Free analyzer of chess games "Creatika" for MS Windows by Arkady Polyakevich. The best chess programs: engines and shells Which chess program is best for analyzing games
With a huge number of servers where you can play online with a "live" opponent - from playchess.com to chess.rc-mir.com or chesshotel.ru (yes, there are a legion of them), it's not easy to find a web interface to a decent chess "engine" offhand I was only able to find this:
1. Play chess online with the "engine" Shredder:
P.S. According to the professional, the online engine has little to do with the power of a real Shredder :)
2. Play chess online with Rybka engine:
Code for connection on your site:
P.S. It seems that the second script is buggy - it hangs stably after several moves.
Loading scripts, especially on slow connections, can take a while... In case of problems, press the F5 key in your browser to refresh the page. Applications require images and Javascript enabled in the browser, as well as support for the floating frame tag.
We don't count flash drives either, these are not developing engines with bases, but there are basically 3-4 of them everywhere - Spark Chess, asisChess, Flash Chess.
The engines and the UCI protocol are well written.
If you have anything to add, please let me know. It is desirable that the codes be the same as in this article - a simple tag
P.S's
3 . They also suggested how if not to play with the engine, then analyze:
If there is only a computer with the Internet, but you want to play or analyze with a more or less decent program, then where to go? Free and without registration. So far, I don’t see anything better than entering the tournament viewer in such situations and entering what you want to watch as a side option. The local engine, although it thinks little time, still finds errors in the games of the elite.
1. Click on any Games.
2. Scroll to the top.
3. Make a move (directly with the mouse on the board or select under the board).
4. Choose an answer option.
A more or less real Stockfish plays there with a calculation depth of up to 20 moves. On other sites (tournaments) you can analyze online in a similar way, only to get to any game.
Chess Game analysis using chess engines
Part one
translated Russian by Google
Analysis of a chess game
Using ChessBase Engines
(Part one)
Steve Lopez
Chess software "explosion" over the past fifteen years has provided many positive benefits to chess players, not the least of which is the ability to play chess anytime at will. You don't have to wait for weekly chess club meetings or take your chess to the park hoping to get some casual games.
But the most important feature provided in almost all chess software packages is often the most overlooked: the ability to have a chess engine analyze your game and provide personalized information on how to improve your game.
The whole reason I bought my first computer back in the early 1990s was to get access to this feature. I've been playing chess and studying chess books for years, but have never had the experience of someone looking at my games and showing me where I went wrong. After I got a computer and some chess programs and started using them to analyze my game, I learned a lot about my own shortcomings as a player. I decided to correct these problems and my results on the board improved significantly.
In this short article series, I will show you how to do the same. While the specific step-by-step instructions will apply to the use of chess engines within the ChessBase Chess interface (used by Fritz, Hiarcs, Junior, Shredder, Rybki, etc.), the principles we will be discussing apply to any chess program that has ability to analyze the game. In the first article, we'll look at the basics of creating and using the game's Analysis Features using the "Full Analysis" feature in ChessBase's production playing program interface. The second article we'll discuss using the same interface is the "Blundercheck" feature, which also provides a complete analysis of the game with output appearing in a slightly more complex (but also more useful) form. In the last article, we'll talk about the practical application of chess program feedback, such as how to use this information to help you improve your own playing skills.
If you want top-notch analysis with your chess engine, there are some things you'll want to do before starting a chess program. Do not run any other programs while your chess engine is analyzing - you are weakening the engine by doing so. This offer also includes any "background" (i.e. "Quit and Stay Resident") programs that may be running, such as screen savers, antivirus programs, "rumble guard", etc.
The next step is to run the chess program (as noted above, we will use the ChessBase production game programs for this article). Press F3 to access the list of available engines and select the one you want to use, we will be using Fritz in this article.
Complete Analysis
Once you have chosen your chess engine there are several different ways to proceed. One has to go to the game database list, double click on the game you want to analyze (load it into the main screen chessboard) and then go to the Tools menu, select "Analysis" from the menu, and then "Full Analysis" from the submenu. I do not recommend this procedure for several reasons. First, you cannot access the full range of "Full Analysis" options using this method. Second, you need to remember to manually save the game to the database after the analysis is complete.
Instead, I recommend the following procedure (which actually saves a couple of steps anyway). First load the database in which you have saved the game you want to analyze - hit F12 to open the game list box, and if the correct database is not shown, go to File/Open/Database to select the correct one. Once you have downloaded the appropriate database, find the game you want to analyze in the list and with one click on it is to place the cursor over the game to highlight it in the list. Then go to the Tools menu, select "Analysis" and then "Full Analysis" in the submenu. To do this, the following dialog will be displayed:
There's a fair few things to consider here! This dialog allows you to set the timing parameters and control the analysis output of your chess engine. Although this dialog may look complicated at first glance, it is really very easy to use. Let's take a look at the various sections of this dialog and explore what these options do.
Calculation of time and threshold
The first thing you should consider is the "Calculation" and "Threshold" options. Generally, the more time you allow your engine to calculate the deeper ("further ahead") it will look in that position - and you will therefore have a better analysis in response. However, there are potential downsides to setting the calculation time to either too high or too low.
First we need to understand what timing actually means. The value in this field is in seconds. If you set this to, say, "30", it means that your chess engine should (theoretically) analyze each move in the game for about thirty seconds on average. In practice, however, it does not work that way. Setting the value to "30" does not mean that the program will stop the analysis when it hits the thirty-second mark and drop the best option, it is in the game score. What this means is that when the thirty-second digit is reached the program will complete the analysis of the current depth layer before providing its analysis and move on to the next move. If the program has just begun, say, the tenth layer at the twenty-eight second mark it may take two minutes or more before it finishes estimating that tenth layer and moves on to the next move.
Thus, we can understand why setting the calculation time parameters too high can be a drawback - it may require many, many hours of calculation for the program to complete its analysis. However, setting the parameter too low (for example, to the value "5") will force the program to fully analyze the game very quickly (in a few minutes), but the quality of the program's suggestions will be quite low.
Adequate settings will vary from machine to machine and will require some experimentation on your part to discover. Chess engine analysis games are best done overnight - it's going to take hours for the program to provide decent quality analysis (six hours isn't too long). The trick is to find a convenient time for analysis without tying up your computer for ten, twelve or more hours. Start with the value "60" (as shown in the picture above). If you find that your program finishes parsing fairly quickly (say, within two hours on a 40-move game), you'll want to bump up the timing. However, if you start the analysis process, go to bed, come back at eight o'clock, and the program is still analyzing the middlegame of a 40-move game, you'll need to reduce the Calculate Time setting accordingly.
The threshold is given in increments of 1/100th pawn - in other words, the threshold value of "1" is equal to 0.01 pawn. Threshold allows you to control how much analysis the chess engine provides and the circumstances under which it will show you the best move. As it analyzes, the program will evaluate each position in the game and find the best move in each position. This will assign a numerical value to each position (ie "If White plays these changes, he will be better by 0.75 pawns").
