Novag Obsidian vs Saitek Travel Champion 2100

Wardy · Post by **Wardy** » Mon Nov 11, 2013 7:51 pm

I get the feeling that the next few moves will steer the result of the game.

TC is holding up well with 3 moves per cycle of 2400mAh rechargables basically matching the performance of alkaline batteries according to the manual and still having life in them too.

Though I do hope the next match is on mains power

sje · Post by **sje** » Mon Nov 11, 2013 8:34 pm

After 15 minutes: (-0.78) 19 Re2

Wardy · Post by **Wardy** » Tue Nov 12, 2013 6:24 am

Obsidian - TC2100
3r1rk1/1bq1bp1p/p5p1/1p1PR3/2n5/3Q1N1P/PPB2PP1/R1B3K1 w - - 0 1

Analysis by Deep Fritz 12:

1. -/+ (-1.00): 19.Re1 Bf6 20.Bh6 Bxb2 21.Rab1 Rxd5 22.Qb3 Bg7 23.Bxg7 Kxg7 24.Be4 Rc5 25.Bxb7 Qxb7 26.Qc3+ Kg8 27.Rbd1 Rc6 28.Ng5 Qc7 29.Qd4 h6 30.Ne4
2. -/+ (-1.16): 19.Re2 Bf6 20.Bh6 Nxb2 21.Qd2 Bxd5 22.Bxf8 Bxf3 23.Bh6 Nc4 24.Rae1 Bxe2 25.Qxe2 Bg7 26.Bg5 Rc8 27.Bb3 Bc3 28.Rd1

(Ward, 12.11.2013)

sje · Post by **sje** » Tue Nov 12, 2013 6:50 pm

It finished at least 1 1/2 hours early this time:

[Event "Correspondence Game"]
[Site "http://hiarcs.net/forums/viewtopic.php?p=62613"]
[Date "2013.10.30"]
[Round "-"]
[White "Novag Obsidian"]
[Black "Saitek Travel Champion 2100"]
[Result "*"]

1 e4 e5 2 Nf3 Nc6 3 Bb5 a6 4 Ba4 Nf6 5 O-O Be7 6 Re1 b5 7 Bb3 O-O 8 c3 d6 9 h3
Na5 10 Bc2 c5 11 d4 cxd4 12 cxd4 Bb7 13 Nc3 Qc7 14 dxe5 dxe5 15 Nd5 Nxd5 16
exd5 Rad8 17 Rxe5 Nc4 18 Qd3 g6 19 Re1 *

[fen]3r1rk1/1bq1bp1p/p5p1/1p1P4/2n5/3Q1N1P/PPB2PP1/R1B1R1K1 b - - 1 19[/fen]

sje · Post by **sje** » Tue Nov 12, 2013 8:02 pm

[+0.904/15/22:35.346/918,322,244/0] 19... Bb4 20 Rd1 Rfe8 21 Qb3 Bf8 22 Bd3 Rxd5 23 a4 Na5 24 Qc3 Qxc3 25 bxc3 Nb3

sje · Post by **sje** » Tue Nov 12, 2013 10:48 pm

sje wrote:[+0.904/15/22:35.346/918,322,244/0] 19... Bb4 20 Rd1 Rfe8 21 Qb3 Bf8 22 Bd3 Rxd5 23 a4 Na5 24 Qc3 Qxc3 25 bxc3 Nb3

[+0.890/16/2:38:31.924/6,284,066,970/0] 19... Bf6 20 Bh6 Bxb2 21 Bxf8 Bxa1 22 Rxa1 Kxf8 23 Qc3 Kg8 24 a4 Rxd5 25 Re1 Rd8 26 axb5 axb5

sje · Post by **sje** » Wed Nov 13, 2013 12:42 am

sje wrote:
sje wrote:[+0.904/15/22:35.346/918,322,244/0] 19... Bb4 20 Rd1 Rfe8 21 Qb3 Bf8 22 Bd3 Rxd5 23 a4 Na5 24 Qc3 Qxc3 25 bxc3 Nb3
[+0.890/16/2:38:31.924/6,284,066,970/0] 19... Bf6 20 Bh6 Bxb2 21 Bxf8 Bxa1 22 Rxa1 Kxf8 23 Qc3 Kg8 24 a4 Rxd5 25 Re1 Rd8 26 axb5 axb5

