> Its advantage over Houdini... wait, what? Does Houdini have a playing style? It seems to me it just plays solidly and waits for the opponent to make a mistake, then it heavily punishes it, it doesn't seem to me a Gambit engine at all.
Yes I was thinking the same thing! Since when does Houdini sacrifice??
1. d4 Nf6 2. c4 g6 3. Nc3 Bg7 4. e4 d6 5. Nf3 O-O 6. Be2 e5 7.
O-O Nc6 8. d5 Ne7 9. Nd2 a5 10. a3 Nd7 11. Rb1 f5 12. b4 axb4 13. axb4 Nf6 14.
f3 f4 15. c5 g5 16. Nc4 Ng6 17. Bd2 h5 18. h3 Rf7 19. cxd6 cxd6 20. Ra1 Rb8 21.
Be1 Bf8 22. Bf2 b6 23. Nd2 g4 24. hxg4 Rh7 25. Nb5 Ne8 26. Ra7 Bd7 27. gxh5 Nh4
28. g4 fxg3 29. Bxg3 Bh6 30. Kh2 Bf4 31. Bxf4 Ng6 32. Bg3 Nf4 33. Rxd7 Rxh5+
34. Kg1 Qg5 35. Kf2 Rh2+ 36. Ke3 Qxg3 37. Rg1 Rxe2+ 38. Qxe2 Qxg1+ 39. Qf2 Qg5
40. Qh2 Nf6 41. Rxd6 Rc8 42. Rxf6 Qxf6 43. Qg1+ Kf8 44. Qh2 Rc1 45. Kf2 Qg6 46.
Nd4 Rd1 47. N2b3 Rd3 48. d6 Ke8 49. Nc6 Qg7 50. Qg3 Qxg3+ 51. Kxg3 Ng6 52. Nca5
bxa5 53. Nxa5 Rxd6 54. Kg4 Rb6 55. Nc4 Rxb4 56. Ne3 Kf7 57. Kg5 Rb3 0-1
i'm with the OP about playing styles. the strongest engines play the lamest openings if you live for tactics and have no understanding of positional play. if you ONLY play gambits, fritz will suck all of the life out of an aggressive position until it gets that 1 point back instead of exploiting the initiative.
i really used to like the aggressive setting on the rebel demo i tried back when i still had win98 as it really was pushy and would end games at least a dozen moves faster than fritz which takes 40 moves to beat a 1400 or so player versus 25 or less with rebel. it's too bad it can't be used with XP. i would have bought a used CD to have an opening repertoire that doesn't count beans.
sure, at the grandmaster level gambits might be unsound, but at the amateur level, they're a lot better. there's nothing like chasing a +400's king around the board as you just keep developing pieces. you can't get that out of "sound" or as i call 'em boring openings.
Watching R3 Dynamic with contempt play the King's Gambit is always enjoyable.
Houdini is a terrible gambit engine. No personality whatsoever and far too defensive.
If Critter gave us access to piece values, king safety and mobility I think we could create a killer gambit engine.
> I still think Rybka 3 Dynamic with a very high contempt setting is the best gambit engine around. The crazy Zappa personality is too insane for engine matches but it can be useful for analyzing unbalanced positions when you need an engine that knows how to make use of initiative.
What contemp value do you recommend to turn Rybka 3 Dynamic a great gambit engine?
If it's to produce games against other engines, Contempt 50 should suffice. Maybe Moz has in mind even higher contempt, though.
To use it for games, change the setting "Contempt Play" to a higher value, to use it for analysis, change the "Contempt Analyze" to a higher value. Since you are doing this for the style, and the dynamic version has a better playing style anyway, it's recommended that you use the Rybka 3 Dynamic version when adding Contempt.
This setting was removed from Rybka 4.
Where is this value located in Rybka 3? Why was it removed from Rybka 4?
> Why was it removed from Rybka 4?
I don't know, my guess is that Vas just messed up the code, because: the same happened with Persistent Hash and Time Management, when Vas made Rybka work for the Cluster, some code was no longer functional, so he had to rewrite it for the UCI version.
