Video Chess, the First Chess Video Game?

[Atari2600Lives]

goldenband Thanks for the update and thread revive , read through the entire thing , fascinating stuff guys!

 

I have loved chess most of my life but as a kid I had a terrible experience with Atari 2600 chess when I went over to play Atari at my friends house and his dad was paying the chess game on the machine , needless to say 10 hours later the game was still gong on and I got on my bike and went home having never touched a controller .

[Retro Rogue]

Fascinating stuff, but the lawsuit story reportedly originated in the March 1983 edition of IEEE Spectrum, in which Larry Kaplan (who didn't program Video Chess, but I'm pretty sure did other great stuff, not just Bridge) told an anecdote (http://2600connection.atari.org/faq/faq_atarivcs.html#general19). Bob Whitehead and Larry Wagner had teamed up for Video Chess, but Whitehead says he knew of no such lawsuit, but heard repeatedly from corporate executives who said there had to be a chess cartridge because there was a chess piece pictured on a common version of the 2600 box.

 

There had to be a chess cartridge because there was a lawsuit over the chess piece being on the box and no chess game available. We have the full story in the book.

[+cvga]

Easy to learn how to play Chess.. getting good at it takes a little longer.

 

 

 

I can't believe I just saw this thread. My 17 year old never showed any interest in even learning how to play but my 6 year old has been playing for almost a year. He plays against himself on a real chessboard when I'm busy. I need to have him play the atari

[ccc---]

I wouldn't underestimate Videopac (Odyssey2) Chess. The C7010 Chess module is basically a complete Chess computer and uses the G7000 (Odyssey2) just as input/output device basically.

 

Would be great if goldenband could play that too. I'd be very curious how it compares to VCS and Intellivision Chess.

[+thegoldenband]

goldenband Thanks for the update and thread revive , read through the entire thing , fascinating stuff guys!

 

Thanks! Sorry about your traumatic childhood experience with Video Chess, though -- I suspect that it would've traumatized me if I'd tried to take it on as a kid in real time.

 

There had to be a chess cartridge because there was a lawsuit over the chess piece being on the box and no chess game available. We have the full story in the book.

 

Hey, I'm glad to hear that story's not an urban legend, and that you've been able to document it.

 

I can't believe I just saw this thread. My 17 year old never showed any interest in even learning how to play but my 6 year old has been playing for almost a year. He plays against himself on a real chessboard when I'm busy. I need to have him play the atari

 

And the Intellivision too! I actually think that port offers a more humane experience, though on the highest difficulty setting it's still quite absurd.

 

I wouldn't underestimate Videopac (Odyssey2) Chess. The C7010 Chess module is basically a complete Chess computer and uses the G7000 (Odyssey2) just as input/output device basically.

 

Would be great if goldenband could play that too. I'd be very curious how it compares to VCS and Intellivision Chess.

 

I'd love to have a copy I could play, but my understanding is that it's not compatible with NTSC systems. Is that correct, or is there a way to make it work? Unfortunately I'm on an older Mac and I don't think the version of O2Em available to me can run Chess (it doesn't even support the Voice).

[Mr SQL]

I wouldn't underestimate Videopac (Odyssey2) Chess. The C7010 Chess module is basically a complete Chess computer and uses the G7000 (Odyssey2) just as input/output device basically.

 

Would be great if goldenband could play that too. I'd be very curious how it compares to VCS and Intellivision Chess.

Seems the O2 hardware could have handled a chess port on it's own, surprised they went that route. Anyone have any idea why??

[SoulBlazer]

You should try the newest Chessmaster game availble on XBox Live. It seems to be quite good.

[ccc---]

I'd love to have a copy I could play, but my understanding is that it's not compatible with NTSC systems. Is that correct, or is there a way to make it work? Unfortunately I'm on an older Mac and I don't think the version of O2Em available to me can run Chess (it doesn't even support the Voice).

 

Unfortunately the C7010 Chess module will not work on NTSC consoles (Odyssey2), because of mid-screen VDC updates. And so far O2EM doesn't support it either ...

