Author: Robert Hyatt
Date: 07:37:52 06/19/99
Go up one level in this thread
On June 19, 1999 at 00:55:01, Ed Schröder wrote: >>Posted by Robert Hyatt on June 18, 1999 at 20:16:44: >> >>>I found this message on the Rebel Site where the events of round six were >>>reported: >>> >>>Ed and I took the opportunity to play some 10 minute blitz games against Deep >>>Blue Junior. Amazingly Rebel crushed the IBM supermonster with 3-0! Deep Blue >>>Junior had no chance in these games, so one can have his doubts about the >>>playing strength of this machine. >>> >>>I haven't seen anyone else mention Deep Blue Junior. >>> >>>Roger >> >> >>Ed didn't tell "the rest of the story" so I will... >> >>I ask Hsu about this 'machine' a few weeks back, and here is what he told me >>about it: >> >>Some internal IBM folks asked him to develop a 'demo' facility to show off DB. >>He elected to do a web-based interface, which is "stateless" if you know what >>this is all about. In essence, this machine won't play a "game" at all, it >>simply takes a position, searches it for 1 second (which includes mostly the >>time needed to download the chess processors with the state information) and >>then it produces a move. No repetition testing at all, no game history, no >>nothing except for a near-instant search. However, it can take quite a while >>to make a move because _many_ web browsers get pointed at this thing by the >>IBM guys doing demos... >> >>Hsu estimated that it might play at 2200 or so. Which was all that was needed >>for the demonstrations it is used for. It is _not_ "deep blue junior" by any >>measure you would care to name. > >If this is true then why put such a thing on-line? > >I can't imagine a company like IBM putting a crappy version of their image >program for public consumption. It doesn't fit because such a thing would >damage the image they build beating Kasparov. Ed... simply email hsu and ask him. If you don't have his email, ask and I can email it to you. I doubt Hsu wanted to see it called "DB Junior" since it obviously isn't. I suspect the marketing folks thought that the "junior" shielded it from connection with the big iron version. But it does taint the 'DB Junior' name that was still producing performance ratings well over 2700 vs GM players... > >>And putting such nonsense on Ed's web page is _highly_ misleading. >> >>To say the least. >> >>Bob > >Deep Blue Junior is connected to the Internet in the hall of the museum >near to the tournament place of the 9th WCCC. You can play any time >control and we set Deep Blue Jr. to 5 minutes blitz and Rebel to 10 >minutes blitz. Rebel played on my laptop of 333 Mhz. It was surprisingly >to see that in the end Deep Blue Jr. used about the same amount of >time as Rebel for the games. > That is immaterial. But rather than me explaining this for a second time, simple contact Hsu. You will discover that (a) it searches for less than one second per move (computation time) no matter what you do; (b) it is a web- based stateless interface with no repetition or game history. There were other things 'disabled' to make the 'stateless' facility work... But ask Hsu for more details and to feel comfortable with the things I have reported. It only takes one email... >On the day before Chess Tiger played 2 games against Deep Blue Jr. >but then the machine played indeed on a 1 second per move level >which apparently is the default setting. Tiger won the first and lost >the second game but the game scores unfortunately were lost. Tiger >played on a Pentium 150 notebook. > >Christophe plans to play a few more games today against the monster >but I wonder if there is any time left for that as today we have to play >the last round, the blitz tournament and then move to the