He had no time to rewrite persistent hash, and the rewritten time management code was poor. He also added tweaking of piece values for Rybka 4, and he thought a personality that increased own piece values could just replace contempt, so contempt was "deprecated" ("superseded" by piece value tweaking, the problem is that there's no way to change Rybka 4's king safety, while high contempt of Rybka 3 would increase aggressiveness towards the enemy king).
Since he lost the code of Rybka 3, its contempt style is probably lost forever.
As a result, the basic engine is very strong, but also very unpolished, and with few features aside from the engine. If there is an R5, it will follow the same development process...
> If it's to produce games against other engines, Contempt 50 should suffice. Maybe Moz has in mind even higher contempt, though.
I think a contempt of 30-50 is the sweet spot for Rybka 3 Dynamic. That's the range I use any time I'm analyzing openings like the KG, Halloween or Evans Gambits. When using a high contempt in IDeA you have to be careful and always supplement your analysis with another strong engine by creating a secondary tree that serves as a check on the Dynamic analysis. It's time consuming but analyzing in this way will reveal interesting discrepancies that are often exploitable in correspondence games. For example, an engine like Houdini will often completely miss some clever idea that Dynamic finds and having two trees makes it easy to spot those differences. Of course, it works both ways. Houdini is just as likely to completely refute some idea that the high contempt R3 Dynamic thinks is wonderful.
playing style matters in gambits too. i'd use rebel aggressive long before any fritz or crafty engine, but will check zappa out afte5r reading everything in this thread.
But that Petrosian has another GM named after him says a lot.
His style may be called boring. But Om you can compare it to Dravid's style in Cricket - A Sound style, Dravid may not win u a game the way Sehwag does. But when it comes to saving games he sure is the key. So i used to like Petrosian because he believed in playing sound chess and not speculating unlike Tal,Lasker etc.
So what if it is boring, it takes extraordinary skills to hold back when u see a juicy sacrifice on board.
>So i used to like Petrosian because he believed in playing sound chess and not speculating unlike Tal,Lasker etc.
Petrosian was unique prophylactic player and can't be compared to any other player other than his idol Aron. Lasker is another player who used psychology as his tactics rather than playing it on board. He was a philosophical player and a calculating giant. No wonder he was the longest reigning champ.
>Do you still thinker 5.3B inert better in sytle then 5.1c?
It's not that Thinker 5.3B has a better style than 5.1C, it's that it's considerably stronger so that it gets the chance to show off its style against much stronger opposition.
Example case: Rybka 3 Dynamic Contempt 15 is an engine of a great playing style too, but when I matched Thinker 5.1C against it, Thinker was destroyed over and over so that it never had a chance to show its style. With Thinker 5.3B it was different, it was strong enough to play some beautiful wins against Rybka 3. Future versions of Thinker would not win on an exciting manner so that's why consider 5.3B Inert the peak.
As a recap:
Thinker 5.3B Inert plays the best style against stronger engines (say, >2950 elo engine)
Thinker 5.1C Passive plays the best style(?) against weaker engines (say, <2900 elo engine)
Thinker 5.4A(?) Active plays the best style against human opponents (it swindles which is suicide against engines).
Basically, if Thinker 5.1C Passive is losing too many games, you're using the wrong version.
> I guess i can use the 5.1c for older engines if need be :) Theirs many ways to handicap these great gambit engines old or new with 1 Cpu or more! :)
>Why do you like Cm9000 tribute is it a setting you made?
No, the setting is by Graham Banks, and it comes from this page:
Okay, to clarify, CM9000 Tribute is my favorite one from those in THAT page, as those are all the settings I've tested. I made those styles play a couple of games and liked Tribute the best, but of course maybe a better playing style exists outside of those, there's a gazillion Chessmaster personalities out there and I just don't have the time to test them, but I stick to my Tribute recommendation from the top 9 (+including default personality).
>In the meantime i be buying a ipod touch so i can test the new chesstiger! Hopefully other tacticial engines follow!
>Have you tested Glass aggressive personality? its around 2300
I've never heard about Glass before.
>Well fernando from talkchess forum thinks its te new CSTAL!! http://talkchess.com/forum/viewtopic.php?t=39820
Thanks for the tip, will check it out!