[+0.939/17/4:40:15.363/11,192,758,101/0] 19... Bf6 20 Bh6 Bxb2 21 Bxf8 Bxa1 22 Rxa1 Kxf8 23 Qc3 Kg8 24 a4 Bxd5 25 axb5 axb5 26 Re1 b4 27 Qxb4 Bxf3 28 gxf3

It would be possible to power your TC2100 with a 6 volt sealed lead/acid battery. One of these batteries would keep a TC2100 running for months. This would be simpler to connect compared to a wall-wart solution.

A 6 volt lantern battery would be less expensive and could power the machine for several weeks. But these batteries aren't rechargeable.

My Obsidian is loaded with fresh alkaline cells; these will remember the game and allow operation in event of a power outage. But no power also means no Internet connectivity, so then no way to communicate the moves.

Wardy · Post by **Wardy** » Wed Nov 13, 2013 7:15 pm

Bf6

This was a close call. All the way through to 9 ply the TC stuck with Bf6, at the commencement of it's 10 ply search it switched to Bxd5 which it stuck to until the 3rd move under consideration at 11ply.......... where it switched back to Bf6..........this only happened with 10 mins to go to time control!

It was such a new change the TC hadn't even managed to post up a continuation.

My own analysis of Bxd5 showed that white had no issues after Bh6....., Qc3 with Bf6 black still has a slight advantage IMHO

sje · Post by **sje** » Wed Nov 13, 2013 11:14 pm

After 2 1/2 hours, the Obsidian says 20 Bb3 with a -0.78 score.

This game might have gone differently if I had used the Novag Citrine instead; although it has the same program, its book is twice as large and that could have affected its opening.

Wardy · Post by **Wardy** » Thu Nov 14, 2013 6:44 am

If wiki is correct then the tc2100 does have a large book, even bigger than the clones that followed. However I don't think the obsidian has been disadvantaged, with the tc2100 book set to tournament it's restricted to very safe lines, also the obsidian did find book moves all by itself, I doubt the tc2100 line went much deeper. I'll check this after the match.

There is a long way to go yet, the obsidian has played some odd looking moves with the rook capture and Qc3,these aggressive moves probably score well against humans and do give it an entertaining style. The tc2100 is bound to make some errors itself yet, all is not lost.

Does the citrine suffer from the same move stack overflow issue do you know?

Not sure if wiki has been updated or if the Google German translator has got better but it now states that the tc2100 doesn't have the h8 bug??

sje · Post by **sje** » Thu Nov 14, 2013 11:00 am

Yes, the Obsidian is like other Novag models going back to the Super Constellation with a tendency towards speculative play which works well against humans.

The Obsidian has the same program as the Citrine other than the code for running the interface. So the Citrine also has the same move stack limitations as the Obsidian.

The Schachcomputer wiki has some test positions for the h8 bug; you'll have to try them after the game.

After 14:15, the Obsidian as at 11 ply and still says 20 Bb3.

Wardy · Post by **Wardy** » Thu Nov 14, 2013 11:38 am

We can always have a tc2100 v citrine match next, with the tc as white perhaps on the same settings?

I'll have a play with the H8 bug after this game.

Wardy · Post by **Wardy** » Thu Nov 14, 2013 12:03 pm

Did a quick blitz shoot out after 20.....Bb3 and the resulting end game came down to a rook and pawn ending where black was unable to leverage the extra pawn into a win.

Hopefully both machines can play this well!

sje · Post by **sje** » Thu Nov 14, 2013 3:12 pm

After no more than 18:20, the search concluded with the move 20 Bb3.

To run a match with the Citrine, I'd like to have done with pondering active. I would also need to get a UPS to keep the machine running during outages.

Wardy · Post by **Wardy** » Thu Nov 14, 2013 4:58 pm

Ok, that might need to wait till I get my hands on a GK2100 for ease.

If anyone has a good one give me a shout!