[ List Earliest Comments Only For Pages | Games | Rated Pages | Rated Games | Subjects of Discussion ]
Comments/Ratings for a Single Item
SMIRF
. Program that plays various 8x10 chess variants.[All Comments] [Add Comment or Rating]Hi Derek, SMIRF currently is designed to cover 8x8 and 10x8 variants, which somehow compatibly are including traditional chess as a sort of superset. Nonetheless SMIRF has included some additional features like supporting some different kinds of castling or the absence of Chancellors in Janus Chess. Thus supporting more piece types actually is not intended. Nevertheless it is thinkable to double the supported gaits from 8 piece types to 16 in a follow-up version of SMIRF. But this will not be in the nearest future. P.S.: not to forget: the future of SMIRF depends on voluntary donations.
This is request for future versions of SMIRF. I notice that there are some games played upon the 10H x 10W board that are related to Capablanca Chess (as evident by the fact that they are played using all or most of the same pieces). If I am correct in presuming that CRC is one of your favorite games, then you may find the inclusion of this 3rd board size desirable. You can be sure that players who use your program would find it desirable.
Hi Derek, your proposal will not be forgotten, but it has to wait until there would be some more serious donations from other people. Indeed I worked on that idea already conceptually. But there are some obstacles. E.g. the row nr could be encoded as a marked bit within a single byte, but only as long as using not more than 8 rows. So a 10x10 board would qualify a lot of procedures as no longer usable then and would demand for an appropriate rewrite of a lot of code. Additionally I am still missing a matching 10x10 variant compatibly including the rules of traditional chess.
This is a request for future versions of SMIRF. A 4th castling option would be useful occasionally: no castling. Corner Chess 8 x 8, my first (and worst) chess variant ever invented is a game where only standard chess pieces are used in which castling is totally unnecessary. [The king is very well protected at the opening setup.] Corner Chess 8 x 8 /play/erf/CornerCh.html
Hi Derek, beside of the fact, that Corner Chess is a 8x8 setup option within SMIRF, there is no need for to have a separate no-castling option. Because one could switch into the edit mode and deactivate any castling potential (shown by special marker dots) simply by clicking at the matching file letters of involved rooks. Activating of castlings is done the same way. Regards, Reinhard.
I never even saw those black dots in any game that index the pieces involved in castling because I chose a black border (for my board). Besides, SMIRF cannot enable castling with the strange opening setup used in Corner Chess 8 x 8 even as the symbol for a castling option must exist within the X-FEN string. Nevermind! All is well.
Hi, I have a suggestion for a variant to be added, If SMIRF was able to play Alice Chess it would probably be the only program (besides Zillions of Games) to play the variant. There isn't even any sites on the net to play it live. There can be two options to show the board : either as two seperate board besides each other, or both boards superimposed on each other, with board B pieces up-side-down or something. -- Another thing: I've noticed a problem with modern castling in the program. Whan I put a random starting position in a 10x8 board with modern castling (I believe it is called Orthodox Castling by Chess480 inventor, whatever ..) the king on the file h or i can't castle with the rook on his right, and so is the case on the a-side of the board. As I unserstand this castling, the king should go to the corner square and the rook placed next to him. And I believe a 'free castling' option would make trying out variants which don't exist (like Grotesque Chess) easier. The castling movement can be notated like O-g for example if the king ended up on the g file, or O-h. Thank you for reading, Yours, Sibahi
Aside from a brief, cryptic note left by Reinhard Scharnagl at this web site, I have no idea why ChessBox & SMIRF have disappeared. Does anyone really know anything about this mystery? ChessBox http://www.chessbox.de
Because of a vanishing feed back and an increasing rate of being insulted, I decided to leave the scene for a while. I will work for a new version of SMIRF for Mac OS X during the next time and may return then. Regards, Reinhard.
Greg Strong wrote on Thu, Dec 28, 2006 12:01 AM UTC:I am very sorry to hear that you have had problems. Unfortunately, this is common on the Internet. It is sad that people who work hard on something, and share it for free, must be repaid with abuse. You should know that there are many people who do appreciate what you have done and who respect your accomplishments. I hope that you will reconsider and make this fine program available again. Your work *is* appreciated by most people, even if there exists the occasional person who does not (probably because of their own commercial interests.)
Hi Greg, people who know me, nevertheless could keep contact e.g. via email. I also think it over to redesign a new homesite later at www.10x8.net (allready working empty) where you also can use my new email address prefixed by ReScharn. Regards, Reinhard.
SMIRF pages at http://www.chessbox.de are online again. Nevertheless SMIRF is migrating to Mac OS X.
Greg Strong wrote on Fri, Jan 19, 2007 02:22 AM UTC:Excellent! I'm glad that this fine program is available again. I hope you continue to provide the current Windows version even as new versions develop. I must ask... Why change development to OS X? Mac computers can run Windows programs! They already can run your program (I think.) Why make SMIRF less useful by converting it?