>Anyway thanks for replying as always :) I wonder what Thinker active will do with old engines. If the old engines see the swindle :D
Oh, indeed, Thinker Active should also show the best playing style against engines that are as weak as a weak human.
could someone please either recommend a useable FREEWARE engine as well as settings that will make it value aggression and especially the initiative in gambit lines unlike MOST boring engines like wishy washy fritz that suck all the life out of an edgy position to either plan on getting the point back or even retreating for no reason?
i'm not looking for a 2900 rated engine as i play real humans and few over 1500 for now. i just want to expand my opening theory as i keep getting in trouble against -200s in the opening as i only understand tactics & targests. i REALLY liked the way that the old demo of rebel used to play with the aggressive setting. it's too bad it can't be had for free with XP. i liked it's style even in "low rated saturday through thursday mode".
if i were to buy an engine, and i've looked for a used copy many times on ebay, it would be gambit tiger. any help in finding an engine that values speedy development and attacking in the opening would be greatly appreciated. looking at engine settings & trying to decode them when there are no tutorials for any of them that i've found only gives me a headache.
glaurung is supposed to be aggressive and it looks like it has settings for making it even moreso. the mobility feature sounds particularly interesting as i can't stand closed game straightjackets and eventually lose a position trying to get my pieces out and attacking. any suggestions for tweaking that? (w-32 version)
> 'm not looking for a 2900 rated engine as i play real humans and few over 1500 for now
Buy good books about tactics and forget computers.
i've tried reading books to get positional knowledge, but even with THREE of them, still can't figure out a pawn ending as they're written in some unfathomable to me abstract code.
tactics are solid principles. back when i was a mid 1500, some snotty 1750 at FICS that mocked me with "if all you know is tactics, you suck" attitude only to lose two games in a row to me with tactics, although one could argue that my two pawns assisted attack on his back rank with a rook was positional, i only looked at it in terms of "how can i make these pieces rip him a new one?" i won the very highest rated game ever against a 2500 (!!!) on caissa in a 15 minute game with a totally wrecked 0-0-0 as soon as i could find a square that his queen couldn't check me on and grab another pawn and could gain the initiative with a rook check followed by an x-ray attack on his rook with my queen. after he decided to trade queens, all i had to do was protect the back rank from his pawn rush whilst shielding mine from his king. it was particularly satisfying as he kept taunting me in the opening with "where's your gambit now?!" jabs because he played a line that doesn't allow for them and he became a logoff weasel making me wait 9 minutes for my win.
it was a REALLY lucky game that my pieces were on the right square to get some counter play, but luck had nothing to do with this game against an 1100 last night: 1.e4 e5 2.Bc4 Nf6 3.d4 Nc6 4.c3 Nxe4 5.Bd5 (a book move, but it's fugly and just the type of thing i'd like to avoid as i see no reason for it as it wastes a tempo instead of finding a way to get my pawn SAFELY on e5 so i could Bxf6+ and get the knight back with Qf3+... a big part of the loss was my trying to find a way to make that happen as it was the only target i could see) 5...f5 (there goes that plan... the beginning of the end where yet again i lose an annoying pawn game where all i'm doing is defending myself to a loss) 6.Nf3 Bd6 7.Nxe5 Bxe5 8.dxe5 (hoping in vain that my opponent has some other plan like d6 or a fianchetto, but who ends all of my setting up with the proper Nxe5 reply) 8...Nxe5 9.Nd2 Nc5 10.O-O c6 11.Bb3 d5 12.Nf3 Ned3 (NOTHING I COULD DO could break those two annoying knights apart!) 13.Nd4 O-O 14.c4 Qf6 15.Nf3 Be6 16.Bg5 Qg6 17.Bc2 Nxb2 (the positional nightmare continues and defending c4 is futile after Re8 and by this point, i'm so frustrated i can't think clearly anymore anyways because those knights are really whizzing me off) 18.Qc1 Nxc4 19.Nd4 Ne4 20.Nxe6 Qxe6 21.Bxe4 Qxe4 22.Re1 Qd4 23.Re7 h6 24.Be3 Qf6 25.Rxb7 Nxe3 (forcing the fugly pawn takes reply because my a1 rook is pinned) 26.fxe3 Rfe8 27.Rab1 f4 28.exf4 Qd4+ 29.Kh1 Re2 30.Qg1 Qe4 (i'm livid now... i can't connect my rooks on the 7th rank or defend mine or weaken what's left of my castle with either g3 or h3 to prevent the eventual corridor mate i know is coming) 31.R1b4 Qe6 32.Rc7 Re8 (every attempt to attack SOMETHING in the game is refuted... ARRRRRRRGH!) 33.Rb1 Re1 34.Rxe1 Qxe1 35.h3 Qg3 36.Rxc6 Qxf4 37.Rc7 ??? (i'm just so frustrated now all i can do is make one last attempt at connecting on the 7th rank because i can't defend and was blind with rage) 37...Qxc7 resign 0-1 and like so many infuriating games, i spend the ENTIRE game defending against a -200 that understands position better than i do without ever getting the opportunity to attack ANYTHING. i could see all of their tactics coming except maybe playing hide & seek with his knights if that could be called a tactic, but i knew i had to take at least one of them as that's the exact kind of position i'd love to be in myself... looking for forking opportunities with the "a pair of knights is a pair of rooks on the back rank's worst nightmare" tactic.