[ccc---]

Seems the O2 hardware could have handled a chess port on it's own, surprised they went that route. Anyone have any idea why??

 

More power probably.

 

The C7010 has an ELO rating of 1650. Is there any info on the ELO rating of other Chess video games?

[Mr SQL]

More power probably.

 

The C7010 has an ELO rating of 1650. Is there any info on the ELO rating of other Chess video games?

ccc,

that's pretty good but I would think a dedicated unit could do better than that even for the timeframe, and also that the O2's native hardware is sufficient for a decent port like it's competitors (192 bytes of RAM vs 128 on the Atari and just 64 on the Fairchild).

 

Having played VCS Chess I'm tremendously impressed with the engine (perhaps 1400?) ; indeed it's got better game play than many other 4k Chess programs such as the Colour Chess cartridge for the CoCo.

 

Chessmaster 2000 on the PC and the C64 from the 80's is awesome and will beat most Chess engines today; that engine is indeed close to a 2000 ranking, perhaps 1900. Later editions (chessmaster 3000 and beyond) did not improve much upon it.

 

And while we're at it Deep Blue is over rated and morphed into a modern day multiplayer networked version of the Turk; It lost to the PC chess games Fritz in 1995 despite running on a mainframe so the 1989 match against Kasparov was obviously a joke. Then in the late 90's, it seems to me IBM either paid Kasparov to throw the match the third time around or just plain cheated - Kasparov felt there were a bevy of high ranking human opponents assisting the machine and asked for a rematch and to see previous games deep blue had played to analyse for signs of human intervention - IBM responded by immediately dismantling Deep Blue

[+Gemintronic]

IBM was simply following the best practices of query optimization:

 

Filter as much as possible

Obviously the hacked in expert tree based on high ranking human opponents can be removed.

 

Select only the fields you need

Previous games are redundant since Kasparov won

 

Be careful with joins

Dismantling Deep Blue is the obvious choice for this one!

[Mr SQL]

IBM was simply following the best practices of query optimization:

 

Filter as much as possible

Obviously the hacked in expert tree based on high ranking human opponents can be removed.

 

Select only the fields you need

Previous games are redundant since Kasparov won

 

Be careful with joins

Dismantling Deep Blue is the obvious choice for this one!

LOL Loon, you crack me up!

 

IBM actually made outrageous claims tanamount to having invented a neural network.

 

In all fairness the Turk chess machine invented in 1770 did contain a neural network and as Deep Blue appears to be an upgraded networked version of the Turk it would include multiple neural networks running in parallel

 

http://en.wikipedia.org/wiki/The_Turk

 

Another possiblity is that IBM had genuinely created an artificial neural network but then dismantled it (after the stock soared) so they could send it hurtling forward in time to one day run the Enterprise thereby turning Kirk into a Dunsel; I'm going with that one

[Tin_Lunchbox]

 

<text clipped for brevity>

 

Then in the late 90's, it seems to me IBM either paid Kasparov to throw the match the third time around or just plain cheated - Kasparov felt there were a bevy of high ranking human opponents assisting the machine and asked for a rematch and to see previous games deep blue had played to analyse for signs of human intervention - IBM responded by immediately dismantling Deep Blue

 

I remember that. Kasparov felt that he was playing not really a computer designed to play the best possible game of chess, but rather a computer designed to beat Kasparov, in other words to recognize Kasparov's various personal stratagems and defenses and pre-programmedly react to those. It's hard for me to say whether that would be "cheating." I guess if they're coding Deep Blue's subroutines with remarks like "IDENTIFY, DEFEAT KASP. GAMBIT 06" I would be critical of that.

[Mr SQL]

I remember that. Kasparov felt that he was playing not really a computer designed to play the best possible game of chess, but rather a computer designed to beat Kasparov, in other words to recognize Kasparov's various personal stratagems and defenses and pre-programmedly react to those. It's hard for me to say whether that would be "cheating." I guess if they're coding Deep Blue's subroutines with remarks like "IDENTIFY, DEFEAT KASP. GAMBIT 06" I would be critical of that.