Reinhard, please consider adding Chess256 to Smirf. It is rather easy to implement, and it is a good training concept for 'orthodox' chess players who will have no help from opening theory from the first move, while the positions are very similar to normal chess. http://hem.passagen.se/melki9/randompawn.htm My BlindChess can already play Chess256, but its a rather weak DOS program. But it's good for testing Chess256. http://hem.passagen.se/melki9/blindc.htm /Mats
SMIRF http://www.10x8.net The new SMIRF pages are up and running as well. This is a mirror web site to the old SMIRF pages.
Indeed, Derek, there is actually only a twin of chessbox.de at 10x8.net. I plan to build a site dedicated to 10x8 chess programming and a coming Mac OS X based successor of SMIRF named 'CapTen' for Capablanca and 10x8.
Smirf is a strong program, and the graphics is attractive. However, it would benefit greatly from moving the centre pawns two steps forwards, instead of moving the flank pawns two steps. Flank operations should begin only after the situation in the centre is clarified. Moving a flank pawn two steps weakens the position much more than just moving it one step, especially if it is the kingside flank pawn. /Mats
Hi Mats! I do not know to wich variant you are referring. SMIRF actually does not use any opening library even for established starting arrays, to not hide possible weaknesses during opening, which should be covered by an improved evaluation function, which in SMIRF is not stage depending but constant during the whole game. Regarding that aspect SMIRF already is handling openings not badly, so in your case it would be recommended to post me such a critizised example privately. To your prior suggestion to include some special variant in SMIRF: I am about to rewrite the whole SMIRF project and to migrate it to Mac OS X. I will remember such proposals and see then, what could be done. At the first sight your variant is matching the FullChess criteria to be included. Regards, Reinhard.
Reinhard, in the following game played between Smirf as white and Zillions as black (1.6 GHz, 10s/move), Smirf moves both his flank pawns, wholly without motivation, and put his bishop ahead of a centre pawn on its initial position. This kind of play is strategically indefensible. In games against humans the game is strategically lost, even after only one or two of these positional blunders. It sometimes adopts this style of play also in normal chess. On the other hand, it is easy to create a tenable position as white. Always move (1) the kingpawn one step, (2) the queenpawn two steps, and (3) the queenbishop-pawn two steps. This position is good regardless what black does, and white can always play for a win. As black one can always make the steps (1) and (2) and have a good position, regardless of white's moves. In almost all cases one can also make (3). These moves could be rewarded in Smirf regardless of variant. Then Smirf will always begin with a strategically tenable opening, until an opening book is developed. /Mats (for Zillions I used my rules file http://hem.passagen.se/melki9/capablanca.htm ) Zillions Save Game File Version 0.02 HCC RulesFile=CAPABL~1.ZRF VariantName=Bird's Chess 1. Knight i1 - h3 1. Pawn d7 - d6 2. Knight b1 - c3 2. Pawn f7 - f6 Black H M1 3. Pawn d2 - d3 White H M2 3. Pawn g7 - g5 Black H M3 4. Bishop c1 - e3 4. Pawn c7 - c5 Black H M4 5. Pawn g2 - g3 5. Knight i8 - h6 6. Bishop h1 - d5 6. Pawn e7 - e6 7. Bishop d5 - e4 7. Knight h6 - f5 8. Bishop e4 x f5 8. Pawn e6 x f5 9. Pawn j2 - j4 9. Bishop h8 - g7 10. Pawn a2 - a4 10. Knight b8 - c6 11. Pawn j4 - j5 11. Pawn g5 - g4 12. Knight h3 - f4 12. Chancellor d8 - d7 13. Pawn j5 - j6 13. aRook j8 - i8 = Rook 14. Pawn j6 x i7 14. Rook i8 x i7 15. Pawn h2 - h3 15. Knight c6 - d4 16. aQueen e1 - d2 flip Z6 = Queen on d2 16. Pawn h7 - h6 17. Knight c3 - d5 17. Pawn g4 x h3 18. Pawn i2 x h3 18. Pawn b7 - b6 19. Bishop e3 x d4 19. Pawn c5 x d4 20. Queen d2 - b4 20. Bishop c8 - b7 21. Queen b4 x d4 21. aRook a8 - c8 = Rook 22. Chancellor d1 - e3 22. Bishop b7 x d5 23. Knight f4 x d5 23. aQueen e8 x e3 = Queen 24. Knight d5 x e3 24. Archbishop g8 - e6 25. Queen d4 - h4 25. Archbishop e6 - g5 26. Queen h4 - h5 26. Bishop g7 - h8 27. Archbishop g1 - f3 27. Bishop h8 - g7 28. Archbishop f3 - d4 28. Rook c8 - c5 29. Pawn b2 - b4 29. Rook c5 - e5 30. Knight e3 - c4 30. Rook e5 - d5 31. Archbishop d4 - c6 31. Chancellor d7 - c7 32. Pawn b4 - b5 32. Archbishop g5 - i3 33. Queen h5 x j7 33. Rook i7 x j7 34. aRook j1 x j7 = Rook 34. Archbishop i3 - h2 35. King f1 - e1 @ e1 0 0 35. King f8 - g8 @ g8 0 0 36. Pawn d3 - d4 36. Chancellor c7 x c6 37. Pawn b5 x c6 37. Rook d5 x d4 38. Knight c4 - e3 38. Rook d4 - e4 39. aRook a1 - a3 = Rook 39. Rook e4 - e5 40. Rook a3 - c3 40. Rook e5 - e8 41. Pawn c6 - c7 41. Rook e8 - c8 42. Knight e3 x f5 42. Bishop g7 - f8 43. Knight f5 - e7 43. Bishop f8 x e7 44. Rook j7 - j8 44. King g8 - f7 @ f7 0 0 45. Rook j8 x c8 45. Archbishop h2 - i1
Hi Mats, it is not clear, which version of SMIRF you are using. I found here a differing better behaviour. Nevertheless SMIRF was not intended to play well in Blitz (though it mostly plays acceptable), it has been designed to think rather positionally, thus it needs some thinking time to create really good moves. It should make sense to discuss such questions to games in different chess fora or by private mail. Additionally SMIRF is able to create game PGN files, which are more recommended to be used in postings, because they could be marked, copied and pasted directly into SMIRF to be replayed. Regards, Reinhard.
