Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: To Ed - what time do Rebel10 get ?

Author: Harald Faber

Date: 04:29:40 08/09/98

Go up one level in this thread


On August 07, 1998 at 03:29:15, Ed Schröder wrote:

>>>Solution times (PII-266) of R10 for the 3 positions...
>
>>>Rxd7  0:08
>>>Rxc6  0:02
>>>Rxa7  0:08
>
>>>Still not perfect but definitely better than R9.
>
>>>- Ed -
>
>>Ed, could you explain why Rebel9 plays all the moves Rebel10 played in the >Anand game (Rebel black, tournament) except for only one move? This is amazing >and certainly no reason to get Rebel10 when the moves are 99% the same. I know >this question is difficult to answer, you need sales to survive. Don't see it >as an attack but as a reasonable question.
>
>I don't feel offended as you can ask me whatever question you want but I
>hope you don't mind I will give a strong answer before it becomes a hot
>item here as you are the second person who is asking me this. Your
>conclusion is entirely wrong at least in my point of view.
>
>I did not check game-7 Anand-Rebel with R9. I checked this game only how
>effective anti-GM did its work with R10. So I can not judge if your 99%
>is right or wrong.

As I wrote, I let R9 analyze 3min/move on K6-200 28MB hash.

>If true this must be an exceptional case.
>Have you checked the other 7 games?

No. I'll check #8 for it is the 2nd tournament game. Blitz and rapid are not
important for me.
But of course I'll give the result of #8 too as it is finished.

>Next, recently I have posted the differences between R9 and R10.
>Did you miss this?

Do you mean the few solution results ? In this case it is not significant if a
key move is found in 5 or 30 sec, at least in tourney.

>The statistic showed a similar result when I compared R6 with R7,
>R7 with R8 and R8 with R9. 10-15% different moves per version.
>Now you can argue about this percentage. Too low? Too high?

10-15% would be OK for me.

>Personally I think this 10-15% is a fine number. If you for example
>get a number of 30% different moves you may strongly wonder how good(?!)
>your previous version was :-)

This question is always interesting. :-)
But of course I'd be proud of how good the actual version is. :-)

>I like to keep the good moves Rebel played and to do something about
>the bad moves. Feel free to disagree with my opinion on this 30%. Surely
>I can produce a R10 which plays 30-40% different moves and still plays
>good chess but I assume people want the best Rebel available so they
>get the best Rebel as I (as the author) can produce and judge this.

I think best comparison will be further tournament games. I'd say if R9 plays
over 95% the moves R10 plays strength is not the point to buy Rebel10, but...

>Last item, I am a little bit surprised by the question(s). If you have
>a look at the Rebel10 Diary on my home page you will notice that on
>the Rebel10 chess engine more work is done than on R7,R8 and R9 all
>together as several new and useful functions are added to the engine.
>No need to list them here just have a look yourself.
>- Ed -

I know you always add senseful features, but computerchessmaiacs always want
also a stronger engine. :-)



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.