Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Junior 5 vs Fritz 5: significantly different programs

Author: blass uri

Date: 03:35:27 10/23/98

Go up one level in this thread



On October 22, 1998 at 17:31:50, Alessio Iacovoni wrote:

>On October 22, 1998 at 14:35:34, blass uri wrote:
>
>>
>>On October 22, 1998 at 14:15:48, Alessio Iacovoni wrote:
>>
>>>On October 22, 1998 at 11:21:49, blass uri wrote:
>>>
>>>>
>>>>On October 22, 1998 at 10:44:08, Larry Field wrote:
>>>>
>>>>>I'm looking for a chess program and have seen good reviews of both Fritz 5 and
>>>>>Junior 5.  However I see that Junior 5 is a 32-bit version of the Fritz 5
>>>>>"interface" and doesn't include some of the Fritz features (endgame and voice).
>>>>>Can someone explain how these two programs work together or separately for me?
>>>>>
>>>>>Also, can Junior be played with the voice and endgame features?  Is it
>>>>>necessary, for any reason, to own both Junior 5 and Fritz 5?
>>>>
>>>>Junior5 and Fritz5 are different programs.and do not play the same moves.
>>>>If you want to analyse positions Fritz5 will give you sometimes different output
>>>>from fritz5.
>>>>
>>>>Uri
>>>>
>>>>>
>>>>>Thanks,
>>>>>Larry Field
>>>
>>>Junior 5 at the moment has the most advanced and bug-free interface. It includes
>>>Junior 5 and Crafty (two fast searchers) and optionally Fritz 5 or Hiarcs. It is
>>>32 bit.. fast and will host also future 32 bit engines. As to the engines.. all
>>>fast searchers are basically the same... Junior, Fritz (or even Crafty) are not
>>>that different... so I'd suggest to go for the most advanced interface (at the
>>>moment Junior 5). In the future you can still add Hiarcs, which is quite
>>>different, has a very unique style of play and is pleasant to play with.
>>
>>I do not think that all the fast searchers are similiar.
>>
>>If Junior is more similiar to Fritz than to Hiarcs than it is easy to prove it.
>>
>>We can find how many times Junior5 suggest a different move from hiarcs6 and how
>>many times Junior5 suggest a different move from Fritz5.
>>
>>It is interesting to do a list of distances between programs (the % of times
>>they suggest different moves).
>>
>>If I have time I will do it for Fritz and Junior.
>>
>>Uri
>
>That's a great idea Uri! I'm sure it'll turn out that i'm right and you're
>wrong.

I did it for the game young-Rebel and I found that in most of the cases Junior5
and fritz5 play different moves(at least this is the case for the first 40
moves.
I found 43 cases out of 73 when the engines suggest different moves.
(I used the blunder check only with Threshold=0 for both sides).

I gave Fritz5 30 seconds per move and I gave ny beta version(16 bit) of Junior5
36 seconds per move because it is slower than the 32 bit by 20%

Here are some moves from this game with the ideas of fritz5 and Junior5

(They did not check moves 1-2,5 and the third move of white because it was part
of Junior's book)

1.d4 d6 2.c4 e5 3.e3 Nd7(Junior5 3...exd4 0.12/12 Fritz5 3...Nf6 0.09/10)
4.Nc3(Junior5 4.Nc3 0.28/13 Fritz5 4.Nc3 0.16/10) 4...Ngf6 (Junior5 4...c6
0.26/13 Fritz5 4...f5 0.16/10) 5.Nf3 Be7 6.Be2(Junior5 6.Bd3 0.31/13 Fritz5
6.Bd3 0.13/12) 6...0-0(Junior5 6...0-0 0.21/13, Fritz5 6...exd4 0.19/11)
7.0-0(Junior5 7.0-0 0.19/14, Fritz5 7.e4 0.25/10) 7...Re8(Junior5 7...c6 0.21/14
Fritz5 7...exd4 0.22/11) 8.Qc2(Junior5 8.Bd2 0.26/14 Fritz5 8.d5 0.19/12)
8...Bf8(Junior5 8...c6 0.20/14 Fritz5 8...exd4 0.22/10)

Uri



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.