Author: Pedro Gomes
Date: 13:40:16 08/22/05
Go up one level in this thread
On August 22, 2005 at 13:17:19, Thomas Logan wrote:
>On August 22, 2005 at 08:57:56, Pedro Gomes wrote:
>
>>On August 22, 2005 at 05:53:13, Thomas Logan wrote:
>>
>>>Now
>>>
>>>THOMAS-KBZT8WLT, 40'/40+40'/40+40' 0
>>>
>>> 12345678
>>>1 Shredder 9 1100111½ 5.5/8
>>>2 Toga II 1.0 0011000½ 2.5/8
>>>
>>>[Event "THOMAS-KBZT8WLT, 40'/40+40'/40+40'"]
>>>[Site "THOMAS-KBZT8WLT"]
>>>[Date "2005.08.20"]
>>>[Round "1"]
>>>[White "Shredder 9"]
>>>[Black "Toga II 1.0"]
>>>[Result "1-0"]
>>>[ECO "B90"]
>>>[Annotator "0.31;0.58"]
>>>[PlyCount "114"]
>>>[TimeControl "40/2400:40/2400:2400"]
>>>
>>>{W=20.1 ply; 729kN/s B=18.1 ply; 1,391kN/s} 1. e4 {[%emt 0:00:00]} c5 {
>>>[%emt 0:00:00]} 2. Nf3 {[%emt 0:00:00]} d6 {[%emt 0:00:00]} 3. d4 {
>>>[%emt 0:00:00]} cxd4 {[%emt 0:00:00]} 4. Nxd4 {[%emt 0:00:00]} Nf6 {
>>>[%emt 0:00:00]} 5. Nc3 {[%emt 0:00:00]} a6 {[%emt 0:00:00]} 6. Be3 {
>>>[%emt 0:00:00]} e5 {[%emt 0:00:00]} 7. Nf3 {[%emt 0:00:00]} Qc7 {[%emt 0:00:00]
>>>} 8. a4 {[%emt 0:00:00]} Be7 {[%emt 0:00:00]} 9. a5 {[%emt 0:00:00]} O-O {
>>>[%emt 0:00:00]} 10. Be2 {[%emt 0:00:00]} Nbd7 {[%emt 0:00:00]} 11. O-O {
>>>[%emt 0:00:00]} b5 {[%emt 0:00:00]} 12. axb6 {[%emt 0:00:00]} Nxb6 {
>>>[%emt 0:00:00]} 13. Nd2 {[%emt 0:00:00]} Bb7 {[%emt 0:00:00]} 14. Bxb6 {
>>>[%emt 0:00:00]} Qxb6 {[%emt 0:00:00]} 15. Nc4 {[%emt 0:00:00]} Qc5 {
>>>[%emt 0:00:00]} 16. Ra5 {White last book move [%emt 0:00:00]} Qc7 {
>>>[%emt 0:00:00]} 17. Qd3 {[%eval 31,18] [%emt 0:01:58]} Rad8 {[%emt 0:00:00]}
>>>18. Bf3 {[%eval 29,17] [%emt 0:01:26]} Rd7 {[%emt 0:00:00]} 19. Rfa1 {
>>>[%eval 48,17] [%emt 0:02:01]} Rc8 {Black last book move [%emt 0:00:00]} >
>>
>>
>>In this position white already has advantage.
>>How can you test black´s strenght in this case ?
>>There is a basic rule for testing chess engines:
>>
>>Use opening books for only 5-6 first moves, in any case never more than 10 (20
>>plies)
>>
>>HFP
>
>The problem is many engines are designed to play with their opening books
>
>No one test a Shredder against a Junior or a Fritz with a 5-6 move opening book
>
>At least that is so in my opinion
>
>When an engine plays in WCC it is not limited to a 5-6 move book
>
>When the sdss test a computer they do not limit to a 5-6 move book
>
>So When I test I test with the book the engine comes with
>
>If the oponnent does not have a book or a very small one I try to provide one
>that has a good reputation
>
>In this case I used the deep Fritz 8 book. Fruit played very well with that
>book in 100 games against Shreder and Fritz bilbao
>
>While a good book I suspect like other books it has some bad lines
>
>In this particular case I had no time to examine the position
>
>however I am open to suggestions for a book for Toga
>
>Tom
I can tell you only what I think is better for testing chess engines, and I am
not interested what other peoples do.
If you test engines, book is not importnate. In this case book is only a tool in
a test process. When I test engines I observe both engines score after exiting
the their books, and if the score is not balances I stop the game. As I said: I
test engines, not their books.
If one engine´s book insite in "bad" posititions after book moves, I replace its
book with mine, in Arena.
Without Arena I would use Nunn tests, which offers balances starting positions,
and would play the same position for every engine, with white and black.
HFP
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.