The threshold shows the difference between the best line of play the chess engine finds and the move that was actually played in the game. For example, if you set the threshold value to "50", the program will display an alternative on a case-by-case basis in which the best line of play (as judged by the program) is better than the actual move by half a pawn or more.
So what value should you assign to the threshold? If you are a beginner chess player I recommend a value of "100", this will force the program to show you tactical misses where you have lost a particular material (eg a pawn or more). It's unlikely that a novice player will be able to figure out why a particular move is better than a fractional pawn value, and beginning players should focus on tactics anyway, so setting "100" will work pretty nicely, showing you the tactical mistakes you've made.
For the intermediate to advanced player I usually recommend a value of "30". Strong chess players and computer chess experts usually estimate the loss of tempo as equivalent to about a third of a pawn. Using a value of "30" will indicate these types of temporary loss of positional errors (as well as any other significant positional errors in nature).
Some players use very low values (like "1"), but I don't find this to be very helpful. If you're playing a "perfect game" (as if such a thing really existed), most of the moves you play can be improved by a chess engine of 0.05 to 0.10 pawns, and that's just too close a shave, of which most human players can reap any significant benefit.
Other options
Once you've set the "Calculation" and "Threshold" options, it's time to move on to the other switches in this dialog. The "Annotations" window allows you to select different forms that annotations can take. Let's start at the bottom of the list. "Delete old annotations" means exactly that - the program will remove any existing annotations in the game score. If you previously manually add any text, symbolic, or graphic commentary to a game (or any other previously annotated game you choose), this checkbox will cause that comment to be removed - so use this switch wisely.
Going back to the top of the list, "Verbose" means that the program will add some conventional language verbal commentary to the game. It is important to note that this comment is very rudimentary - the program will not give a nine-point dissertation on why you were not able to properly work with Maroczy's structure to bind the opponent's pawns. We'll show an example of a verbal comment to a program a little later.
"Graphical" means that the program will display colored arrows and squares on the board where it deems such a comment necessary. This usually takes the form of showing weak squares (by coloring them), or a control square (for example, you may see many arrows in an isolated pawn, showing the pieces that attack and defend that pawn).
"Training" allows the program created timed training of the subject at critical points in the game. These are usually in the form of tactic challenges in which you are asked to find the best move in a position. Please note that the program will not generate these questions in every game - in my experience I have seen them created once every twenty to twenty four games I have had the analysis program.
Database link
In the picture above, you'll notice that "Opening Link" is shown in halftone and unavailable. This is because I didn't assign a "Reference Database" before I created the illustration. You assign such a database by clicking " Reference-DB" button (visible at the bottom of the dialog) and database selection. Selecting the "Open Links" option will allow you to opt out of the program installed opening options from other games in your game account, as shown below:
In this picture you can see where the program added three alternatives to the game (as you often see in chess books and magazines) and even designated the move 5...e6 as a "theoretical novelty" (which doesn't mean that 5 ... e6 was necessarily a good move, only that this move was not found in the reference database games).
Note that when choosing a reference database, that database you choose must have an opening key attached to it in order for this feature to work properly. I've also found that the feature works best if the reference database is the one containing the opening-only games used in the game being analyzed - otherwise the program sometimes crashes into annotations at the very beginning of the game, which is from other unrelated openings.
You can select any or all of the options in the "Notes" section, selecting one does not "override" any others.
The radio buttons in the "Side" window are self-explanatory - you can choose a chess engine to analyze both player's moves or only one player's moves. My strong suggestion is that you always select "Both" - the program will work much better if you do so, and it's always beneficial for you to see how your opponent could have improved his game by punishing his mistakes.
You can only select one option in "Sid", selecting an option here prevents you from selecting any of the others.
Finally we come to the "Storage" options. "Replace" means that the program will physically replace your game in the database with a new, annotated version (for example, if you have game analysis program No. 320 in the database, old game No. 320 will be replaced A new version). "Add" means that the program will add the game to the database "tacking it on" as last game in the game database list (for example, you are analyzing game #320 in the 2474 game database. The program will parse Game #320, leave the current #320 untouched, and add its parsed game to the database as the 2475th match in the list). The disadvantage of using Append is that you end up with the same game twice in the database, once in its original form, and a second time in its annotated form.
Start analysis
After you set the options and the selected options in this dialog, click the "OK" button and the chess engine will begin to analyze your game. The display screen will change from the "game list" window to the main chessboard screen. This step is currently being evaluated is highlighted in the panel designation with a dark cursor. If you follow the process for a few minutes you will notice something interesting: the program starts to analyze at the end of the game and works backwards through the moves. As the program finds the best variations it will insert them into the game to score as many variations as possible can be re-played. When analyzing the completion process, the program will return to the database "list of games" display (if you started the analysis process from the list of games, as I recommended above), the selection cursor of the newly annotated game is how you "will know that the process will be completed.
When the analysis is completed, double click on the game score to download the game. Have you noticed that the program often uses a symbolic comment to show its appreciation of the proposed changes and the moves actually played. To understand analysis, you must know what these symbols mean:
You can see how much better the recommended line is by comparing the estimate of the move actually played with the estimate of the suggested changes to the chess engine:
Here we see an interesting phenomenon: the chess engine will sometimes show faint lines to illustrate a point. In this graphic we see that this move is actually played, 18.cxd5 leaves white with a significant lead. But if White were to capture the d5-pawns with the rook instead of (18.Rxd5), he would only be left with an equal game after Black's reply 18...a5.
Here is a screenshot of the notation panel to give you an idea of the type of comments that the chess engine will provide in the ChessProgram interface:
You can see that the commentary text (created because we chose "Verbose" as the "annotations" option) is very brief and is intended mainly to draw our attention to interesting and/or important points in the game. Sometimes the text describes the purpose of the move (as is the case with the notes after White's seventh move and Black's twelfth move). In other cases, the program's comment text simply warns us of points where one player is in trouble (white 21 and 23 moves). And sometimes the program will use text to indicate places where a player can improve his game (such as a change on White's 31 moves).
Now that we know how the "Full Analysis" option works in the chess interface, we'll look at how to "fine-tune" the Analysis and get even more specific information, albeit in numerical rather than verbal form. This "Blundercheck" analysis option will be explained in the second part of this article series.
(Part two)
Steve Lopez
In the first part of this article series, we covered the "Full Analysis" feature in the ChessProgram ChessBase interface (used by Fritz, Hiarcs, Rybka, Junior, and Shredder). This feature allows you to analyze your games and will provide general information about where you might have gone wrong in your game. The second form of analysis is called "Blundercheck", and is in many ways similar to the "Full Analysis" feature. "Blundercheck" will analyze your games and show you where you (and your opponent) made mistakes, but its output is in digital form, not verbal. This is the traditional display mode of chess analysis; it's been around since the first commercial PC chess software came out in the 1980s. This traditional numerical analysis is, in many ways, a much more accurate method of analysis in that it will show you the exact (up to 1/100th of a pawn) difference between the move you made and the recommended move that the chess engine determines to be best. Instead of a symbolic annotation that shows, in general terms, how much better the variations are recommended, numerical scores will show you exactly how your move and recommended lines differ.
