Deep Junior – Deep Fritz
Fritz vs Junior Match Elista, 2007
1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6 6.Bg5 e6 7.f4 Qb6 8.Qd2 Qxb2 9.Rb1 Qa3 10.e5 h6 11.Bh4 dxe5 12.fxe5 Nfd7 13.Ne4 Qxa2 14.Rd1 Qd5 15.Qe3 Qxe5 16.Be2 Bc5 17.Bg3 Bxd4 18.Rxd4 Qa5+ 19.Rd2 0-0 20.Bd6 Re8 21.0-0 f5 22.Qg3 Nc6 23.Qg6 Qd8 24.Bc4 Kh8 25.Ng5 Qxg5 26.Qxe8+ Kh7 27.Bf4 Qg6 28.Qxg6+ Kxg6 29.Bxe6 Nf6 30.Bc4 Kh7 31.Bc7 a5 32.Ra1 a4 33.Bb5 Ne4 34.Rd3 Nb4 35.Rd8 a3 36.Bd7 Nxc2 37.Rf1 Bxd7 38.Rxa8 Be6 39.Be5 Ng5 40.Rf2 Bb3 41.Rxf5 Kg6 42.Rf1 a2 43.Ra7 Ne3 44.Rxb7 Bd5 45.Rxg7+ Kh5 46.Rc1 Nxg2 47.Kf2 Nh3+ 48.Ke2 Nh4 49.Ba1 Be4 50.Rc4 Bb1 51.Rg3 Ng5 52.Rb3 Nf5 53.Kd1 Ne4 54.Kc1 Kg5 55.Rf3 Kg4 56.Ra3 Kf4 57.Rc8 Nf2 58.Rcc3 Ne4 59.Rf3+ Kg4 60.h3+ Kg5 61.Ra5 Ned6 62.Ra4 Ne4 63.Rb4 Nfd6 64.Kb2 Nf5 65.Rb5 Ned6 66.Rc5 Nb7 67.Rc4 Nbd6 68.Rg4+ Kh5 69.Rgf4 Kg6 70.Kc1 Kg5 71.Rg4+ Kh5 72.Ra4 Kg6 73.Rff4 Nc8 74.Ra5 Ncd6 75.h4 Kf7 76.Re5 Kg6 77.Kb2 Kf6 78.Kb3 Kf7 79.Ra4 Kg6 80.Ra6 Kf7 81.h5 Kf6 82.Kb4 Kf7 83.Ra7+ Kf6 84.Rd7 Kg5 85.Rxd6 1-0
Junior is better.
This has to be one of the best computer vs computer games I’ve seen. Deep junior displays so much activity in exchange for the three pawns!
I thought the game was very interesting as well. Fritz played some stereotypical computer chess, materialistically gobbling up no less than three poisoned pawns. Kasparov has said that computers are unable to correctly assess positions with uneven material, for Fritz this seems true, but Junior might represent the next generation of computer chess. The endgame was also interesting, with Fritz seeming to wander around while Junior steadily gained ground. Another computer weakness being inability to plan, computers being unable to understand concepts such as fortresses.
Now I might sound critical of computer chess, so I want to offer something positive – I don’t think any human could sacrifice three pawns against Fritz and convert it into a win. Only a program like Junior could be accurate enough to hold onto the advantage. In some ways Junior’s play (with some decidedly uncomputer like sacrifices) was more human than a human!
The first 26 moves of this game were already known from Shirov-Ftacnik, German Bundesliga. Nice win by Junior, but it was not Junior who came up with the 3-pawns sacrifice idea. That was Radjabov.
I did not know the triple gambit idea was human, which makes it more believable, but no less impressive. I wonder at which point Deep Junior was able te measure acurately his compensation. In any case, being able to hold such an advantage against Deep Fritz, as bob hu pointed out, says a lot about how far chess software has come.
So this proves that the Sicilian, poisoned pawn variation is not so drawish anymore? Maybe computer chess will prove that there is more to be discovered in this variation.
No bias intended, but greater processing power could be helping Junior. FIDE should have stipulated similar hardware characteristics to make an objective assessment of the outcome of the match possible.
Deep Fritz is running on an eight-core machine and searching 13-14 million nodes per second, reaching a search depth of 20-21 ply. Deep Junior is employing the latest Intel Server technology with 16 cores. The program is running at 24 million nodes per second and consistantly reaching search depths of 24 ply. More on the technology and the match tomorrow, a rest day for the humans but a full round for the computers.
Notes by Alex Kure of the Deep Fritz team
The engines are playing on completely different hardware. This combined with the few games to be played in this match means that we can conclude NOTHING from the result of this match!
If thats the case…..8 core..vs 16 core…um..whats the point of this match? It says absolutely nothing about the software since they’re competing from different levels. Its sort of weird…like one of them is fighting with an arm tied behind their back.
Oh well…..least nobodys feelings are getting hurt.haha.
And now that it’s 3-1 to JUNIOR, I expect the FRITZ team to go into damage-limitation mode.
There are pros and cons about a free choice of hardware. On the upside, this improves the quality of the chess. However, this means that less to nothing can be said about the relative quality of the engines.
At the cost of chess-quality, which is zero-cost to me as I would not miss it anyway, I’d prefer to see a genuine software contest.
I would also like to see a 24-game match, which would give stats twice as accurate. The machines should be switched on left to play the match ’round the clock’ with minimal ‘pit stops’ in between games.
g
I would also like to see, particularly in the case of computer-computer games, resignations delayed so that I, and thousands of other scholastic players, could see the win more clearly.
It’s less clear in Game 4 than in Game 3.
At the cost of a little more pain at the board, there’s a major teaching/learning opportunity here.
g
An unfair contest between the junior and fritz. Now it is difficult to judge which is a better engine. I prefer same hardware for both the programs and 24 matches bewteen them.
Additionally i would also want to know the result in rapid matches to judge which algorithm is better in search and evaluation techniques.