Reinhard, I used the latest downloadable version. All I say is that there should be knowledge built in so that it takes charge of the centre in the opening. Likewise, in pawn endgames there must be knowledge about the opposition. If there is no such knowledge, then all games are strategically lost. Of course, Smirf is likely to win anyway because it is so strong, but chess programmers must learn to honour the laws of chess. The game example I gave turns the stomach of a cunning chessplayer. It's like a musician who must bear to listen to false play. It is an interesting AI project, but I don't think you can do without knowledge. Chess programmers tend to see chess as an algorithmic experiment, and they disregard the laws of chess, expecting the program to find the best move unaided by knowledge worked out during the centuries. If the programs have knowledge then 10s/move on a 1.6 GHz machine is clearly good enough. Capablanca said that he counted one move forwards, but he had an immense knowledge. /Mats
Hi Mats! It seems necessary to explain something of SMIRF's concept. This program is designed to gain a maximum output at a minimum input. You might have noticed, that SMIRF's single multivariant engine is sized only about 60K (72K for the bonus version). Its intention is to reach success WITHOUT implemented chess knowledge (during the first stages of its development). Instead the goal is to create something like a machine intelligence REPRODUCING human behavior NOT by design but maybe only by chance. If you notice a less successful behavior (as you have reported) the solution of this thus would not be to implement chess knowledge as done in a lot of other chess programs, but to make the inner functions more appropriate, on what I am always working. There is nearly nothing less intelligent (despite of being successful) as to have a lot of looking up knowledge inside of a program. Nevertheless SMIRF should be able to handle sufficiently such things like Zugzwang, passed pawns and opposition by its own means.
If you had a way of storing the program's experiences of different positions, then I would understand your concept (i.e. a learning file). But to build hundreds of years of collected understanding into wholly abstract algorithms, that I don't believe in. Chess is too deep a game for that. Possibly it would work with gomoku, but not chess. Nevertheless, it's an interesting experiment. I wish you good luck. /Mats
Please let me explain another way ... The purpose of SMIRF is to fairly play ALL 960 FRC starting positions and ALL 12,118 CRC starting positions equally and extraordinarily well. Obviously, it would be impossible to generate high-quality opening books of adequate depth and width for all of these 1000's of games within a survivable time. So, to attempt such a project is not even being realistically considered. Moreover, some people are skeptical that any of the starting positions with FRC pieces upon the 8x8 board or CRC pieces upon the 10x8 board are at all or significantly superior compared to other unknown permutations. In other words, they do not consider Chess any better than the other 959 FRC variants nor do they consider any of the 2 dozen CRC variants given names by their inventors any better than the other 12,000+ CRC variants. [Personally, I think otherwise that Chess & Opti Chess are the best FRC & CRC games, respectively. In the latter case, I admittedly lack impersonal objectivity.] In my playtesting experience, SMIRF is vastly better than merely 'an impractical experiment that plays Chess poorly' as you describe it (more or less). You are missing the point. You obviously do not fully understand that there is a strong correlation between the quality of moves generated by a pure search intelligence program (such as SMIRF) and the time, plies or positions that must be invested to achieve worthwhile results. When testing one program against another or testing one set of material values for pieces against another using the same program, I NEVER use a time per move of less than 1 hour ... running a dual-CPU (2.4 Ghz per CPU) server. I regard quickly-obtained results as random, too replete with bad moves and analytically uninteresting (since a sub-genius such as I could have won the game playing either side). Given reasonable conditions, SMIRF & ChessV make moves well beyond my capacity as a player. By comparison, the Zillions Of Games program makes bad moves even when given 24 hours per move or more.
25 comments displayed
Permalink to the exact comments currently displayed.