Thursday, October 18, 2007

Martillo vs. SMIRF 1.66g, 1-0 in 63



Here is a link to the game (with SMIRF 1.66g score and comments)

Martillo vs. SMIRF

The Martillo programmer has had problems getting his software to run on the hardware that will be used for the upcoming 2007 Gothic Chess Computer World Championship. Finally, this evening, he was able to get it to run. He requested a casual game to be played against SMIRF, which I did do.

The time controls were G/30 + 15 seconds per move.

I will post the complete game here tomorrow for everyone to review.

My comments on the game:

Both programs showed similar evaluations and Principal Variations for much of the opening. Castling did not appear in the P.V. for quite some time as each program moved their pieces around the board. Both were reaching roughly the same search depths, with neither really outsearching the other.

There was some odd form of "Archbishop chasing Queen" where I thought some of the moves were unusual. Martillo finished a pawn ahead as a result of the run-around. As each program moved progressively faster, it seems that Martillo was able to search 1- and 2-ply deeper, until its score grew. SMIRF's score fluctuated +/- 0.4 pawns through all of this, thinking it was +0.4 one moment, down to +0.2, then to -0.1, back up to +0.1, and back and forth, finally down -0.4

Meanwhile Martillo showed about +2.3 or better, growing to +4.3 at one point. Yet, the score never increased for Martillo, and never got worse for SMIRF. Move after move, +4 for Martillo, about -0.8 for SMIRF at that point.

Finally, towards the end, Martillo picked up the material and checkmated on move 63.

11 comments:

Smirf said...

Well, there are indeed weaker older versions of SMIRF. And I could not read here, whether it has been the faster MS- version or the slower donationware BC- version.

As for the installation problems: the newer versions should even be installable at MS Vista, or XP 64.

Make an installation attempt using:

http://www.10x8.net/down/SmNewSetup.exe (weaker donationware version)

And - if need be - send me reports on its weaknesses ...

GothicChessInventor said...

The 1.66g version was the version you gave me with the registration keys. I don't know if this was the stronger version or the weaker version.

Everyone who has submitted their tournament versions of the programs has copies of every other program that will play in the tournament. So far we will use SMIRF 1.66g for the tournament since no other has been substituted.

Final submissions for the tournament must be made 2 days before the start of the event with changes permitted one time per week during the running of the event.

Smirf said...

Ed, if that is the sent version, then it will be a faster MS- engine. Nevertheless the development of SMIRF has been continued. Thus currently a version even better than the one of the last blog game here could be downloaded by those, who own a key set like you do.

To use the most recent published version following has to be done: a) save the current INI file from SMIRF's installation folder for to prevent any lost of the keys, b) uninstall the SMIRF application, c) install the new SMIRF GUI system using: http://www.10x8.net/down/SmNewSetup.exe , d) replace the SMIRF engine by a more recent version using: http://www.10x8.net/down/SmirfEngine.dll

In the case of having an actual GUI and a personal key set, merely step d) is necessary.

This moment I am trying to add a few amount of opening moves to the SMIRF engine, thus there will be additional updates during the next weeks.

geography_teacher said...

That was an interesting game. Was this the first game ever for Martillo? Is it pronounced like the Spanish Mar-TEA-yo or is it as in English, Mar-TILL-oh?

I liked the Archbishop fianchetto move shown in the diagram! Is this special code in Martillo to play like that?

Smirf programmer - why did smirf play 34 Nxf4 when its knight was hanging?

Martillo programmer - why did your program play 43. Kj1 instead of 43. Ki2?

Nice game!

Victor Vector said...

I don't think the Martillo programmer has ever posted on here. And what are all those numbers in the games replay list?

Smirf said...

