Interfacing the Novag Citrine

[IanO]

First, I haven't gotten xboard to talk to the Citrine -- yet.

However, manual testing shows that the pieces on the board do not have to be moved when controlling the Citrine over the serial link. The obvious exception is when using referee mode.

Furthermore, the Citrine can be set to run in Auto or Demo mode where it plays against itself without the need for moving the pieces. The last game played can be retrieved over the interface. Well, at least the last 112 ply can be retrieved.

Sadly, the "S" (show running information) command implemented in some Novag models does not work. The Citrine LCD module shows this data, but it can't be accessed over the serial link.

None of the S1, S2, S3, or S4 commands work? Bummer.

Was the AutoNovag project I published last week of any help? It was sufficient to autoplay under WinBoard; it shouldn't be hard to adapt it to the more sane Unix-style serial devices.

Ian

Ian

[Karmazen & Oliver]

Thanks.

I've also located the undocumented "J" command that tells the Citrine to start searching for a move. I don't think there are any more secrets to be discovered.

----

No response yet from FICS on getting an account for the Citrine.

----

The Citrine is actually kind of fun to play, even for a fish like me who has to use the easy levels. Not too much playing time is needed to become acclimated to the piece capture movement implementation. The auto sensory experience is to me far superior to square tapping, and I think this is more of a deciding factor than just having wood pieces or a "wooden" decal board.

Novag advertises the Citrine's strength at 2330 elo but its experiences with Symbolic show that Steve B's number of 2048 elo as being rather more accurate. Then again, the latter figure is measured versus computer opponents; perhaps a higher figure is called for when playing against human adversaries.

My Cookie Cat is quite unhappy with this new competitor for my attention. I keep the pieces safely bagged when not in use so they might escape her nocturnal predations.

ehh ¡ a lot of people said during a lot of moths that citrine have near 2100 points... and I... ( oliver ) was the first that "do the test BT-2650-2450 and do a extesive review...

the conclusion was. CITRINE have 2000~2025 points... after this... the firts evaluation a lot of optimist 2100, down to 2080... and now 2049... to do that ELO ratings use a rare games versus "other" computer with have a "sobrevalorate" elo ratings...

I say... before and NOW that citrine have 2000~2025... NO MORE.

bye. from Spain. OLIVER.

[sje]

None of the S1, S2, S3, or S4 commands work? Bummer.

Was the AutoNovag project I published last week of any help? It was sufficient to autoplay under WinBoard; it shouldn't be hard to adapt it to the more sane Unix-style serial devices.

Yes, I tried all the Sn commands. I might retry some items without the Citrine LED module in line to see if that might make a difference.

My intermediate goal is to get the Citrine to talk with Lisp directly; success here would save a lot of time and effort in the long run. My hope is that this can be done without having to fiddle with system calls to set the line attributes.

[sje]

I say... before and NOW that citrine have 2000~2025... NO MORE.

If I can get a FICS account for the Citrine, then we will soon thereafter have more data.

[sje]

Here's the Citrine playing against itself at 40/9000:

Code: Select all

[Event "Autoplay test"]
[Site "At home"]
[Date "2008.10.17"]
[Round "-"]
[White "Novag Citrine"]
[Black "Novag Citrine"]
[Result "0-1"]
[TimeControl "40/9000"]

1. d4 Nf6 2. c4 e6 3. Nf3 b6 4. a3 Bb7 5. Nc3 d5 6. cxd5 Nxd5 7. Qc2 Nxc3 8.
bxc3 Qd5 9. e3 Nd7 10. Bd3 Nf6 11. O-O Be7 12. e4 Qh5 13. Qa4+ Kf8 14. Re1 Kg8
15. Rb1 a6 16. Bg5 h6 17. Bxf6 Bxf6 18. Qd7 Rc8 19. Rec1 Bg5 20. Rxb6 cxb6 21.
Qxb7 Rc5 22. Qb8+ Kh7 23. Qxb6 Rhc8 24. Nxg5+ Qxg5 25. Qb2 R5c6 26. e5+ Kh8 27.
Rb1 Qd8 28. c4 Rxc4 29. Bxc4 Rxc4 30. Rd1 Qc7 31. f3 Rc3 32. Qb4 Kh7 33. Rd2 a5
34. Qb1+ g6 35. Qb2 Qc4 36. Kf2 Kg7 37. Kg3 Rb3 38. Qa2 Qa4 39. Rc2 g5 40. h3
Rxa3 41. Qc4 Qxc4 42. Rxc4 a4 43. h4 Ra2 44. hxg5 hxg5 45. Rb4 Kg6 46. Rc4 Ra1
47. Rc5 Rd1 48. Rc4 a3 49. Ra4 Rd3 50. Ra7 Rb3 51. Ra8 Rc3 52. Ra7 Rd3 53. Ra4
Kf5 54. Kf2 Kf4 55. Ke2 Rb3 56. Kd2 Kg3 57. Kc2 Re3 58. Kd2 Kf2 59. Ra5 a2 60.
Kc2 Kxg2 61. Kd1 a1=R+ 62. Rxa1 Rxf3 63. Ke2 g4 64. Ra8 Rf2+ 65. Ke3 g3 66. Rc8
Rf3+ 67. Ke2 Rf2+ 68. Ke3 Rf3+ 69. Ke2 Rf4 70. Ke3 Rf2 71. Ra8 Kg1 72. Rh8 g2
73. Rc8 Rf5 74. d5 Rxe5+ 75. Kf3 Rxd5 76. Ra8 Rd3+ 77. Ke2 Rh3 78. Ra1+ Kh2 79.
Kf2 Rb3 80. Re1 e5 81. Rd1 f5 82. Rc1 f4 83. Ke2 g1=R 84. Rxg1 Kxg1 85. Kd2 e4
86. Kc1 Rd3 87. Kc2 f3 88. Kc1 f2 89. Kc2 f1=Q 90. Kb2 Qe2+ 91. Kc1 Rd1# 0-1

The above took about eleven hours; there were no glitches with the Citrine, the cabling, or the monitoring software.