i lose way more lower rated games than i should getting picked off just like that because my opponent keeps blocking every attempt at attacking and mobilizing. i understand tactics and even if i'm positionally clueless, i can at least MEMORIZE positions if not understand the subtleties behind them. i have to credit my former 1650 rating ENTIRELY to studying and memorizing the theory from a couple scandinavian books and learning at least the basics of the wing gambit & krejcik alekhine which i recently DESTROYED a high 1400 in by simply planning what i needed to do after Bxf7+ and developing pieces when i forced him out in the open. i owe that FUN FUN FUN victory entirely to "book learnin" and not getting my pieces on the worst squares possible as in that game where no book could ever help, but some home brewed opening study would have given me a much better development where i could make something happen when the time is right. everyone's mind works differently. mine sees tactics like they're painted in dayglow under a blacklight, but getting into the position where i can use them is the problem. i can't do that without opening theory and a lack of it is holding me back. without it, i'll just keep making the same kind of "i don't know what the heck i'm doing" in the opening mistakes.
i'm simply positionally inept. my mind doesn't do abstractions. tactics are solid concepts. for my total inability to grasp a pawn ending, i INSTANTLY grasped the two bishops ending when i read that chapter in a book... same thing as a rooks ending, only diagonal. i really don't get why so many people seem bent on talking me out of correcting what i know to be my biggest weakness in chess... position. i'll never be able to figure it out on my own and i STILL can't draw or win a pawn ending
etc. read that books and forget about engines, they will not teach you a shit about chess.
instead of pointlessly telling me to do something that i know won't help (i have how to reassess your chess, petrosian's legecy and best lessons of a chess coach collecting dust because they're mostly about positional concepts that are voodoo science to me and that make my head hurt, why not just answer my question instead of evading it and trying to hold me back from improving the ONLY way I KNOW that works... tightening my opening play. tactics are useless if i get creamed in the opening like that game against an 1100 that is embarrassing in the extreme, but a perfect illustration of why i need to find solutions to the same problems i see over and over and can NEVER figure out because my mind just doesn't get abstractions.
if those books work for you, more power to you. all they do for me is aggravate me because the speak a language i don't. they don't help me one bit and so far no one here is either as you keep refusing to answer my question.
how many ways and times must i ask, "what is the most gambit friendly freeware engine & tuning out there that i can use to patch all of the holes in my opening theory and stop losing games before the 10th move?" and get an actual answer instead of evasions and non-answers? trying to is just like banging my head against the wall in openings when i don't know how to reply to a subtle move that looks harmless to me as it's not an attacking move. i'm not going to beg and plead. i'll just find somewhere else to ask this simple question and get a direct answer if no one here is willing to for what reason, i'll never know. i never withold information that i have that can help someone else if i have it.
i also want to totally get rid of ALL of the fritz theory i've already created as that's probably part of why i'm getting in trouble in those lines to an extent because they're too positional with no bite. i envy all of you that can grasp the subtleties of position that just go in one ear and out the other no matter how many times i try to read them while i can breeze through one of those tactics problems books by what's his name in something like 15-20 minutes. the ONLY way to improve my game efficiently is expanding my opening theory and ESPECIALLY steering it towards "unsound gambits" after trying openings like the grand prix attack, c3, advance & smith morra and hating them all until i tried the wing gambit and instantly clicked with it even if there's hardly any theory for it. aggression is my comfort zone and defending or simplified endings are sure paths to losing for me like another recent game where my rook and king couldn't stop a bishop & 2 pawns from promoting.
i've been wasting my time trying to get a straight answer on this for almost TWO MONTHS by now with the only actual help being from someone suggesting a zappa mexico tuning when that's not freeware
Ok you memorize 10 moves in any opening, you play e4 and I answer a6 what you do?
