Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: But Shredder 5, SHOULD be better than last years programs!

Author: Enrique Irazoqui

Date: 06:09:45 08/27/00

Go up one level in this thread


On August 27, 2000 at 08:56:18, Uri Blass wrote:

>On August 27, 2000 at 08:43:33, Enrique Irazoqui wrote:
>
>>On August 27, 2000 at 08:19:14, Rajen Gupta wrote:
>>
>>>unless shredder5 is rigorously tested either by the SSDF or by enrique irazoqui
>>>against other top and not so top programmes over a large number of games, one
>>>cannot confidently proclaim that shredder5 is the strongest so far, let alone
>>>stronger overall than shredder4. so far fritz6 and junior 6  taken on all comers
>>>and come out on top.this result has been duplicated by enrique in the cadaques
>>>tournament.unless there is proof to the contrary i would still consider fritz6
>>>nad junior6 to jointly hold the title of the ''stongest single processor PC
>>>programmes''. a newer version does not necessarily mean stronger(Mchess pro,
>>>rebel and genius are notable examples of newer versions being weaker than their
>>>predecessors.
>>>
>>>Shredder5, no doubt, is strong, but is it the strongest?
>>>
>>>rajen
>>
>>Hi Rajen,
>>
>>The new Shredder will be stronger than Shredder 4, I am sure of it. The current
>>beta is between 30 and 60 points stronger than S4, and the final version will be
>>even stronger. That's about the engine alone. Sandro's book could significantly
>>contribute to Shredder's performance, particularly in comp-comp, although much
>>less than years ago, before programs had learners.
>>
>>Sandro posted that S5 will be 80 points stronger than F6a, which means about 150
>>more than S4. I don't believe it for a second. Programs improve and the new
>>Shredder will be great, but miracles are something else.
>
>We will have to wait and see.
>
>150 elo improvement is not a miracle.

In one year it is not real, it is not even close to reality. A more
knowledgeable voice than mine would confirm.

Now, that aplies to the engine. The contribution of a special opening book is
something else. But if a book can add up so much to engine's performance it
would be the right time to quit comp-comp, at least for me.

Enrique

>It is a very good result but it is not impossible.
>
>It is clearly possible in 3 years and I see no proof that it is impossible in 1
>year.
>
>I am also not sure if 80 points stronger than Fritz6a will be enough to be
>number 1 in the ssdf list.
>
>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.