Hello:
I saw an interesting thread on Rybka Forum yesterday:
http://rybkaforum.net/cgi-bin/rybkaforu ... ?tid=26785
It can be an interesting alternative to Interactive Deep Analysis (IDeA)... well, I do not know how IDeA exactly works, but I think it uses a MonteCarlo approach.
According to the author, his tool uses a modified SF 2.2.2 specially made for this task. I do not have intention to try it, but if someone is interested then this tool could be useful:
http://chessengine.blogspot.it/
Regards from Spain.
Ajedrecista.
New MonteCarlo free analysis tool.
Moderator: Ras
-
Ajedrecista
- Posts: 2177
- Joined: Wed Jul 13, 2011 9:04 pm
- Location: Madrid, Spain.
-
AdminX
- Posts: 6363
- Joined: Mon Mar 13, 2006 2:34 pm
- Location: Acworth, GA
Re: New MonteCarlo free analysis tool.
Interesting name for a chess tool. LOL 
"Good decisions come from experience, and experience comes from bad decisions."
__________________________________________________________________
Ted Summers
__________________________________________________________________
Ted Summers
-
jshriver
- Posts: 1371
- Joined: Wed Mar 08, 2006 9:41 pm
- Location: Morgantown, WV, USA
Re: New MonteCarlo free analysis tool.
AgreeAdminX wrote:Interesting name for a chess tool. LOL
-
Vinvin
- Posts: 5312
- Joined: Thu Mar 09, 2006 9:40 am
- Full name: Vincent Lejeune
Re: New MonteCarlo free analysis tool.
Do someone already use and test it ?
-
jefk
- Posts: 1084
- Joined: Sun Jul 25, 2010 10:07 pm
- Location: the Netherlands
- Full name: Jef Kaan
Re: New MonteCarlo free analysis tool.
[quote="Vinvin"]Do someone already use and test it ?[/quote]
just tried it, surely interesting concept, and promising tool.
some preliminary findings:
startup requires some time, to determine the first moves considered, done in a dos console. then during analysis you dont get graphical output, only after pressing quit results are ported to the graphical interface.
So i could use it, having hardly looked at the readme instructions,
but on the other hand its not highly easy to use (yet?), although
probably easier for the avereage user than Idea in Aquarium.
But also works different. would be interesting to compare analysis
results with the two methods.
As general comment i wonder how useful this tool would be for
middlegame positions, compared with Idea, or compared with just
letting Houdini3 run for a night or so in infinite analysis.
For complicated endgame positions, Monte Carlo seems to have
an advantage above the conventional engine analysis as
the latter doesnt see fortresses, and similar difficult structures.
But shouldnt then -for endgame analysis- not also endgame
tablebases be used in the (engine) analysis within the MC search ?
And Stockfish, the engine used in this cockroach thing doesnt
have endgame bases (yet?) isnt it. But in general it looks like a
promising 'tool'. so promising some programmers might start
copying the concept, but then you are the first so have an
advantage i suppose.
jef
PS in fact one could wonder to what extent MC search could/should
replace or supplement conventional search during slow/standard
games to achieve better results.
just tried it, surely interesting concept, and promising tool.
some preliminary findings:
startup requires some time, to determine the first moves considered, done in a dos console. then during analysis you dont get graphical output, only after pressing quit results are ported to the graphical interface.
So i could use it, having hardly looked at the readme instructions,
but on the other hand its not highly easy to use (yet?), although
probably easier for the avereage user than Idea in Aquarium.
But also works different. would be interesting to compare analysis
results with the two methods.
As general comment i wonder how useful this tool would be for
middlegame positions, compared with Idea, or compared with just
letting Houdini3 run for a night or so in infinite analysis.
For complicated endgame positions, Monte Carlo seems to have
an advantage above the conventional engine analysis as
the latter doesnt see fortresses, and similar difficult structures.
But shouldnt then -for endgame analysis- not also endgame
tablebases be used in the (engine) analysis within the MC search ?
And Stockfish, the engine used in this cockroach thing doesnt
have endgame bases (yet?) isnt it. But in general it looks like a
promising 'tool'. so promising some programmers might start
copying the concept, but then you are the first so have an
advantage i suppose.
jef
PS in fact one could wonder to what extent MC search could/should
replace or supplement conventional search during slow/standard
games to achieve better results.