Blundercheck
The name "Blundercheck" means that the program will only show you basic errors, but this is not the case. This analysis mode was originally conceived as a way for advanced players to check their own analysis, for example a chess writer can comment on a game and use "Blundercheck" to show errors in his variations as a means of "double checking" his work for "blunders" in your analysis. But "Blundercheck" is much more useful as a means for average players to get a better idea of what the chess engine is showing them.
Let's first show how to create and use the "Blundercheck" function, then we will descibe the power of the engine. Just like in the previous article for "full analysis", we'll follow similar first steps to get to the "Blundercheck" dialog. After launching your ChessProgram interface, press F3 and select the chess engine you wish to use for analysis. After selecting an engine, press F12 to open the game list window. If the proper database does not appear, go to the File menu and select Open /databases and use the file selection dialog to navigate to the appropriate folder and database file. Double click on the file name to open the database.
Once the game list is displayed, click once on the game you want to analyze, this will place a black cursor bar on that entry in the game list. Now go to the Tools menu, select "Analyze" and then "Blundercheck" to display the following dialog:
Some parts of this dialogue will be familiar to you (after reading the first part of this series). We've already discussed "Side Analyze" but I'll reiterate my best advice to always use "AND" - the program works much better in this mode and it will also alert you to mistakes your opponent has made (and which you might would be punished).
"Storage" was also discussed in the previous article.
"Exit" is a new feature unique to the "Blundercheck" feature. "Annotations as text" means that the lines played by the chess engine recommends being presented as text annotations - that is, they will look strictly like text and will not automatically be replayable on the chessboard when you familiarize yourself with the game. Hence Istrongly recommend another option: "annotations as variations". This will result in the recommended lines of the chess engine, which will be presented as replayable variations (as they appear when you use "Full Analysis") - you will be able to play variations on the chessboard when you familiarize yourself with the game later.
"Time" setting is the same as "Calc. Time" in "Full Analysis" mode and the same recommendations apply here as well. An additional setting here is "Depth", the engine will always analyze to the layer depth you set in this field - never less, never more. "Time" and "Depth" are mutually exclusive, you can set one or the other, but not both. My recommendation is to use the "Time" setting instead of the "Depth" setting; the use of the latter often results in short changes that are cut off in "forcing" lines (eg, in a series of checks or captures).
"Threshold" works in "Blundercheck" mode exactly as it does in "Full Analysis" and the same recommendations apply here.
A set of checkboxes follows "Threshold" and gives you a nice bit of latitude in how the chess engine will display its output. "Write full variation" is an interesting setting. Checking this box means the chess engine will display a complete change (with steps for both sides) when it finds an improvement over what you or your opponent actually played the game. If you uncheck this box, then the program will only display the initial step when it finds something better than the game itself. Seeing only the first move is not very profitable for the average player, you will often find yourself asking, "Why was it better to move?" Hence, I urge you to keep this box, so that you will see "observation" moves to the best of that initial movement.
"Remove old annotations" works the same as in "Full Analysis" and applied to previously annotated games, the program will remove everything before the annotation from the gamescore.
"Training" was described in the previous article, this allows the program to generate timed training questions as part of its analysis. This tends to happen about every twenty to twenty-four games on average.
"Store scores" must be checked: this allows the program to add its numerical scores to the changes it inserts into the gamescore. Unchecking this option actually defeats the entire purpose of the "Blundercheck" feature for the average player. The option to omit these numerical scores is included primarily for grandmaster-level players, who are presumably able to make these scores on their own.
"Check the backbone" tells the program to evaluate the steps that were actually played in the game of two players. This should always be checked. "Check Variations" checkbox for players who want to have the chess engine "recheck" any changes they manually add to the gamescore; this field is useful for authors/commentators who wish to have an engine check their work for errors.
After you set the options in this dialog, click the "OK" button. The program switches to the main chessboard screen, jump to the last move in the game (as described in the previous article), and start the analysis. This will be a step back in the game, adding variation and numerical analysis at the points where she finds the best line of play. When the program finishes analyzing the game it will return to the game list screen, where you will see the game again highlighted with a black cursor. Now you can double click on this game to download it and see, analyze the game:
Note that Fritz (the chess engine used to analyze this particular game) inserted a variation of five points into the game on which he found the best move for any player (based on the "Threshold" setting provided when the analysis parameters were set). It is interesting to note only the text of the annotation reads: "The last move of the book." This means that 4.Be2 was the last step that was found in the opening of the book that was loaded at the time the analysis was started. Black's reply, 4...a6, was not found in the book open.
To understand the numerical analysis of a chess engine, let's take a closer look at one of the options it provided:
Analysis with reference to Black's seventeenth move (17... exd5). Numerical analysis is always given from White's point of view, positive numbers mean the position is favorable for White, while negative numbers mean the position was better for Black. In this case we see that White is enjoying the 94/100ths advantage of the pawn (0.94) after the actual move of black 17...exd5, this means that White is almost a full pawn ahead according to the chess engines. But Black could improve on this play with 17...Bxc3. The program continues to give a recommended option in which it assumes the best to play for both sides. After this sequence of moves that ends with 22.Qc4, White will still enjoy the 44/100ths advantage of the pawn - but note that this advantage is much less than after the actual play. With 17...Bxc3, White was almost a full pawn ahead, but in the change after 17...Bxc3, White's advantage would have been less than half a pawn. Black would have reduced White's advantage by exactly half a pawn (0.94 - 0.44 = 0.50) if he had played a bishop capture instead.
The number "13" after the numerical variation score tells us how deep the search went ahead of the chess engine arrived at the given score. In this case, the engine analyzed the position after 17.d5 at a depth of thirteen layers (half moves) to come to its conclusion on how Black could have improved his play.
We can easily see how accurate this information is compared to the Full Analysis output. While "Full Analysis" is a bit more readable, "Blundercheck" gives us more accurate information. We know exactly how much better the suggested variation is compared to the move actually played, and we also know exactly how deep the search engine is to arrive at its score. Therefore, "Full Analysis" is well suited for beginners or inexperienced players, while intermediate and advanced players will enjoy and benefit from the accuracy of the analysis provided by "Blundercheck".
In the final article of this series, we will learn how to apply the information engine to the challenge of improving our own chess game.
(Part Three)
Steve Lopez
In this part, the final installment of a three series of articles, we'll look at how you'll use a chess engine (such as Fritz, Rybka, Shredder, Junior, and Hiarcs) to help you improve at chess. This is not going to be a software tutorial per se, we won't be looking at "click on x to make y happen" instructions since we covered those in the previous two articles. We instead explore how you will use the output generated by the game's analysis features to help you improve your chess.