closing >dinner. > >I heard some rumors the blitz tournament will not take place because >the impossibility of some participants to move from one place to another >which would be a too big advantage for that team. But you never know >for sure in ICCA tournaments as things can change very easily by hour. > >I really loved the initial programmers meeting before the tournament. >When the issue of accelerated paring for the first rounds was discussed >the ICCA asked for a voting by all present participants and the majority >(90%) voted for a normal Swiss tournament which was accepted by the >ICCA. Head off for Jaap, Tony and David! > >As to the games Rebel played against Deep Blue Jr. I was shocked as >Rebel beat the monster on his own specialty namely tactics. Also the >machine played some horrible positional chess at times. If this is not >the real thing IBM better can put it off. Here are the games. See for >yourself. It isn't the 'real thing' unfortunately. This is just another effect of marketing types taking control of something for publicity, without regard for 'fact'... I don't think Hsu has had control since the _first_ DB vs Kasparov match... > >Ed Schroder > >--------------------------------------------------------------------------- > >[Event "Game-1"] >[Site "5 minutes blitz"] >[Date "1999.06.18"] >[Round "?"] >[White "Rebel 10.5"] >[Black "Deep Blue Jr."] >[Result "1-0"] >[WhiteElo "?"] >[BlackElo "?"] >[ECO "A50"] > >1. d4 Nf6 2. c4 d5 3. cxd5 e6 4. dxe6 Bxe6 5. Nc3 Bb4 6. Nf3 Qd6 7. a3 >Bxc3+ 8. bxc3 Qc6 9. Bb2 Qb5 10. Qb1 Nc6 11. e4 Qb6 12. Bd3 Na5 13. O-O >Nc4 14. Bc1 Na5 15. d5 Bd7 16. Be3 Qxb1 17. Raxb1 b6 18. Bf4 c6 19. Be5 >cxd5 20. exd5 Rc8 21. Ng5 Nxd5 22. Bxg7 Rg8 23. Rfe1+ Be6 24. Bh6 Ke7 25. >Bf5 Rc6 26. Bxh7 Rgc8 27. Be4 Nxc3 28. Bxc6 Nxb1 29. Bd5 Rh8 30. Nxf7 1-0 > >Here the game was interrupted and Deep Blue Jr. refused to play any further >whatever we tried. We also were unable to figure out how to play with the >black pieces so Rebel only played the white pieces in all 3 games. > >[Event "game-2"] >[Site "5 minutes blitz"] >[Date "1999.06.18"] >[Round "?"] >[White "Rebel 10.5"] >[Black "Deep Blue Jr."] >[Result "1-0"] >[WhiteElo "?"] >[BlackElo "?"] >[ECO "A56"] > >1. d4 Nf6 2. c4 c5 3. d5 d6 4. Nc3 e6 5. e4 Be7 6. Be3 O-O 7. h3 exd5 8. >cxd5 Qb6 9. Rb1 Qa5 10. Bd2 Bd7 11. Be2 Re8 12. f4 c4 13. Qc1 Bd8 14. Nf3 >Bb6 15. Rf1 Rc8 16. Qc2 Na6 17. a3 Re8 18. e5 dxe5 19. fxe5 Nxd5 20. Ng5 >g6 21. Rxf7 Be6 22. Rxh7 Bf5 23. Nxd5 Qxd5 24. Qxc4 Qxc4 25. Bxc4+ Be6 >26. Bb5 Bf5 27. Bxe8 Rxe8 28. Rc1 Nc5 29. g4 Rxe5+ 30. Kd1 Be6 31. Nxe6 >Kxh7 32. Nf8+ Kg8 33. Nxg6 Re6 34. b4 Ne4 35. Nh4 Nf2+ 36. Kc2 Nxh3 37. >a4 Nf2 38. g5 Nh3 39. Kb3 Re8 40. g6 Nf2 41. Nf5 Ng4 42. Rh1 Bd4 43. Nxd4 >Kg7 44. Rh7+ Kxg6 45. Rxb7 Ra8 46. Nc6 a6 47. Rb8 Rxb8 48. Nxb8 Kf5 49. >Nxa6 Ke6 50. b5 Kd5 51. a5 Ne5 52. Bb4 Nd3 53. Nc7+ Ke5 54. a6 Nxb4 55. >Kxb4 Kd6 56. a7 Kxc7 57. a8=Q Kd6 58. Qe4 Kd7 59. Kc5 Kc8 60. Kc6 Kd8 61. >b6 Kc8 62. Qe8# 1-0 > >[Event "game-3"] >[Site "5 minutes blitz"] >[Date "1999.06.18"] >[Round "?"] >[White "Rebel 10.5"] >[Black "Deep Blue Jr"] >[Result "1-0"] >[WhiteElo "?"] >[BlackElo "?"] >[ECO "C10"] > >1. e4 e6 2. d4 d5 3. Nc3 dxe4 4. Nxe4 Nf6 5. Nxf6+ gxf6 6. Nf3 Bd7 7. Bd3 >Bc6 8. O-O Qd6 9. Be3 Nd7 10. c4 O-O-O 11. a3 Rg8 12. b4 Ne5 13. d5 Bxd5 >14. cxd5 Qxd5 15. Be2 Qe4 16. Bd2 Bh6 17. g3 Qf5 18. Ra2 Nxf3+ 19. Bxf3 >Bxd2 20. Rxd2 Rxd2 21. Bxb7+ Kxb7 22. Qxd2 Qg5 23. Qd7 Rg7 24. Qe8 a6 25. >a4 Qd5 26. b5 a5 27. Re1 Qc4 28. Rd1 Qc2 29. Rf1 f5 30. Qf8 Qc3 31. Rd1 >h5 32. Rd8 Qe1+ 33. Kg2 Qe4+ 34. f3 Qe2+ 35. Kh3 Rxg3+ 36. Kxg3 Qe1+ 37. >Kh3 Qf1+ 38. Kh4 Qxf3 39. Kg5 Ka7 40. Qc5+ Kb7 41. Kf6 Qd5 42. Rxd5 exd5 >43. Ke7 1-0
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.