Deep Junior defeated Deep Fritz in a 6-game match which took place in Elista. The first 2 games were drawn. Junior won game 3 and 4 and the final 2 games ended in draws. The final match score was 4-2 in favor of Junior. Congratulations to the Junior team!
Deep Fritz – Deep Junior [A30] Game 6
Ultimate Computer Challenge, Elista, RUS (6), 11.06.2007
1.Nf3 c5 2.c4 Nf6 3.g3 b6 4.Bg2 Bb7 5.0-0 g6 6.Nc3 Bg7 7.d4 cxd4 8.Qxd4 d6 9.Rd1 Nbd7 10.Be3 Rc8 11.Rac1 0-0 12.Qh4 a6 13.b3 Rc7 14.Bh3 Qe8 15.g4 Qa8 16.Bd4 e6 17.g5 Ne8 18.Bxg7 Nxg7 19.Rxd6 Bxf3 20.exf3 Qxf3 21.Ne4 Nh5 22.Bg2 Qf5 23.Rcd1 Nc5 24.Nf6+ Nxf6 25.gxf6 h5 26.Rxb6 Rd7 27.Rxd7 Nxd7 28.Rxa6 Nxf6 29.Qd4 Qb1+ 30.Bf1 Qf5 31.h3 Kg7 32.f4 Re8 33.Qe5 Qb1 34.Ra7 Qc1 35.a4 Qd1 36.Rb7 Qf3 37.Bg2 Qg3 38.Rb8 Rxb8 39.Qxb8 h4 40.Qe5 Kh7 41.Qxf6 Qe3+ ½-½.
So:
Kramnik lost to Deep Fritz
Deep Fritz lost to Deep junior
So that makes Deep Junior the
World Chess Champion?
I don’t know how the creator of Rybka has done it…he is an IM in his own right, but to have a chess program match, or any human vs chess program match and that program not be Rybka is a facade and everyone knows that Rybka is better….for now.
The offer made by Rybka’s creator was fair and has went unanswered by FIDE or the computer chess “powers that be.”
Rather than having the WCC play a program (i.e. Kramnik vs Fritz 10) that we all know to be much weakther than Rybka…why not have a match with a team of human GM’s vs Rybka or simply play an odds match.
Computer chess software has surpassed the best human player(s). There is no denying it. So, let’s spice it up a bit and have a team of GM’s play against Rybka or have a “Rybka vs the world” match. I don’t know…I’m just brainstorming. But, the simple fact is: Humans have created super strong chess software and they have surpassed us. It’s no longer a question of “human vs computer” which is better? but rather, “What can our best GM’s learn from these program to make human GM’s better?”
Only a GM can answer how Rybka can help them improve. I’m no GM and far away from having a GM’s understanding of the game. But, I guess most have Rybka so it must influence their playing style in some manner.
Just some randome thoughts. Thanks.
I too am frustrated by all these computer matches that do not include Rybka. I want to see Rybka play and soundly defeat all the other chess programs.
I do not have a rybka program. But I know it is the best one and best in many important ways. not just a tweek or two. The program is solidly better. A big jump better.
The longer this goes on the more I get frustrated with ChessBase. This is not an attempt to present the truth but an attempt to conceal the truth.
This whole ordeal of computer matches are stupid. How can you really compare two programs running on two machines with different specifications? Not to mention this Rybka guy complains that he hasn’t been included in the matches.
Really, if you want to have computer vs. computer match, you need to run them in the same environment. period.
What a waste of money and time. I thought that the developers of Fritz and Junior are smarter than that… As well as they can claim that Junior is the “best” program in the world, they can also claim that Fritz hasn’t lost to Junior because they are running on slower machine.
Did anybody notice from the report that Junior was running on a 16-way and Fritz was only running on an 8-way CPU machine. Something like 25 million positions a second versus 16 million. No wonder Junior won!
Susan, why do you publish every game of the Fritz-Junior marketing event and do not mention the ongoing Computer Chess World Championships? Fritz and Junior both avoid it for obvious reasons. Chessbase does mention that Junior is Computer World Champion but they have NO news on this year’s championships. And you follow chessbase’s strategy to make every non-chessbase chess nonexistent. Are you afraid? I am sure you are not!
The HARDWARE was different, giving Junior a huge advantage. Junior had around twice the processing power, and was processing about 20 million nodes a second, compared to Fritz’s 13.
This was not a fair match anyway. The two programs were running on different H/W, so it was more of a match of H/W processing speed than a measure of quality of the programs themselves. Deep Junior was running on h/w with twice as many processors, running at much higher clock speed. No wonder it calculated twice as many positions per second than Fritz and saw more.
Why wasnt Rybka invited to this first oficial computer chess championship? Politics. Interests. There is a lot of politics in the world of chess.
And now Junior 10 will have an important sale increment. A lot of people will try to get the winner program, as it happened with Fritz 10 last year. ( probably myself will buy the Junior 10 too . :o)
One thing to note, Junior has been known to manipulate its opening library to steer opposing machines into positions that Junior is strong in. This trick to manipulate results gives Junior the impression of strength it does not have, and it melts away in a testing cauldron that contains a lot of games, enough to achieve statistically meaningful results. This should be considered for matches like this that are short and played on different hardware. These tests are manipulated and do nothing to determine the best engine, they should be ignored, they are nothing but publicity stunts meant to fool and manipulate you. Don’t let them get away with it!
Do you really believe if Rybka and Fritz competes at normal(a pentium 4, 3 GHz dual core machine with say 2 GB of ram) computing power, rybka will win? I would say FAT chance!
The newest version of Ryka is simply the best chess program ever designed. It is, and has proven itself to be, far superior in overall chess strength than any rival.
Both Junior and Fritz are inferior to Rybka. So, there is a reason that Rybka ins’t invited to the WCCC. It would easily win.
Rybka’s creator’s challenge was more than fair. Just go back and read it on this blog. They’ve not responded.
And to remind the person who questions why Susan posts the Junior-Fritz results and not the Computer WCC, read what it states on the first page of this blog: That she posts those things that interest her. If you want to follow the computer WCC go ahead. Plus, she gives every Saturday as an “open forum” for us. Any visitor to this site can enlighten the rest of us on a topic if it isn’t mentioned.
Remember, this is her blog and GM Polgar can post whatever she wishes on her own blog.
Pretty simple.
Ratings list exist. For example:
http://web.telia.com/~u85924109/ssdf/list.htm
To say, hardware WAS the reason for the loss is not so obviously clear. A better hardware translating to better moves, is something Fritz will have to prove. For now, Junior is the deserving “best program in the World”, unless the Fritz/Rybka team is able to prove otherwise. IMO, blaming hardware for the loss is just a “face saving” excuse from the Fritz team.
i thought hydra was the best am i wrong??
No. Hydra is best solely becouse it runs on rate of 216,000,000 positions per second.
That’s about 10 times more than Deep Junior and 20 times more than Deep Fritz.
Rybka,Junior,Fritz are all better programs than Brutus-Hydra assembly.
More likely Fritz could not make effective use of the extra processors if it were running on 16. I can’t believe that getting better hardware was out of reach for them, especially *renting* it for a few days.
Rybka can run on multiprocessors (as well as Junior) with no problem.
But assholes from UA know that their machine would comfortably lose against better programs and simply avoid match on equally powerfull hardwares.
Do you know when Hydra last time played in any kind of tournament?
Take a look at their site:
http://www.hydrachess.com
So pompose and irritating!