| View previous topic :: View next topic |
| Author |
Message |
Luca Hemmerich
Joined: 23 Sep 2009
Posts: 344
Location: Germany
|
 Posted: Fri Jun 04, 2010 11:25 am Post subject: ChessMind 0.82 |
  |
|
I just released ChessMind 0.82.
Changes to ChessMind 0.80:
- better at tactics, this at the cost of lower search depth
- improvements in evaluation, does not sacrifice pawns without any reason as often as 0.80
- some more UCI options (explained in the readme file)
- improvements in time management
- smaller aspiration windows (also adjustable)
The strength should be around 2100 CCRL now (at least I hope so  ).
Download |
|
| Back to top |
|
 |
Swaminathan
Joined: 09 Mar 2006
Posts: 6488
|
| Posted: Sat Jun 05, 2010 2:56 am Post subject: Re: ChessMind 0.82 |
|
|
Thanks, Luca. I believe it's about 2150. Decent engine suitable for most National Masters.
0.82 has improved +12 points over the previous version.
 |
|
| Back to top |
|
|
Graham Banks
Joined: 26 Feb 2006
Posts: 19755
Location: Auckland, NZ
|
| Posted: Thu Jul 15, 2010 6:55 pm Post subject: Something appears to be wrong |
|
|
| metax wrote: |
I just released ChessMind 0.82.
Changes to ChessMind 0.80:
- better at tactics, this at the cost of lower search depth
- improvements in evaluation, does not sacrifice pawns without any reason as often as 0.80
- some more UCI options (explained in the readme file)
- improvements in time management
- smaller aspiration windows (also adjustable)
The strength should be around 2100 CCRL now (at least I hope so ).
Download |
Hi Luca,
it appears here that something is not right with this version. It comes straight out of book thinking it's over 2 pawns ahead in every game! 
As a result, it's performing poorly as one would expect in this tournament I'm running:
http://kirill-kryukov.com/chess/discussion-board/viewtopic.php?f=7&t=5255
Cheers,
Graham.
_________________
My email addresses:
gbanksnz at gmail.com
gbanksnz at yahoo.co.nz |
|
| Back to top |
|
|
Dr.Wael Deeb
Joined: 08 Mar 2006
Posts: 8618
Location: Amman,Jordan
|
| Posted: Thu Jul 15, 2010 7:22 pm Post subject: Re: Something appears to be wrong |
|
|
| Graham Banks wrote: |
| metax wrote: |
I just released ChessMind 0.82.
Changes to ChessMind 0.80:
- better at tactics, this at the cost of lower search depth
- improvements in evaluation, does not sacrifice pawns without any reason as often as 0.80
- some more UCI options (explained in the readme file)
- improvements in time management
- smaller aspiration windows (also adjustable)
The strength should be around 2100 CCRL now (at least I hope so ).
Download |
Hi Luca,
it appears here that something is not right with this version. It comes straight out of book thinking it's over 2 pawns ahead in every game!
As a result, it's performing poorly as one would expect in this tournament I'm running:
http://kirill-kryukov.com/chess/discussion-board/viewtopic.php?f=7&t=5255
Cheers,
Graham. |
It's good to have a high level of self-confidence Graham 
_________________
It ain't about how hard you can hit,it's about how hard you can get hit and keep moving forward,how much you can take and keep moving forward. |
|
| Back to top |
|
|
Graham Banks
Joined: 26 Feb 2006
Posts: 19755
Location: Auckland, NZ
|
| Posted: Thu Jul 15, 2010 7:49 pm Post subject: Re: Something appears to be wrong |
|
|
| Dr.Wael Deeb wrote: |
| Graham Banks wrote: |
| metax wrote: |
I just released ChessMind 0.82.
Changes to ChessMind 0.80:
- better at tactics, this at the cost of lower search depth
- improvements in evaluation, does not sacrifice pawns without any reason as often as 0.80
- some more UCI options (explained in the readme file)
- improvements in time management
- smaller aspiration windows (also adjustable)
The strength should be around 2100 CCRL now (at least I hope so ).
Download |
Hi Luca,
it appears here that something is not right with this version. It comes straight out of book thinking it's over 2 pawns ahead in every game!
As a result, it's performing poorly as one would expect in this tournament I'm running:
http://kirill-kryukov.com/chess/discussion-board/viewtopic.php?f=7&t=5255
Cheers,
Graham. |
It's good to have a high level of self-confidence Graham |
Indeed, but to be over confident? 
_________________
My email addresses:
gbanksnz at gmail.com
gbanksnz at yahoo.co.nz |
|
| Back to top |
|
|
Graham Banks
Joined: 26 Feb 2006
Posts: 19755
Location: Auckland, NZ
|
| Posted: Sun Jul 18, 2010 9:15 am Post subject: Can anybody else confirm this? |
|
|
Can somebody else who uses ChessMind confirm this behaviour?
The previous version was okay.
_________________
My email addresses:
gbanksnz at gmail.com
gbanksnz at yahoo.co.nz |
|
| Back to top |
|
|
Ben-Hur Carlos Langoni
Joined: 16 Oct 2008
Posts: 449
Location: Milky Way
|
| Posted: Sun Jul 18, 2010 11:26 am Post subject: Re: Can anybody else confirm this? |
|
|
| Graham Banks wrote: |
Can somebody else who uses ChessMind confirm this behaviour?
The previous version was okay. |
Hi Graham,
Yes I can confirm that. I am using ChessMind to test Redqueen and it appears that it is losing every game because of that. It seems like a problem in the evaluation function regarding dynamic terms, perhaps mobility or King safety. The engine sometimes evaluates equal positions to 1 or 2 pawns ahead, then when it starts losing material the score drops to something more realistic.
Regards, |
|
| Back to top |
|
|
Graham Banks
Joined: 26 Feb 2006
Posts: 19755
Location: Auckland, NZ
|
| Posted: Sun Jul 18, 2010 6:27 pm Post subject: Re: Can anybody else confirm this? |
|
|
| bhlangonijr wrote: |
| Graham Banks wrote: |
Can somebody else who uses ChessMind confirm this behaviour?
The previous version was okay. |
Hi Graham,
Yes I can confirm that. I am using ChessMind to test Redqueen and it appears that it is losing every game because of that. It seems like a problem in the evaluation function regarding dynamic terms, perhaps mobility or King safety. The engine sometimes evaluates equal positions to 1 or 2 pawns ahead, then when it starts losing material the score drops to something more realistic.
Regards, |
Thanks Ben-Hur. 
_________________
My email addresses:
gbanksnz at gmail.com
gbanksnz at yahoo.co.nz |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