For your answer there are not gambit friendly freeware engines. Satisfy now?
i'm not looking to book up on EVERY possibility over the board as that would turn into millions of positions by move #5 i best, but to just start improving in the lines i actually DO see a lot like 1.e4 d5 2.exd5 Nf6 3.Nc3 Nxd5 4.Nxd5 Qxd5 5.Nf3 which usually follows with a Bc5+ or queenside pawnstorm and me clueless about where to move my queen short of retreating back to d8 an waste a tempo which is exactly the kind of thing i'm loathe to do. i tried playing it like the scandinavian gambit too with 3...c6 4.dxc6 Nxc6 and kept getting in trouble with Qf3 and a bishop lined up on f7 threatening mate, or a flat out castle grabbing Bxf7+ sac or with the really annoying e6 blocking my queen's bishop from f5 or the classic g4 pin. that's what i'm talking about, my opponents forcingf me into the same annoying positions that cramp my development that i don't know how to reply to.
a6 doesn't bother me at all. it doesn't stop me from developing normally. if i can, i just open classically in an unfamiliar position and against ANYTHING but 1.e4 and 1.g4, i play the trusty (but boring) stonewall formation. i really want to stop playing that and try something like the albin countergambit, but i'd have to build an entire repertoire.
maybe you don't need to study openings, but i do. that's the only way i can get into playable situations most of the time. i just can't look at a pawn formation and think... OK, i really need to control c6 and open up the e file as there's nothing on c6 but an empty square. my mind just looks at PIECES and what they can do. i have an idiot savant type brain when it comes to abstractions and the concepts in chess books are just that... abstractions that i can't VISUALIZE. tactics... this piece does this, then that one does that... OK... no prob!
i'll only say this one more time, i wasn't able to make the jump from 1400 to 1650 a decade ago until i started studying openings. they filled in the gaps in understanding for MY THINKING STYLE (not yours!) needed to improve. memorizing openings is easier than understanding them for me. then, i'm ready to exploit whatever opportunities come to me like the game where i waxed a 2500 rated player in 15 min. a REAL 2500 BTW as i checked his game history to see that he was a regular player. all i needed was that one tactic to gain the initiative and once i have it, i rarely let go of it.
> 1.e4 d5 2.e5
play c5 or Bf5 and you got yourself an improved french or caro kan.
> 1.e4 d5 2.exd5 Nf6 3.Nc3 Nxd5 4.Nxd5 Qxd5 5.Nf3
> and me clueless about where to move my queen short
and why you need to move your queen? white already made you a favor exchanging in the center
If this positions are your problem you do not need gambits or memorizing openings, you need to play fundamental basic concepts.
i just hate the idea of Qxd5 anyways. that's why i switched to the Nf6 line as soon as i learned about it. i love using my queen, but only after everything else is out or i have some kind of clear target. without something to attack, i just never know what to do
i can't stand the carokann or french! i call 'em trench weasel defenses
> i can't stand the carokann or french! i call 'em trench weasel defenses
Then you must play something else, d5 is not for you.
The thing is your queen in the center is well placed the only piece that could harass her is the bishop prior a b3 move or pawn c4 just move it back to d6 and you are more than ok. The moves that give you problem are in fact harmless variation for black so I recommend forget about openings and start learning some strategy.
besides, if i dropped the scandinavian. i'd never get to play one of my favorite lines... the scandinavian gambit
> i was told that contempt has nothing to do with aggression in that engine.
True, contempt only increases aggressiveness on the Rybka 3 engines, others only use this setting to avoid draws.
>i REALLY liked the way that the old demo of rebel used to play with the aggressive setting. it's too bad it can't be had for free with XP.
Look at my suggestion in the other thread, Pro Deo is the free version of Rebel.
Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill