Komodo 2.03 running for the IPON
Komodo 2.03 running for the IPON
Ponder ON rating list: http://www.inwoba.de
-
- Site Admin
- Posts: 1226
- Joined: Wed Jun 09, 2010 7:49 am
- Real Name: Jeremy Bernstein
- Location: Berlin, Germany
- Contact:
Re: Komodo 2.03 running for the IPON
It's still too early to say definitively, but the new Komodo looks like a huge leap forward. Fascinating is that it's getting so pummeled by Stockfish -- looks like the engine was tuned to be a Rybka-killer. Can you comment on _how_ Komodo is losing to Stockfish? In the absence of the games (yes, that again), it's impossible to draw any conclusions.
Jeremy
Jeremy
Re: Komodo 2.03 running for the IPON
I barely look at the games at all. The only thing I realizerd it that Komodo might gain some points with a better endgame. There is definately room for improvement.Jeremy Bernstein wrote:It's still too early to say definitively, but the new Komodo looks like a huge leap forward. Fascinating is that it's getting so pummeled by Stockfish -- looks like the engine was tuned to be a Rybka-killer. Can you comment on _how_ Komodo is losing to Stockfish? In the absence of the games (yes, that again), it's impossible to draw any conclusions.
Jeremy
Actually to judge something about "pummeled" is impossible out of 50 games. Even out of 100 at the end it is quite bold. Have a look at the statistics I offer for download. You can see out of the result vs the individual engines with just 100 games a range in result of 100+ Elo is nothing unusuall. So I think that it is impossible to draw anything about superiority of individual engines regardless if the games are available or not.
The only conclusion which I consider valid is the overall result, the individual results with just 100 games are statisticaly too "noisy"
Bye
Ingo
Ponder ON rating list: http://www.inwoba.de
Re: Komodo 2.03 running for the IPON
ALL tests or tournaments without the games bieng posted to validate results are at best unreliablie and at worse highly suspectIWB wrote:I barely look at the games at all. The only thing I realizerd it that Komodo might gain some points with a better endgame. There is definately room for improvement.Jeremy Bernstein wrote:It's still too early to say definitively, but the new Komodo looks like a huge leap forward. Fascinating is that it's getting so pummeled by Stockfish -- looks like the engine was tuned to be a Rybka-killer. Can you comment on _how_ Komodo is losing to Stockfish? In the absence of the games (yes, that again), it's impossible to draw any conclusions.
Jeremy
Actually to judge something about "pummeled" is impossible out of 50 games. Even out of 100 at the end it is quite bold. Have a look at the statistics I offer for download. You can see out of the result vs the individual engines with just 100 games a range in result of 100+ Elo is nothing unusuall. So I think that it is impossible to draw anything about superiority of individual engines regardless if the games are available or not.
The only conclusion which I consider valid is the overall result, the individual results with just 100 games are statisticaly too "noisy"
Bye
Ingo
Re: Komodo 2.03 running for the IPON
I fully agree here and the best advice is not to look at these sites. Best is not to pay attention at all, it is just supporting the makers.veritas wrote:
ALL tests or tournaments without the games bieng posted to validate results are at best unreliablie and at worse highly suspect
And believe me, there is no irony in my sentences - actualy I am suppriesed how many people are interested despite the fact that there are no games available. But I have learned that nearly nobody is interested in conditions, not to speak about games.
Bye
Ingo
PS: Hihi, 'total posts one'. To make an account just to make that posting ... and then I agree, you could have used your normal account.
Ponder ON rating list: http://www.inwoba.de
Re: Komodo 2.03 running for the IPON
IWB wrote:I fully agree here and the best advice is not to look at these sites. Best is not to pay attention at all, it is just supporting the makers.veritas wrote:
ALL tests or tournaments without the games bieng posted to validate results are at best unreliablie and at worse highly suspect
And believe me, there is no irony in my sentences - actualy I am suppriesed how many people are interested despite the fact that there are no games available. But I have learned that nearly nobody is interested in conditions, not to speak about games.
Bye
Ingo
PS: Hihi, 'total posts one'. To make an account just to make that posting ... and then I agree, you could have used your normal account.
your sarcasams less usefull than 1000s of letters and numbers purporting to be tournaments and probaly just as accurate
2 questions , not difficult ones
1) why when there are many that wouls (contary to your opinion ) appreciate and use games from your tournaments , as either to data base or for makeing books , do you refuse to post said games ?
its not difficult , im sure with a little practice you could manage it
2) WTF would anyone much less myself bother to make fake account to comment on your or any other posts?
ok a 3rd question in 2 parts
Are you paranoid Do you have "somehing " to be paranoid about
Re: Komodo 2.03 running for the IPON
Still trying to provoke. Sweet.
Rgds
Ingo
Ponder ON rating list: http://www.inwoba.de
Re: Komodo 2.03 running for the IPON
Hi,
Komodo 2.03 Dc SSE42 with an impressive win over its predecessor!
Have a look here: http://www.inwoba.de
Bye
Ingo
Komodo 2.03 Dc SSE42 with an impressive win over its predecessor!
Have a look here: http://www.inwoba.de
Bye
Ingo
Ponder ON rating list: http://www.inwoba.de
Re: Komodo 2.03 running for the IPON
And today, 2011.06.17 the IPON has the most hits of all times!
Thanks for that!
Bye
Ingo
Thanks for that!
Bye
Ingo
Ponder ON rating list: http://www.inwoba.de
Re: Komodo 2.03 running for the IPON
IWB wrote:And today, 2011.06.17 the IPON has the most hits of all times!
Thanks for that!
Bye
Ingo
questions to difficult for you then
as for above from you... ok , enjoy eating crap as 50 millions flys cant possibly be wrong can they
one more "hit " for you