Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Fritz 5.16 weaker say some - OPINIONS PLEASE

Author: blass uri

Date: 18:16:54 10/06/98

Go up one level in this thread



On October 06, 1998 at 19:35:50, Charles Milton Ling wrote:

>On October 06, 1998 at 16:56:23, blass uri wrote:
>
>>
>>On October 06, 1998 at 15:29:23, Charles Milton Ling wrote:
>>
>>>I have the vague, uncertain, and not scientifically substantiated feeling that
>>>Fritz 5.16 is slower than before the upgrade.
>>
>>I found that in the following position:
>>
>>r4rk1/2p3pp/p7/1p1pq3/8/2P2N2/PPQ2KPP/R1B5 white to move
>>
>>Fritz5.16 is significantly faster than fritz5 in avoiding kg1 that is losing
>>after Rxf3(I do not know about fritz5.03 because I did not download this update)
>>
>>Fritz5 needed more than an hour when fritz5.16 needs less than 15 minutes on my
>>pentium200MMX
>>
>I have a PentiumII 266 with 64 MB RAM.  Giving Fritz 44 of those (hash tables
>filled in about 3 minutes), he needed 45 minutes to fall out of love with Kg1,
>preferring Be3, but still giving Kg1 a 0.09 rating (at 15 ply, by the way).
>Strange, isn't it?  Of course, I have no idea what might have happened before
>the upgrade.

I have more ram but it is not the reason for what happened at depth 15.

I tried to give my fritz5.16 44MB hash tables(44*1024Kbytes=45056Kbytes)
and my results at depth 15(I asked it to analyze and not to play):

1)Kg1 has evaluation of   -2.34 after 14 minutes and 46 seconds
fritz5 looked at 214418 Knodes to discover it.

2)after 17 minutes and 20 seconds Fritz prefers Qd3 with an evaluation of
-2.31(this is not a final evaluation and fritz is "thinking" about Qd3).
Fritz5 looked at 257269 Knodes to discover Qd3

>>
>>
>>>  It seems to me that the ply
>>>numbers take longer to rev up than they used to.
>>
>>The ply numbers are not important
>>It is possible that the ply number take longer because the program sees more in
>>low plies.
>>
>Doesn't (for example) 8/15 mean that absolutely everything is examined over 8
>half-moves?  What can you see that is more than everything?

maybe I can see more in the selective half moves

>Thanks for your comments, Uri!  I still like Fritz, rest assured.  But I wonder
>why your Fritz is better than my Fritz...
>Charley

I do not know

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.