Tin_Lunchbox,

the support team was networked to Deep Blue just like the Turk above. They were supposed to just tweak it between games; clearly they couldn't resist tweaking it during the game with helpful input and were positioned to do exactly that.

 

The guy inside the Turk looks like he's sitting there too but looking at the design no one would believe he's not participating - Deep Blue has the same design but we are to believe they used the honour system despite that Kasparov was able to identify the points in the game where humans assisted the machine.

 

Then we have IBM's tales of the fantastic breakthrough milestones to contend with - the ideas you pointed out of "recognising various personal stratagems and defenses" in Chess is descriptive of a learning neural network which is what made it such an outrageous claim! It's clear IBM did create a neural network but their actions and the evidence all point back to the design.

 

Here's some Deep Blue techno babble from IBM claiming the Deep Blue technology has now been reassembled and expanded to take it to the next level by Winning at Jeapordy! But that sounds like Google and not the neural network (artificial or otherwise) that beat Kasparov.

[Tin_Lunchbox]

<stuff clipped for brevity>

the support team was networked to Deep Blue just like the Turk above. They were supposed to just tweak it between games; clearly they couldn't resist tweaking it during the game with helpful input and were positioned to do exactly that.

 

You have evidence or a solid source that Deep Blue's support team tweaked its play while a game was in progress?

[Mr SQL]

You have evidence or a solid source that Deep Blue's support team tweaked its play while a game was in progress?

Yes, Kasparov After passing the Turing test IBM refused to provide any evidence to support their outrageous claims of building an artificial neural network.

[+thegoldenband]

It's been a while since I've brawled with a console chess program, but I had a win over SNES Chessmaster today. It's basically a reskinned version of NES Chessmaster, but the underlying engine is actually one of the stronger ones I've seen, at least in a cart-based console. Even Level 1 can put up a tough fight, and after losing my first 2-3 attempts, I drew the next game and then won this one.

 

Anyway, here's the game. It starts out with a bizarre opening (to take the computer out of book), but transposes pretty smoothly into a Benoni-style position. Notable mistakes from the computer include 4. Bxa6? giving up the two bishops, 8. Ng5? wasting time, 12. Qe2? giving up his dark-squared bishop, and 16. Nb1?? which is the culmination of a series of bizarre errors that drop a full rook.

 

[Date "2014.07.01"]

[White "SNES Chessmaster, Level 1 (60 moves in 5 minutes)"]

[black "thegoldenband"]

[Result "0-1"]

1. f4 Na6 2. Nc3 c5 3. e4 d6 4. Bxa6 bxa6 5. Nf3 Nf6 6. d4 Bb7 7. d5 g6 8. Ng5 Bg7 9. O-O h6 10. Nf3 Rc8 11. Be3 Ng4 12. Qe2 Nxe3 13. Qxe3 Qb6 14. b3 c4 15. Qxb6 axb6 16. e5 cxb3 16. Nb1 b2 17. Nd2 bxa1=Q 18. Rxa1 Rxc2 19. Re1 dxe5 20. fxe5 Bxd5 21. a3 O-O 22. Rb1 b5 23. Re1 Rfc8 24. Rd1 Rc1 25. Rxc1 Rxc1+ 26. Kf2 Rc2 27. Ke3 Bxf3 28. Nxf3 Rc3+ 29. Kd4 Rxa3 30. Nd2 Ra2 31. Kd3 Rxd2+ 32. Kxd2 Bxe5 33. g3 a5 34. Kd3 a4 35. Kc2 b4 36. Kb1 b3 37. Kc1 a3 38. Kd2 a2 39. Kd3 a1=Q 40. Kc4 b2 41. Kd5 b1=Q 42. Kc6 Qa5 43. Kd7 Qb7+ 44. Ke8 Qaa8#

[+thegoldenband]

And here's a victory on Level 2. Once again, I promoted three pawns to Queens!