Instead of commenting a game with one outdated engine, I post another fresh game between currently published ChessV and an internal testing release of SMIRF, playing a game of Embassy Chess:

[Event "Embassy Chess 40 sec / move"]
[Site "CHESSBOX-MAC-XP"]
[Date "2007.10.22"]
[Time "22:08:06"]
[Round "A"]
[White "ChessV 0.9.3 (64+32 MB)"]
[Black "SMIRF MS-169c (128 MB)"]
[Result "0-1"]
[Annotator "RS"]
[SetUp "1"]
[FEN "rnbqkmcbnr/pppppppppp/10/10/10/10/PPPPPPPPPP/RNBQKMCBNR w mKQkq - 0 1"]

1. f4 Nh6 {(10.19) +0.031} 2. g3 d5 {(11.00) -0.018} 3. Nh3 c6 {(10.02=)
-0.051} 4. Bg2 Cf6 {(09.04=) +0.086} 5. Nc3 Na6 {(09.37) +0.307} 6. j4 g6
{(09.01=) +0.449} 7. d3 Bg4 {(09.03) +0.406} 8. Nf2 Be6 {(09.01=) +0.781} 9. e4
d4 {(09.17) +0.957} 10. Ne2 Qd7 {(09.35) +0.744} 11. e5 Cg7 {(11.02=) +0.256}
12. Nh3 Bg4 {(10.01=) +0.346} 13. Bf3 Bxf3 {(10.01) +0.424} 14. Mxf3 c5
{(09.34) +0.516} 15. Ng5 Ch5 {(09.16) +0.598} 16. Mf1 O-O-O {(09.32) +0.662}
17. Ch3 Cg4 {(09.46) +0.625} 18. Ci5 Cxi5 {(09.32) +0.326} 19. jxi5 f6
{(11.01=) +0.330} 20. ixh6 fxg5 {(09.10) +0.609} 21. hxi7 Bxi7 {(11.01=)
+0.525} 22. Ng1 gxf4 {(10.01=) +0.711} 23. Bxf4 Bxf4 {(10.33) +0.732} 24. Mxf4
Nb4 {(10.01=) +0.926} 25. Kd2 Mxf4 {(10.19) +2.188} 26. gxf4 c4 {(11.02=)
+2.402} 27. a3 Nd5 {(11.01=) +2.879} 28. Rxj7 Rxj7 {(12.01=) +4.029} 29. Qxj7
cxd3 {(12.01=) +4.656} 30. Rc1 Qb5 {(12.05) +6.100} 31. Qf3 Ne3 {(12.02=)
+8.609} 32. b4 Qa4 {(11.13) +9.928} 33. c4 Qxa3 {(11.01=) +13.81} 34. c5 Qb2+
{(11.06) +14.99} 35. Kxd3 Qxc1 {(12.01=) +15.39} 36. Qe2 Rd8 {(11.07) +17.22}
37. Ke4 Qxg1 {(12.27) +22.79} 38. Kf3 Qg4+ {(09.01=) +M~007} 39. Kf2 Qxf4+
{(07.01=) +M~006} 40. Qf3 Qxh2+ {(06.01=) +M~005} 41. Ke1 Rj8 {(04.01=) +M~004}
42. Qf2 Qh1+ {(02.01=) +M~003} 43. Qg1 Qxg1+ {(03.01=) +M~003} 44. Ke2 Qf1+
{(02.01=) +M~002} 45. Kd2 Qd1# {(02.00?) +M~001} 0-1

You will find out, that SMIRF's mating procedure still is unprecise, but sufficient to succeed. ChessV has been improved a lot. It seems to search a little bit more selective than SMIRF does and to tend to slightly over-evaluate centralized pawns.

GothicChessInventor said...

Since the later versions of ChessV do not support Gothic Chess, the 2004 version will be used in this tournament, unless someone recompiles a version that will support Gothic Chess.

I had ChessV removed from sourceforge.net since he used his software product to slander my name and spread lies about Gothic Chess. Greg Strong is now permanently banned from sourceforge.net forever.

Crocodile said...

But i thought you've said that the person who did those things, was not the real Greg Strong.

GothicChessInventor said...

Details are important. Your statement, as it is worded, is incorrect.

I have a version of ChessV that will draw up a dialog box that is essentially one big slander against myself and Gothic Chess. Greg Strong distributed it using sourcforge.net, and I had my attorney contact their legal team.

Strong was forever banned from there as a result.

You are confusing the person who signs on the discussion board with the name "ChessV" with Greg Strong. We now know they are not the same people.

But Greg Strong is guilty of what I stated. He knows it. I know it. The legal team at sourceforge.net knows it.

Cartaphilus said...

So which program is best in your opinion Ed?

GothicChessInventor said...

It's very difficult to say which program is best. Vortex has had its ups and downs, mostly due to my coding it in "spurts" and not testing it thoroughly. SMIRF has shown improvement. The new programs have shocked some of us.

Any program can win it at this point!