Someone once said that "the first step to knowledge is to admit one's ignorance" that statement certainly applies here as well. In order to profit from having a chess engine analyze your games you first need to make a (perhaps painful) recognition that there is a lot about chess that you don't know but need to learn. Over the years I have spoken to more than one user of chess programs who have used parsing features like "ego booster", admiring moves that the chess engine considers "correct" while ignoring unsatisfactory moves (or outright errors) that the software software discovered.This approach may be "Chicken Soup for the Soul", but a real waste of a valuable chess improvement tool.To get the most out of the analysis engine own games You must first admit to yourself that your chess needs improvement - there is no other way.
In fact, the whole process is based on the fact that you have already decided that something is wrong with your chess and you want to correct the mistakes. What we need to do now is to clarify the process: how will we use the engine generated analysis to improve?
Contrary to what several development companies would have you believe, no piece of chess software alone is guaranteed to improve your chess scores. I understand that more than a few players are looking for a "magic bullet": a single book or piece of software that, in itself, make the player a kind of "moment master". Sorry, but it's a faint hope that it just won't happen. What we, as players, as well as learners, need to do is find a way to integrate chess study and chess practice into a method for improvement. In fact, that (and the hard work it entails) is the key to getting better at chess.
Let's take a closer look at this idea. Improvement in chess is actually a three-step cycle of processes:
Practice
Analysis
To study
No one part of the software will help you in all three areas. Chess gaming software excels in helping you with practice (you can play games at your convenience) and analysis (you can also have engines to analyze your game in your spare time). Although some chess programs contain tutorials on various aspects of chess, this is usually geared towards absolute beginners and inexperienced players. For more high level instructions suitable for intermediate players, you need to refer to books and specialized chess learning software. This brings us to another important point. Chess books and chess programs are not mutually exclusive, it's easy enough to combine chess books and software using the best of both effective training development environments. We will return to this idea in a while. First, however, we must consider the "Learning Cycle" to understand the three processes.
Practice applies to any chess game you play. Within the learning cycle, "practice" does not only mean games that "do not count" (such as games against the computer or impromptu games you play for a chess club or against a friend). "Practice" refers to the practical application of existing chess knowledge, that is, the application of what you know to the actual game environment. Anytime you play chess (as opposed to the tactics of solving puzzles or "checkmate in x" problems, etc.) you are practicing what you know. That's what we mean by "practice".
Analysis means looking at your games after you play them, looking at them to discover flaws in your game. While it's always nice to look at a three-move combination that won your opponent's rook and allowed you to win the last game you played (and we all wanted to admire what we did right), it's more important to look at the rest of the game to see is there anything we could do better. It's ironic that chess has a reputation for being a game for egoists, because there are few other pursuits that require you to be as sharply self-critical as chess claims.
Research is exactly what the word means: the process of learning new techniques in order to correct one's shortcomings. "Research" can mean reading books on positional chess, solving tactic problems with CD chess and/or analyzing the games of great chess players, these are any actions we take to increase our knowledge of the "nuts and bolts" of our own way of playing chess.
Hello dear friends. With you dad Zhorik.
In today's article I will tell you about how we spend with Zhorik chess analysis played game online. At what the analysis passes very qualitatively.
The computer shows with arrows the moves (where it is better to go), where the error is. Shows “+” or “-” with numbers, immediately finds mate options in a certain number of moves, sacrifices, combinations and everything like that.
And everything is very convenient. Played - pressed the button - you analyze the game for each move. The computer is not a fool, it analyzes everything very well. Don't think that you are smarter than him. =)
In fact, now there are a lot of different free and paid programs both in Russian and in English, where all sorts of engines are connected. There are services, etc. But personally, both me and Zhorik like to analyze everyone more at lichess.org.
If you play on a computer, then it looks like this:
And if on the phone (iPhone), then like this:
The scheme is simple if you play on the lichess.org site itself. Played, after the game you press - analysis:
And by clicking on each move in the table with the mouse, look at what the computer shows you. The chess game will be analyzed by the Stockfish 8.0 engine. In fact, a very cool engine, so you can be sure of its quality of analysis.
See an example of a game that I played with Zhorik. They tested him for knowledge of the trap in defending the pawn on e5, with the pawn on f6. The game was: 1. e4 e5 2. Nf3 f6 3. Nxe5 fe:
You see, the computer shows with an arrow they say to move the next move, he advises the queen on h5. In terms of position, he also estimates it as +3.6 in favor of White.
In fact, you sit like this after the game, look at your mistakes and understand how easy it was to win, however.))) The opponent made mistakes here and there ... Eh ... I wish I could turn back time. I would arrange for him.)))
Thanks to chess analysis, your level of play increases. You start to find good moves, you start to see sacrifices, good combinations, etc.
Farther. If you don’t play on lichess, but for example, somewhere on another site or even offline with a friend in a chess club in your city or at some competitions, you sit down and write down the game on the form and want to analyze it, then again, with lichess is not a problem.
If you have a pgn file, then you can import it into lichess and analyze it in the same way:
Also, if you do not need analysis from the very beginning of the game, but want to analyze some chess position and find out how it was possible to make a better move than you actually did, then everything is just as simple here.
Come in board editor and choose whose move:
Clearing the board:
Set the desired position by dragging the figures onto the board:
Click the "Analyze" button. As a result, this is what kopm shows me:
Checkmate in 4 moves. With a rook sacrifice.)) These are the pies.
Analyze, practice and improve your playing skills. In conclusion of the article, I suggest that you analyze the games of Sergey Karjakin and Magnus Carlsen.
Watch the games here, and make moves on lichess.org under stockfish analysis. I also advise you to analyze various.
That's all for me. Wait for new articles. We will further describe other possibilities of computer analysis. For example, there is such a mega cool program - Chessbase.
See you soon...
I was inspired to write this program by my once favorite chess site chess.com. My account allowed there to do only "Deep" analysis of my games (2-3 minutes per game), and not "Maximum" (4-6 minutes per game). In any case, the computer analysis on chess.com is slow, because it uses not the stockfish binary engine, but its implementation in a translatable javascript language. Thus, the results of game analysis obtained on the chess.com website are often unsatisfactory. Comparison of the results of analysis of games on the site chess.com and local analysis by binary Stockfish is far from in favor of the first.
Curiously, immediately after the publication of this program, my account on chess.com was closed without any explanation and the return of the annual membership fee! (I had to claim it later myself). Did my program cause them serious financial damage? I would be ashamed in their place to make money on poor-quality analysis chess games(made in addition by the free Stockfish chess engine!) and impose a time limit on the analysis of the game, depending on the amount of money paid for membership. We can only wish chess.com success in their quest for "fair" play!
In addition, an analysis like Chemov's just doesn't exist or doesn't really work in Windows chess programs. In "SCID vs. PC" it doesn't exist, and in Chessbase "deep analysis" doesn't work at all! I wonder what the developers think about this?
Therefore, I decided to write my own chess game analyzer, similar to Chemovsky, only much more nimble, easy to install and easy to configure and use.
The functionality of the analyzer of chess games "Creatika"
Main features and limitations
- Easy to install
- Simplicity and ease of use. Just right-click on the pgn file and select "Analyze". The file resulting from the analysis will automatically open in the graphical application for pgn files by default
- New in version 4.1! Graphical user interface for selecting batches for analysis
- New in version 4.1! The best move of the engine is now stored in the best_moves.db database for quick retrieval. No need to repeatedly waste time analyzing a known position
- New in version 4! pgn-extract.exe and coreinfo.exe are no longer needed. The Creatica chess game analyzer is now able to independently convert from one chess notation to another and determine the type of your processor. Microsoft. NET framework 4.6 is still needed, as is the SQlite library that is included in the distribution
- New in version 4! GUI for settings - no more need to edit the chessgame-analyzer.exe.config file with a text editor
- New in version 4! Deep analysis - the chess engine will play itself with itself several games after each move and determine the best move statistically based on the results of the games played. Requires a powerful computer and enough time to use deep analysis. Played games are stored in the database and can be used to analyze other games
- New in version 4! An improved version of the database SQlite compiler is now included in the analyzer
- New in version 4.1.0.13! SQlite database KingBaseLite.db updated to 01/19. It includes the first 20 moves of KingBaseLite.pgn games played by opponents with an Elo rating of at least 2300 and a rating difference of no more than 200. This database is designed to search statistically the best options in openings
- Attention! SQLite database for version 4.0.0.0 is not compatible with database for version 3.0.0.0! Please download the new database
- New in version 4! A set of chess problems for the Arasan engines in PGN and EPD formats is supported - you can compare the abilities of your favorite engines to solve chess puzzles
- Please note that the format of the annotation template files has changed. Therefore template files for version 3.0.0.0 are not compatible with template files for version 4.0.0.0
- New in version 4.1.0.13! Included the latest Stockfish 10 binaries at the time of release of this version (development build on Feb 8, 2019)
- New in version 3! Ultra-fast search for statistically best options across a large chess database
- New in version 3! Chess database compiler (chessdb-compiler.exe) for quick search of statistically best variants from chess databases in pgn format
- New in version 3! The ability to download a ready-made compiled database for ultra-fast (hundreds of times faster than in Chesbase, for example) search for the statistically best options (includes more than a million games since 2000, Elo rating of players is not less than 2200, details at http://www. kingbase-chess.net)
- New in version 3! Annotation templates allow you to annotate games in any language in any style
- New in version 3! As an example, the program comes with three well-documented templates in Russian and English
- Professional - practically without words, only options with a symbolic score at the end of the option
- Semi-professional - short monotonous comments - used by the program by default
- Amateur - more detailed various comments to your taste
- New in version 3! Ability to display multiple options - set by the multipv parameter in the configuration file
- New in version 3! Symbolic position evaluation at the end of a variant
- Support for any UCI-compatible chess engines
- Auto-tuning of most parameters. Possibility manual setting via config file
- Russian and English interfaces
- Automatic (without user intervention) analysis of a large number of batches
- PGN standard support
- Support for any Windows encoding PGN, ECO files and annotation templates
- New in version 3! Updated binaries (with fixed bugs and increased performance) of the default engine - Stockfish 8, the strongest to date (my build of sources from GitHub from September 7, 2017
- Time per move, rendering depth, number of threads and memory size for the engine
- Possibility to start the analysis of games starting from the initial position, from any move
- Ability to analyze games starting from any position
- Configurable chess engine process priority
- Game classification in Russian or English (ECO code, name of opening and variation)
- The ability to use your own classification file by specifying it as the value of the eco_file parameter
- Support for Fisher chess (chess 960). Chesbase reads them correctly, but Scid vs PC gives errors - does not support.
So, my program is a console application under MS Windows. The results of game analysis can be seen on the screen immediately, but it is more convenient to view them in a graphical chess application that can read PGN files, for example, in Chessbase or .
The program comes with a free stockfish engine, the most powerful to date. You can also configure the application to use commercial engines such as Komodo or Houdini (not supplied with the program), as well as any other UCI compliant chess engine of your choice.
During batch analysis, the following information is displayed:
- program name, version and developer
- titles (tags) of the party
- the name of the chess engine and the names of its developers,
- number of threads and hash size of the engine in MB,
- moves with position evaluation, depth of analysis and comments.
Each move is by default accompanied by position evaluation and calculation depth. If the difference between the estimate of the best move and the move made in the game is greater than certain threshold values, then a comment about inaccuracy, error or gap is displayed. The best move and continuation are also indicated as a variation. At the end of the variation, the estimate of the best move, the depth of calculation, the number of nodes in MU, and the speed of analysis in MU/s are reported. If the engine sees checkmate, then the moves leading to checkmate will also be shown.
When the difference in the assessment of the positions of the parties exceeds one pawn, encouraging comments are included. If the move matches the best move of the engine, then congratulations are displayed. If the difference in score between the best move and the move made in the game does not exceed 0.2 pawns, then a good move and a better continuation are reported. Encouraging annotations differ in their tone depending on whether the party wins or loses.
The results of the analysis are saved in a new pgn file, the name of which consists of the name of the original pgn file and the suffix "-analyzed_<движком>" (or "-analyzed_by_
After the analysis is completed, the results will automatically be displayed in the default graphical chess application for pgn files, such as Chessbase, SCID vs PC, or any other that understands the pgn format.
Download , Installation , Use and Setup
Programs used by the application chess analyzer "Creatika"
Stockfish 10 ( latest builds developers)
The chess game analyzer comes with the free Stockfish chess engine, which is by far the strongest. It consists of four binaries:
- stockfish_10_32bit.exe - for 32-bit version of Windows
- stockfish_10_x64.exe - for 64-bit version of Windows
- stockfish_10_x64_modern.exe - for 64-bit version of Windows running on a computer with a processor that supports POPCNT instructions
- stockfish_10_x64_bmi2.exe - for 64-bit version of Windows running on a computer with a processor that supports BMI2 instructions
By default, the chess analyzer will automatically select the optimal binary.
Lecture read by the Honored Coach of the USSR in Loo on April 17, 2014 at the coaching seminar
Few people dispute the necessity and usefulness of the analysis of the games played. This analysis makes it possible to find committed mistakes, unravel the planned operations of the parties, determine the critical and turning points of a chess battle, and as a result of all this, draw conclusions that help to further improve the game of a chess player.
I started doing this kind of thing fifty years ago, when, after graduating from the Moscow Power Engineering Institute, I had to move away from active game in competitions due to the heavy workload of engineering work in a very serious design bureau. In 1973, I nevertheless returned to the chess world, accepting an offer to become one of the coaches of the USSR national teams.
Well, at the beginning of 1976, fate gave me a wonderful chance - I began to work closely with a little prodigy from Baku, who in a couple of years was recognized by the whole chess world. It was Garry Kasparov.
I will not talk further about the details and method of working with this super talented young man. Let me just say that the analysis of the games played, both joint and separate, was one of the main forms of his training process.
Clara Kasparova, Garry Kasparov, Vitaly Melik-Karamov and Alexander Nikitin (Vilnius, 1984)
Garry's rise fell on the endgame of the pre-computer period of chess, when the quality of analytical work depended only on the chess player's desire to get to the bottom of the truth, sometimes spending a lot of time and effort on it. From an early age, Harry had a great love for chess and a constant desire to learn as many of their secrets as possible. Analysis different kind positions that arose at different stages of this wise game, attracted first the boy, and then the young man to the same extent.
The results of our analyzes were recorded by Garik in thick notebooks. I, having managed to create a punched card file in the Central Chess Club (which is on Gogolevsky Boulevard), entered my analyzes into my own punched cards, which became the first attempt to facilitate the collection and storage of the necessary chess information and, most importantly, the quick extraction of its individual copies from a large array of punched cards, on which the specific information I needed was stored. In the seventies, the flow of information grew larger and larger every year, and there was a real threat of drowning in this stream of new games and analyzes published in books, chess newspapers and magazines, in special issues, etc. The invasion of computers into the world of chess rescued us.
At first, these were primitive game modules, of little use even for the role of sparring partners for a serious chess player. But the whole world was waiting for the appearance of computers capable of solving problems that were not chess at all. Since chess made it possible to study the process of playing from records of chess moves, they became a testing ground for working out serious analytical programs intended for use in science, management, military affairs, and in many other areas of human activity. Analytical programs designed for chess were not far away. True, the first computers only allowed chess pros not to carry a heavy suitcase full of magazines, informants, analytical notebooks, etc. to competitions, but they were not adapted for the work that is now familiar.
My first computer was an Atari 1040, which in 1987 I spent about most of his fee received at the end of the Seville, the fourth match of two Ka. Now I could go to competitions and training camps with a small suitcase where my computer was packed, in a small hard drive of which all the chess information I needed on the trip was stored. I was happy with it and took great care of my computer. Back then, computers were extremely expensive. So, my colleague, returning from Seville, exchanged the same Atari for brand new Zhiguli. My Atari served me faithfully for five years, until in 1992, when I visited the editorial office of the Dutch magazine "New in Chess", I bought "at a reasonable price" an advanced computer for that time on the 386 SL processor. It was a real laptop that allowed you to work with the first versions of "Chess Base" and "Chess Assistant". True, his weak memory did not allow him to think about any advanced forms of work - operative memory of 1.0 megabytes, and the hard drive memory of 40 megabytes allowed him to use it only as a repository of valuable chess information that I had selected. I quickly got out of the habit of writing on paper, I did all the comments of games and analysis of openings on the computer and automatically saved it in its memory. This went on for the same five years, until the parameters of my computer friend began to noticeably yield to his new, more powerful and faster-firing brothers. The coaching problems that I had to solve became more and more difficult, my bank of chess information required moving to a new, more spacious room, and I had to change my computer.
It is generally accepted that computers become obsolete very quickly, and their useful life is approximately two years. I calmly work with my silicone assistants for five years, quickly get used to them, learn their features. They become my friends, with whom it is a pity to part. So, if your chess problems don't get more complicated, don't rush to part with your old computers. No wonder they say that an old friend is better than two new ones.
For the next five years, I had a Pentium 266 computer that allowed me to run serious Chess Base 7.0 and Fritz 5.0 programs. Their use made it possible to feel more confident in the analysis of opening schemes. True, the scores issued by the computers were sometimes very controversial, and I immediately made it a rule not to follow their lead, but to rely on my understanding of the position.
Honored Coach of the USSR, FIDE Senior Coach, International Master Alexander Sergeevich Nikitin
Now, for two years now, I have been closely friends with an ASUS laptop, the heart of which is a dual-core Intel (R) Core TM i3 CPU, which has a clock frequency (2.4 megahertz) quite sufficient for chess research; more than sufficient and hard disk capacity (40 gigabytes). I am quite satisfied with his work: both in terms of the good accuracy of his estimates and in terms of the time they were received. Now there are computers that can get the same results twice as fast, but I don't care, I don't like to rush my friend.
I must say that my experience with the latest analytical programs such as "Houdini 4", "Stockfish 4", "Komodo 5" allowed me to draw conclusions that you should think about.
1. No computer can improve your game. He can warn blunders during the analysis and at the end of it (the termination of the analysis is determined by you, again based on your understanding of the position), give a certain amount of information with your purely formal numerical estimates (there are always several of them), which you must check based on your understanding of the position and, most importantly , comparing with your own, "human" assessment, make final choice.
2. Leading in the analysis should be a person. It must be firmly grasped that the computer is a conscientious accountant, and you are the thinker and chief in the search for chess truth.
3. It's funny to me to see young chess players playing as a Candidate Master, but already having powerful computers with quad-core processors. For them, it's like playing games with crystal or gold chess, instead of the usual wooden pieces. On such computers it is good to spend time on challenging games type " star wars", etc., but that, as they say, is another story. To solve chess problems, dual-core processors are quite enough. It's better to buy good chess books with the money saved.
Next, I will try to simulate the process of analyzing a chess game using a computer. To do this, I took a game of the old masters, not at all long, not infallible, but without rude blunders. It had already been analyzed by Grandmaster Nunn 6 years ago, and I wanted to see if my more powerful computer would help me find something new in covering the events that took place in this interesting and instructive game. I note that during the lecture, of course, I did not show variations longer than 5-6 moves; they are presented below for independent analysis by trainers and students.
Young listeners in Loo
Rudolf Shpilman - Arnold Denker
French C06 Defense
1.e4 e6 2.d4 d5 3.Nd2 Nf6 4.e5 Nfd7 5.Bd3 c5 6.c3 Nc6 7.Ndf3 Qb6 8.Ne2 cxd4 9.cxd4 Bb4+
Some players are itching to start a fight without first maneuvering. True, most of them have ratings that barely climb over 2100 - for this, it is enough to look at a selection of games from the Megabase. In my comments, I operate with examples of the play of stronger chess players who are able to penetrate into the understanding of the position, in computer language, on about greater depth in less time.
The loss of castling here cannot have serious consequences, since the position in the center is stable. In the meantime, it is stable, the pawn e5 provides White with a space advantage, and Black the inconvenience associated with crowding in his own home, part of which is shot through by the enemy. I would like to somehow remove this pawn, but not very decisive ones are frightened by the opening of the game and the threat to their own king. This is how a conflict of interests immediately sets in and a chess Maidan arises.
Serious players prefer to get a small positional advantage, but not lose castling. 10.Bd2 (+0,10) Bxd2+ 11.Qxd2 Qb4 12.Rc1 Qxd2+ 13.Kxd2 Nb6 14.b3 Ke7 15.h4 Bd7 16.Rh3 Nb4 17.Bb1 a5 18.a3 +/=, Keres - Flores, 1939.
10...0-0?! (+1,15)
Rudolf Shpilman, in his meager comments, awarded the castling of the black king with a single question mark and called this natural move a decisive mistake, allowing the standard bishop sacrifice on h7. Indeed, he was not mistaken in speaking of a mistake; there are better moves than what he did. Black's castling in this moment- a move in which there is a lot of provocation, since a more or less experienced player's eyes immediately light up at the sight of the h7 pawn. "To hit or not to hit?" - immediately arises an ancient question.
Let's talk about computers first. My computer assistants - the snooty "Stockfish 4" and the solid "Komodo 5", as I expected, were not united in express assessment of the consequences of the elephant's seductive sacrifice. Stockfish decided on a sacrifice instantly (11.Bxh7+), immediately giving her a powerful support "+1.21". Komodo didn't notice this move at all, compared to the solid general strengthening 11.Be3 and two similar moves given to me to choose from. I decided to give them time to work. Only after 15 minutes, having reached the calculation depth of 25 half-moves, Komodo deigned to see the bishop's sacrifice and its benefits, immediately putting the move on the first line with a score of "+0.79". At the same time, he indicated a variant proof more convincing than that of his competitor Stockfish. Next, I decided to see at what depth of calculation the computers would stop moving further. After an hour of work, they sharply slowed down the speed of their calculations, and it no longer made sense to continue their operation. I had to record the following results of their work: "Stockfish 4" reached a depth of 31 half-moves and retained its former enthusiastic attitude towards the bishop's sacrifice when evaluating the move of the second line (11.а3), which is also quite high (+ 1.05). Komodo 5 was more thoughtful. He reached a depth of calculation of 26 half-moves and retained respect for the bishop's sacrifice, leaving it on the first line (+0.88), far from the second line (11.Be3 with a score of +0.31). At the same time, it's worth noting that Nunn's best "human" move (11.Nf4) has never received the attention of bookkeepers.
Having learned the opinion of the computer, we now continue our purely chess analysis. The main opening idea of the French Defense is to undermine White's pawn fortifications in the center. Since the d4-pawn was successfully repelled, Black must hurry to break the e5-pawn, which unpleasantly hampers the movement of black pieces. This goal is served by advancing the pawn f7-f6, which must be carried out as quickly as possible.
1) 10...f5?(+0.90) A fundamental error that does not require a variant proof. Black refuses to undermine the pawn center and must now endure ordeals in a cramped position without counterplay. The hands make the first moves of the winning plan, giving the head a rest: 11.h4 0-0 12.Rh3 h6 13.Rg3 Kh8 14.Be3.
The computers that followed my moves here give a cruel assessment of Black's opening strategy: "+2.04 or +-". I agree and stop further analysis.
2) 10...f6 (0.56)
Let's see how Stockfish4 works again: 1 minute - calculation depth 20 half-moves, position evaluation "+0.58"; 2 minutes - 22 half moves and "+0.08"; 6 minutes - 26 half-moves and a score of "0.00".
11.Nf4 (11.exf6 (-0.10) Nxf6 12.Bf4 0-0 13.Qb1 Be7 14.a3 Bd7 =)11...fxe5 (11...Ndxe5?! 12.dxe5 fxe5 13.Nh5 0- 0 14.Be3 Qd8 15.Ng3 h6 16.h4 Bd6 17.Rh3 e4 18.Nxe4 dxe4 19.Bxe4 +/-) 12.Nxe6!? g6 (Looks extremely suspicious, but nothing better) 13.dxe5 Ndxe5 14.Be3 Qa5 15.Nxe5 Nxe5 White seemed to be doing everything right, but his hopes for an advantage seem to be fading away. 16.a3 (16.Nd4 0-0 17.Be2 Nc4 =) 16...Bxe6 17.axb4 Qxb4 18.Bd4 Bg4! 19.Qa4+ Qxa4 20.Rxa4 Rc8 21.Bb5+ Nd7 22.Bc3 a6 23.Bxa6 bxa6 24.Rxg4 0-0 25.Rd4 Nf6 26.f3 Ne4 =.
3) 10...Be7!(0.00) The bishop has done its job - the bishop can move closer to the king. 11.a3 f6! ( As an example of an unsuccessful staging of the opening for Black, I present a fragment of the game of two great classics who played this game at the end of their chess career. For some reason, Black immediately went into a deaf passive and meekly perished. 11...Nf8? 12.b4 Bd7 13.Be3 Nd8 14.Nc3 a5 15.Na4 Qa7 16.b5 b6+- (Alekhin - Capablanca, 1938) 17.h4! f5 18.Rh3 h6 19.Rg3+-)12.Nf4 fxe5 13.Nxe6 Bf6. The opening development is suddenly interrupted by the beginning of the battle in the center of the board. 14.Qe2 e4 15.Bf4 Nde5! 16.dxe5 Bxe6 17.exf6 0-0! 18.Ng5 Rxf6 19.Nxe6 exd3.
In this acute position, we began testing the operation of the Stockfish4 accountant in a single-processor stationary computer. After one minute, the engine reached a calculation depth of 18 half-strokes; ten minutes later he was at a depth of 27 half-moves with an estimate of 20.Qg4 "+0.14". After another 20 minutes, he stopped at a depth of 31 half-move with an estimate of the same move "0.00".
20.Qg4 Rg6 21.Ng5 Qxb2 22.Re1 d2. Computers give here an assessment of "=". But for me the position is not so clear - it is too difficult to understand what is happening here, and I rate it "unclear". However, if we continue the variation - 23.Rd1 h6 24.h4 hxg5 25.hxg5 Re8 26.Kg1 Qc2 27.Rxd2 Re1+ 28.Kh2 Ne5 29.Qh3 Rxh1+ 30.Kxh1 Qe4 31.Qc8+ Kh7 32.Qh3+, we get a perpetual check.
11.Bxh7+ (+1.15)
White is in a hurry, maybe it's not in vain. Indeed, on the next move, one of the black pawns - f7 or h7 can go to the 6th rank, and then the possibility of sacrificing the bishop will become irrelevant. Of course, the hand reaches out to “slam” on h7, after which the white queen bursts into the fortress, where the enemy king is hiding, and he has to flee. It was this circumstance that allowed the great lover to attack and consider the bishop's sacrifice justified, for he believed in his ability to attack and believed that “further attacking opportunities will be found automatically. Having sacrificed the bishop, I was convinced that my attack must somehow lead to victory.
However, not all so simple. Black's position is unpleasant because his pieces are in some kind of disorder, making it difficult to organize counterplay. Taking advantage of this, White can look for a way to strengthen the attacking grouping. And then it is easy to pay attention to the jump of the e2 knight.