White actually resigned at move 77 but played on for testing purposes.

[Steve B]

Novag advertises the Citrine's strength at 2330 elo but its experiences with Symbolic show that Steve B's number of 2048 elo as being rather more accurate.

actually my number is the number reported by Selective Search Magazine in their most recent issue
Selective Search has been rating dedicated computers since 1985 and is highly respected by collectors worldwide
of course their main emphasis Today is rating Pc Engines
http://www.elhchess.demon.co.uk/sscontnt.htm

it should also be remembered Steve ,that Novag's rating estimate of 2330 is a USCF estimate and not an ELO estimate so right there we need to deduct 125-150 points to get a comparable ELO figure
when Le Citrine first came out(2006) Selective Search showed the Citrine as low as 2025 ish or thereabouts.. but over the last year or so the rating seems to have settled into the mid 2000 level

will be very interesting to see how the rating settles in on ICC
will you be limiting your challenges on ICC to engines rated <2200 (or some other figure)?

if you let the top engines running on quads and octals play Le Citrine the rating will not be pretty

Concerned Regards
Steve

[sje]

will be very interesting to see how the rating settles in on ICC
will you be limiting your challenges on ICC to engines rated <2200 (or some other figure)?

if you let the top engines running on quads and octals play Le Citrine the rating will not be pretty

Well, I still have to write the glue software and then get an account. FICS has not yet responded, and I'm not interested in paying for an ICC account just for the machine.

I will probably set the match formula to +/- 400 points from the machine's rating. This will let many humans match the Citrine; I won't leave the window wide open as that tends to attract 1200 elo humans that somehow know openings 30 ply deep and also play five man endgames perfectly. (Do they really think they're fooling anyone?)

The match formula will also have to be restricted so that the only time controls accepted are those that the Citrine can understand.

If I can get my old Novag Super VIP to talk to Lisp, I might also get that machine connected to an ICS.

[Steve B]

I won't leave the window wide open as that tends to attract 1200 elo humans that somehow know openings 30 ply deep and also play five man endgames perfectly. (Do they really think they're fooling anyone?)

it never ceases to amaze me how rampant the cheating is on all of the servers
i occasionally like to play unrated games on my Playchess account in the CAFE forum ...and even unregistered guests will cheat with engines in UNrated games
some how ..someway.. these folks get some form of satisfaction on seeing their temporary guest accounts win an UNRATED game with an engine instead of playing themselves

beyond belief regards
Steve

[sje]

it never ceases to amaze me how rampant the cheating is on all of the servers

I don't care so much about the cheating as I care about those who cheat only when playing against programs. This screws up the program's rating and the errors can only be corrected over time by distributing the rating damage to honest opponents.

[sje]

About eight hours of calculations:

Code: Select all

[Event "Autoplay test"]
[Site "At home"]
[Date "2008.10.17"]
[Round "-"]
[White "Novag Citrine"]
[Black "Novag Citrine"]
[Result "0-1"]
[TimeControl "40/9000"]

1. c4 e5 2. Nc3 Nc6 3. Nf3 Nf6 4. g3 Bb4 5. Bg2 O-O 6. O-O e4 7. Ng5 Bxc3 8.
bxc3 Re8 9. Qc2 d5 10. cxd5 Qxd5 11. d3 Bf5 12. Qb2 Rab8 13. Be3 Qd7 14. Bxe4
Bxe4 15. Nxe4 Nxe4 16. dxe4 Rxe4 17. Rfd1 Qe8 18. Rab1 b6 19. Qc2 a6 20. Qd3
Ra8 21. Qd7 Re7 22. Qxe8+ Raxe8 23. Kf1 f6 24. h3 Kf7 25. Kg2 Ke6 26. a4 Rd7
27. Rxd7 Kxd7 28. Rd1+ Kc8 29. Kf1 Re5 30. c4 Ra5 31. Ra1 Ne5 32. Bd2 Rc5 33.
a5 Nxc4 34. Bb4 Rb5 35. Bf8 Rxa5 36. Rxa5 Nxa5 37. Bxg7 f5 38. g4 fxg4 39. hxg4
Nc4 40. Kg2 a5 41. e4 b5 42. Kf3 a4 43. Ke2 a3 44. Kd3 c5 45. f4 Kd7 46. Kc3
Na5 47. Kc2 b4 48. Kb1 b3 49. Bc3 Nc4 50. g5 Nd6 51. e5 Ne4 52. Ba5 Ke6 53. Ka1
Kf5 54. Be1 a2 55. Ba5 c4 56. Bb4 c3 57. Ba3 Nd2 58. e6 Nf3 59. Bb2 Nd4 60.
Bxc3 Nc2+ 61. Kb2 a1=Q+ 62. Kxb3 Qb1+ 63. Bb2 Nd4+ 64. Ka3 Nb5+ 65. Kb3 Kxe6
66. Kc4 Qxb2 67. Kd3 Qc3+ 68. Ke2 Nd4+ 69. Kf2 Qf3+ 70. Kg1 Ne2+ 71. Kh2 Qg3+
72. Kh1 Qh3# 0-1