Random Posts

Wednesday, March 6, 2019

IdeA Analysis With Aquarium

     Complex positions are the only way to play for a win is a quote from ICCF SIM Kostas Oreopoulos of Greece who is rated 2490. There is a site called LIPEAD, a non-profit Peruvian association affiliated with ICCF that represents ICCF in Latin America, that has a two part series by Oreopoulos in which he describes how to prepare and maintain an opening repertoire for correspondence chess using the Aquarium program. 
     The articles warn that it is an involved and difficult task and it assumes you have a fair working knowledge of Aquarium because it uses the program’s IdeA feature. PART 1  PART 2  ChessPub also an old discussion on using IdeA. 
     With Interactive Deep Analysis you start with a position called a root node and the program will keep exploring possibilities and build a tree from that position for as long as you want it to run. You can mark moves as good, bad, dubious, interesting, etc. The program won’t analyze the ones marked bad and will spend extra time on those marked as good or interesting. 
     According to the description of IdeA, instead of analyzing the same position forever, it behaves like a strong chess player analyzing a position. If you ever studied Kotov’s Think Like a Grandmaster, you no doubt remember his well known advice to identify candidate moves and methodically examine them to build an “analysis tree.” That’s how IDeA works. 
     I will throw this in about Kotov’ book. According to the late James R. Schroeder, this book, which caused a sensation when first published in the US by Kenneth Smith, was written because Kotov was weak with Ns, but he wasn’t aware of it and so created an artificial system which he hoped would avoid blunders. Schroeder claimed that because Kotov was weak when it came to playing with Ns he sometimes misjudged a position. 
     The purpose of IDeA is to dig deeply into a position and return as much information about it as possible. IDeA keeps its analysis in a tree structure which is unlimited in size and the user can browse at will, even while the analysis is in progress. And, you can stop the analysis and go back later and it will pick up where it left off. You can do neither of those things with the Infinite Analysis feature in other programs. 
     As I have mentioned in previous posts, I find Aquarium difficult to use and somewhat confusing; one has to read page after page in the manuals just to learn how to perform basic stuff. So, for me, I have done nothing more than tinker with IdeA because most of my correspondence chess has been played on LSS in rapid tournaments and line most players I don’t spend a lot of time selecting moves. Most people play whatever move the engine suggests after only a few minutes of analysis, but I do like to experiment a little bit. Because everybody is playing pretty fast and a lot of opponents have a ton of games going, you can experiment with unusual openings. 
     One such opening has been the Urusov Gambit (1.e4 e5 2.Bc4 Nf6 3.d4) and my current score with it is +6 -0 =5 (including one opponent who lost twice against it)! The Urusov might be a good IdeA project, but if you read the articles you’ll see why I am reluctant to bother with doing it! Back in October, I published an article titled Is Opening Theory Wrong About the Urusov? I relied heavily on analysis by Michael Goeller found on his site HERE
     It’s interesting that in Victor Bologan's Black Weapons in the Open Games: How to Play for a Win if White Avoids the Ruy Lopez, he thinks it’s too dangerous for black to accept the gambit! In all 11 of my engine assisted games nobody declined it, so it would seem Bologan is correct. 
     According to my modified version of the opening book that came with Fritz, in OTB play black accepted the gambit with 3...exd4 over 75 percent of the time and the percentage results for white are +39 -33 =28. Accepting the gambit with 3...Nxe4 is even worse: white scored +69 -16 =15. 
     If black declines the gambit with 3...Nc6 white does even better, scoring percentage-wise +63 -18 =19. The Fritz opening book has zero lines in which black has a plus score. Of course, those statistics are no doubt skewed because the Urusov isn’t played by titled players so the results are not always related to the choice of opening. But, what the stats do show is that unless you’re playing titled players, the Urusov has a lot going for it! 
     Declining the gambit with 3...Nc6 is interesting because it leads to a line also discussed by Michael Goeller called the Perreux Variation of the Two Knights Defense. For a fascinating discussion of this line visit HERE and HERE.  I have also seen this opening classified as the Bishop’s Opening, Ponziaini Gambit. 
     After toying with Aquarium’s IdeA on the Urusov it appears that theoretically the best move for black is to accept the gambit, but accepting it, based on Bologan’s advice and my results in quick games, black might do better with the Perreux Variation despite the horrible results shown in the Fritz opening book. 
     I let Fritz do a deep position analysis on the Perreux Variation using both Stockfish 10 and Komodo 10. The deep position analysis feature in Fritz is good for getting deep and detailed analysis of a critical position, and is especially interesting for correspondence players. 
     DPA generates a detailed analysis tree for a given position and you can determine how deep and broad the tree should be and what moves should be included or excluded from the analysis. It’ similar to IdeA, but, as mentioned, once you stop the analysis, that’s it; you can go back and pick up where you left off. 
     The move most chosen by humans after 1. e4 e5 2. Bc4 Nf6 3. d4 Nc6 is 4.Nf3 which according to the DPA with Komodo and Stockfish results in equality. Instead, they both prefer 4.d5. 
     Here is an interesting game by a couple of masters using the Bishop’s Opening Perreux Variation where white played 4.d5. In the game white managed to establish a winning position. But, as often happens, even with masters, white did not follow up correctly and lost. At move 23 he missed a chance to win a piece and soon reached a position where there was only one move that wins...26.Rxe4. 
     Had white’s tactical antenna been out he might have seen a signpost or two that there may be a winning tactic available. At move 23 black’s B had limited mobility. And, at move 26 the alignment of black’s N, B and Q coupled with the pin on the N was suggestive.  Refer to my post Tactics, the Pornography of Chess

No comments:

Post a Comment