 

[Date "2014.07.01"]

[White "SNES Chessmaster, Level 2 (40 moves in 5 minutes)"]

[black "thegoldenband"]

[Result "0-1"]

1. c4 e5 2. Nc3 f5 3. e4 d6 4. d4 Nc6 5. Nf3 fxe4 6. Nxe4 Nf6 7. Nxf6+ Qxf6 8. Bg5 Qg6 9. Be3 e4 10. Ng1 Be7 11. Ne2 O-O 12. Nf4 Qf5 13. Nd5 Bd8 14. Be2 Ne7 15. Nxe7+ Bxe7 16. O-O Bg5 17. Bxg5 Qxg5 18. Qb3 Kh8 19. Qe3 Qxe3 20. fxe3 Bd7 21. g3 Rae8 22. Kg2 c6 23. Rf4 Rxf4 24. exf4 d5 25. Re1 Kg8 26. cxd5 cxd5 27. Bh5 Re7 28. Kf2 g6 29. Bd1 Kf7 30. Bb3 Bc6 31. Ke3 Ke8 32. g4 Kf7 33. f5 Kf6 34. fxg6 Kxg6 35. h4 Rf7 36. Bd1 Bd7 37. g5 h5 38. Rg1 Bb5 39. Rg3 Rf1 40. Bc2 Rh1 41. Bb3 Bc6 42. Rg2 Rxh4 43. Bc2 Rg4 44. Rxg4 hxg4 45. Kf4 Bd7 46. Bb3 Be6 47. Ke5 e3 48. Bd1 g3 49. Bf3 e2 50. Bxe2 g2 51. Kxe6 g1=Q 52. Bd3+ Kxg5 53. Kxd5 Qe3+ 54. Bc4 Kf6 55. Bb3 Qd2 56. Kc5 Qxb2 57. d5 Qa3+ 58. Kd4 Ke7 59. Ke4 Kd6 60. Kf5 a5 61. Ke4 b5 62. Kf5 a4 63. Bc2 Qxa2 64. Be4 Qxd5+ 65. Bxd5 Kxd5 66. Kf4 Kd4 67. Kf5 a3 68. Ke6 a2 69. Kd6 a1=Q 70. Kd7 Qa6 71. Ke7 b4 72. Kf7 b3 73. Ke7 b2 74. Kf7 b1=Q 75. Ke8 Qbb7 76. Kd8 Qaa8#

[davyK]

Sargon 1 and 2 were on the Apple II - released '79 - guess they are a bit too late to beat Video Chess.

 

 

(EDIT: sorry - just noticed this is an old thread.....)

 

 

 

 

Edited July 2, 2014 by davyK

[+thegoldenband]

(EDIT: sorry - just noticed this is an old thread.....)

 

 

Nothing to apologize for, I'm the one who keeps reviving it!

[Bill Loguidice]

Pretty sure Sargon Chess was 1978 on at least the TRS-80. MicroChess 1.5 was also 1978 on the TRS-80. MicroChess 2.0 was the PET version. The higher version numbers were to differentiate from the non-graphical KIM-1 version, which was 1976. It's probably safe to say that the Atari 2600 had the first console version of chess. I certainly can't think of any other before it.

[+thegoldenband]

I've set aside SNES Chessmaster for a while, but I've spent some time this week battling Chessmaster for Game Boy, and ended up defeating it on the highest difficulty (Level 13) today with the Black pieces. Here's the game score, with animated GIF at the end:

 

[Date "2014.08.21"]

[White "Game Boy Chessmaster, Level 13"]

[black "thegoldenband"]

[Result "0-1"]

