Three Crown Seedings and schedule

[Dann Corbit]

Geschrieben von: / Posted by: Dann Corbit at 01 July 2000 05:32:01:

3 CROWN SEEDINGS...
SEED:
01. ArasanX
02. LambChop
03. GnuChess4
04. Ant
05. InmiChess3
06. ExChess3
07. Knightx
08. Dragon
09. GnuChess5
10. Fortress
11. LDBlanche
12. Amyan

SEED:
\ 1 2 3 4 5 6 7 8 9 10 11 12
round \................................................
1: 12 11 10 9 8 7 6 5 4 3 2 1
2: 7 12 11 10 9 8 1 6 5 4 3 2
3: 2 1 12 11 10 9 8 7 6 5 4 3
4: 8 3 2 12 11 10 9 1 7 6 5 4
5: 3 4 1 2 12 11 10 9 8 7 6 5
6: 9 5 4 3 2 12 11 10 1 8 7 6
7: 4 6 5 1 3 2 12 11 10 9 8 7
8: 10 7 6 5 4 3 2 12 11 1 9 8
9: 5 8 7 6 1 4 3 2 12 11 10 9
10: 11 9 8 7 6 5 4 3 2 12 1 10
11: 6 10 9 8 7 1 5 4 3 2 12 11

[Mogens Larsen]

Geschrieben von: / Posted by: Mogens Larsen at 01 July 2000 08:27:01:
Als Antwort auf: / As an answer to: Three Crown Seedings and schedule geschrieben von: / posted by: Dann Corbit at 01 July 2000 05:32:01:

09. GnuChess5

Why does the GnuChess 5 version on your page only use 2Mb for hash?
Best wishes...
Mogens

[Dann Corbit]

Geschrieben von: / Posted by: Dann Corbit at 01 July 2000 09:35:25:
Als Antwort auf: / As an answer to: Might as well try again. geschrieben von: / posted by: Mogens Larsen at 01 July 2000 08:27:01:

09. GnuChess5

Why does the GnuChess 5 version on your page only use 2Mb for hash?

I have a 32 meg version that I am using. I just have not posted that.
I might just modify the sources myself to that it uses dynamic space, since it is such a pest to recompile for changing hash size.
As it stands, you have to change two constants in a header and recompile the whole mess.


My FTP site

[Mogens Larsen]

Geschrieben von: / Posted by: Mogens Larsen at 01 July 2000 09:39:13:
Als Antwort auf: / As an answer to: Re: Might as well try again. geschrieben von: / posted by: Dann Corbit at 01 July 2000 09:35:25:

I have a 32 meg version that I am using. I just have not posted that.
I might just modify the sources myself to that it uses dynamic space, since it is such a pest to recompile for changing hash size.
As it stands, you have to change two constants in a header and recompile the whole mess.

A 32Mb version would be excellent for me, but dynamic space is the best longterm solution. When you upload a new version post a message here if you don't mind. BTW, have you tried adding a few Kb to Gully's pawnhash, or is that not possible?
Best wishes...
Mogens

[Dann Corbit]

Geschrieben von: / Posted by: Dann Corbit at 01 July 2000 09:54:34:
Als Antwort auf: / As an answer to: Re: Might as well try again. geschrieben von: / posted by: Mogens Larsen at 01 July 2000 09:39:13:

I have a 32 meg version that I am using. I just have not posted that.
I might just modify the sources myself to that it uses dynamic space, since it is such a pest to recompile for changing hash size.
As it stands, you have to change two constants in a header and recompile the whole mess.

A 32Mb version would be excellent for me, but dynamic space is the best longterm solution. When you upload a new version post a message here if you don't mind. BTW, have you tried adding a few Kb to Gully's pawnhash, or is that not possible?

I won't be able to post a GnuChess5 update until Monday (by the way, it's very much stronger than the previous iteration). I'm at home and work is 45 miles off.
I have not looked at G2 pawn hash, but I suspect it's an easy fix. G2 searches like the blue blazes. But there is some serious flaw in the search. I have seen it consistently outsearch another program by 5 plies and still lose. That's strangely incredible to me. It does some other strange things. Once the kinks are worked out, I predict greatness for G2. Of course, I turned down the initial offer for Microsoft stock.
"That crappy OS!? Unix is going to blow it's doors off. It will be lucky to last two more years." I said to the one who showed me the offer.




My FTP site

[Pete Galati]

Geschrieben von: / Posted by: Pete Galati at 02 July 2000 10:16:42:
Als Antwort auf: / As an answer to: Re: Might as well try again. geschrieben von: / posted by: Dann Corbit at 01 July 2000 09:54:34:

I have a 32 meg version that I am using. I just have not posted that.
I might just modify the sources myself to that it uses dynamic space, since it is such a pest to recompile for changing hash size.
As it stands, you have to change two constants in a header and recompile the whole mess.

A 32Mb version would be excellent for me, but dynamic space is the best longterm solution. When you upload a new version post a message here if you don't mind. BTW, have you tried adding a few Kb to Gully's pawnhash, or is that not possible?

I won't be able to post a GnuChess5 update until Monday (by the way, it's very much stronger than the previous iteration). I'm at home and work is 45 miles off.
I have not looked at G2 pawn hash, but I suspect it's an easy fix. G2 searches like the blue blazes. But there is some serious flaw in the search. I have seen it consistently outsearch another program by 5 plies and still lose. That's strangely incredible to me. It does some other strange things. Once the kinks are worked out, I predict greatness for G2. Of course, I turned down the initial offer for Microsoft stock.
"That crappy OS!? Unix is going to blow it's doors off. It will be lucky to last two more years." I said to the one who showed me the offer.

What you could maybe do with GNUchess is name the different compiles by the amound of hashtable they use like what's done with Averno. It works great. You could have a GNUchess8, GNUchess16, GNUchess32.exe etc.
You can't see the toe tag while MS is wearing socks & shoes.
Pete