents.
Your work is interesting in the sense that it clearly shows that Fritz and
Hiarcs are very close, so close that 60 games are not enough to decide which one
is the best.
Christophe
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.