1. e4 e5 2. Nf3 Nc6 3. Bb5 Nd4 4. Nxd4 exd4 5. O-O g6 6. Na3 Bg7 7. f3 Ne7 8. d3 O-O 9. Bf4 d5 10. Qd2 Kh8 {see note} 11. Qa5 c6 12. Bc7 {the game tried to play 12. Bb5-c7, an illegal move!} Qd7 13. Ba4 b6 14. Bxb6 axb6 {the game crashed here and I had to set the position up from scratch} 15. Qxa8 Bb7 16. Qa7 Qc7 {game crashed again!} 17. exd5 Nc8 18. d6 Nxd6 19. Bxc6 Qxc6 20. Nc4 Nxc4 21. dxc4 Ra8 22. Qxa8+ Bxa8 23. b3 Bf6 24. h4 Bxh4 25. Rad1 Bf6 26. a4 Qc5 27. Rf2 Bg5 28. f4 Bh4 29. Rfd2 Bf6 30. a5 Qxa5 31. Kf2 Bc6 32. Re2 Qf5 33. Kg1 Kg7 34. Rf2 b5 35. cxb5 Bxb5 36. b4 h5 37. Rb1 d3 38. c4 Bxc4 39. Rc1 Bd4 40. Rxc4 Qxf4 41. Rxd4 Qxd4 42. b5 d2 43. g3 d1=Q+ 44. Kg2 Qe4+ 45. Kh3 Qdh1+ 46. Rh2 Qg4#

 

This opening (Bird's Defense to the Ruy Lopez) worked over and over against the computer, though on lower levels it would skip 7. f3, playing 7. d3 instead, and reach the key position one move earlier. That's why I deliberately wasted time with 10...Kh8, enticing the computer to play the misguided 11. Qa5? (which was 10. Qa5? in the earlier games). It thinks that after 12. Bc7?? Qd7 13. Ba4 b6 14. Bxb6 axb6 15. Qxa8 it'll win Rook + Pawn for a Bishop, but fails to see that the Queen is trapped and, despite seven moves worth of squirming, will be lost. After that it's just a matter of careful play to see the advantage through.

 

BTW I'm not sure whether it's the program itself or just a problem with my copy, but this game is seriously buggy. On later levels it'd crash once or twice per game, forcing me to set the position back up from scratch, and on two occasions it made flat-out illegal moves, including 12. Bb5-c7?!?! in the above game, ending up with two dark-squared bishops! Since it played 12. Bf4-c7 in every other game, I felt comfortable in inputting that move for the computer. Some routine saying "move the bishop!" must've bit-flipped and pointed to the wrong one, I guess.

 

[SoundGammon]

Computers don't "play" the same way as humans. A human player starts with a plan from the beginning and works forward with it. Whereas a computer thinks from the last move on, changing the strategy from that point in the game and then again after the next move. A human player would work to be in a certain position, say, in mid-game and be there, but the computer would not, unless it was following an already played or set game.

[Keatah]

Does anyone have a screenshot of what it looks like in black and white?

 

[+thegoldenband]

Computers don't "play" the same way as humans. A human player starts with a plan from the beginning and works forward with it. Whereas a computer thinks from the last move on, changing the strategy from that point in the game and then again after the next move. A human player would work to be in a certain position, say, in mid-game and be there, but the computer would not, unless it was following an already played or set game.

 

I'm not sure whether you're replying to something I (or someone else) said, but FWIW the idea of "start[ing] with a plan from the beginning and work[ing] forward with it" doesn't really describe most high-level human play nowadays. A lot of early chess literature emphasized the idea of deep planning, true, but more recent writers like John Nunn and John Watson have observed that grandmaster play is more dynamic and ever-changing than that (especially since the 1950s).

 

Conditions on the board just change too rapidly for any player to expect to consistently see through preconceived goals; of course it happens sometimes, but short-term, dynamic/tactical considerations generally override any long-term plan. Naturally, some of this is the influence of computers, and all top grandmasters these days train with programs like Fritz, Rybka, and Shredder (or whatever's most popular nowadays).

 

As for why Chessmaster headed for the same position for 5-6 games in a row, the first five moves were part of its "opening book". After that, it just seems to have a heavily deterministic algorithm, which is a major flaw in a consumer chess program. The games I played on Levels 9 and 13 were identical for the first 22 moves!