11.Nf4! (+0.90). This calm enhancement of the attacking potential was found by Grandmaster Nunn, who did not believe in the effectiveness of the sacrifice of the white bishop, a piece, by the way, very useful in attacking the king. If the attack bogs down, then the absence of a lookout for white fields can create serious problems for them.
a) 11...f6 12.Nxe6 fxe5 ( 12...Re8 13.Nf4 fxe5 14.Nxd5 +-) 13.dxe5 Ndxe5 14.Bxh7+ Kxh7 15.Nxf8+ Bxf8 16.Nxe5 Nxe5 17.Qh5+ Kg8 18.Qxe5 +-;
b) 11...h6 12.Bb1 Re8 13.Kg1 Nf8 14.h4 Qc7 15.Rh3 Bd7 16.Rg3 Ne7 17.Nh5 Nfg6 18.Nxg7 Kxg7 19.h5 Rec8 20.Bf4 +-;
With) 11...Be7 12.h4 h6 13.Rh3 Re8 14.a3 Bf8 15.Rg3 Ne7 16.Kg1 a5 17.Ne1 Nf5 18.Bxf5 exf5 19.Nxd5 Qc6 20.Nf4 +-.
I “showed” Nunn's move to my accountant when, after 12 minutes of calculated work, he still did not highlight it on the monitor, and the move made by Shpilman remained the favorite, with an estimate reaching “+1.27”. The computer reacted positively to such a “kick”, but without enthusiasm, giving a rating of “+0.94” after 8 minutes of calculation at a depth of 25 half-moves.
11...Kxh7 12.Ng5+ Kg8
The game ends instantly if the king attacks - 12...Kg6?? 13. Qd3+ f5 14. Nf4+! Kxg5 15.Qg3+ Kh6 18.Qg6#.
13.Qd3! (+1.03) Re8 14.Qh7+ Kf8
The first phase of White's attack is over. Now they have to choose from several very tempting sequels the best one. There is practically no one to protect the black king, the only staging post - the square e7 is a saving loophole for the further flight of the monarch, if he succeeds. On the other hand, only two white pieces are involved in the attack so far. But this pair is only good for mating the king around the f7-square. White solves the problem of this point, but not in the best way.
15.Qh8+?(-0.16)
It cannot be said that this malfunction is obvious. However, this is indeed a glitch. The white queen did the wrong thing. Fascinated by the chase, he drives the enemy king out of the danger zone to a relatively safe temporary shelter on d8, from where a path leads to a safer shelter on b8. As a reward, he will get two unfortunate pawns on the kingside and clearing the way for the white passed pawn to move to the queen.
However, further we will see the divergence of the winner's statement: "The idea that every sacrifice should be calculated as accurately as possible is fundamentally wrong. When attacking, only faith in the position (?) and in oneself is required," with the conclusions of careful analysis, which proves that here White missed two much more effective continuations of the attack.
First of all, it should be noted that an attempt to immediately win back a piece given up for the sake of attack by moving 15. a3 reveals a defect in White's position - a weakening of the d4-pawn. This circumstance can be used by Black to great advantage - 15…Nxd4! 16.axb4 Qb5!, and after the forced exchange of queens, White is left without a pawn and without initiative.
Having discussed the bad opportunity, now let's talk about the good ones.
1) 15.Qh5!? The winner, who is very pleased with the attack brought to victory, does not even mention another version of the scenario he invented, when White gets to the f7-pawn a move earlier. This time saving sharply strengthens the attack, making it irresistible, for example: 15...g6 (+2.20) 16.Qh7 Ke7 (16...Nd8 17.Ne4!!+-)17.Qxf7+ Kd8 18.Nxe6+ Rxe6 19. Qxe6 Qb5 20.Be3+- or 15...Nd8 (1.09) 16.a3 Qa6 17.Be3 Ke7 18.Rc1 Ba5 19.Qh7! Rf8 20.h4! Nc6 21.Qxg7 Bb6 22.Ke1! Ke8 23.Rh3!+- Qd3 24.h5 Bxd4 25.Bxd4+-.
2) 15.Nf3! (+1.27). The knight, having jumped back, not only strengthened the defense of the central pawn, but, most importantly, gave way to the bishop, and its appearance on the g5-square threatens to immediately end the battle.
Comps do not immediately realize the power of this maneuver, and for about half a minute I see their indignation. Their favorite (15.Qh5) was displaced from the first line only at a calculation depth of 25 half-moves, and immediately with a knight's jump estimate of "+1.90", and after calculation for 27 half-moves, the estimate reached "+2.16".
Such a combination of defense of the d4-pawn and attack, with the involvement of the dark-squared bishop, was not seen even by Nunn. 15...f6 16.Bh6! Ke7 17.Qxg7+ Kd8 18.exf6 Bf8 19.Qg6 Bxh6 20.Qxh6 Qxb2 21.Re1 Kc7 22.f7 Re7 (22...Rf8 23.Ng5 Nd8 24.h4 b6 25.Nf4 Ba6+ 26.Kg1 Kb7 27.Nfxe6 Nxe6 28.Qxe6 +-)23.f8Q Nxf8 24.Qxf8 b6 25.Kg1 Kb7 26.h4 e5 27.Qf6 Re6 28.Qg7+ Re7 29.Qg5 exd4 30.Nfxd4 Rd7 31.Qf6+-
15... Ke7 16.Qxg7 (0.00) 16…Kd8 (+0.30)
Black should have defended the pawn: 16…Rf8 17.Nxe6! “While my opponent was thinking about his move, I delved deeply into the position and decided that this sacrifice wins immediately” R. Shpilman. It is a pity that the grandmaster did not provide a variant proof of deep penetration. Nunn, however, somehow casually remarks that the winner simply did not worry about calculating the options and relied entirely on his intuition. But in vain ... Something is not visible winning straightaway, except maybe only in case of 17...Kxe6? 18.Nf4+ Ke7 Nxd5+. To unsteady equality leads only indicated by computers 17 … Ke8 18.Nxf8 Nxf8 19.Be3 Ng6 20.Qh7 Bf5. The bookkeepers are mumbling about White's little initiative, and I agree with them.
17.Qxf7? (-1.11)
The ensuing lull invites reflection. I would like to look at the position "in a human way", throwing aside the sheets with computer calculations. Something is not quite right, or maybe not quite right for White. The game for checkmate has died down, and now the most important thing for them is to push the pawn as far as possible, but where?
Nunn managed to find an increase in White's attack using the old method. He considered this position to be the first intermediate result of the bishop sacrifice on h7. Then a legitimate question arose: “What did White achieve in these six moves?” Yes, they got enough pawn equivalent for the sacrificed bishop. But it was not pitiful pawns that were their target when they launched their attack. They drove the king to the center, where he found himself under the protection of the black pieces already standing there. Well, what of it? The threat of capturing the e6 pawn can be easily repelled by Black, at least by placing the knight on f8, and in its place bring the bishop, further strengthening the defense of the king. It is not difficult to understand that the two white pieces, having “destroyed” the royal fortress, need reinforcements to continue the offensive. But it takes time to bring reserves into battle (about 3-4 moves), and in this short time period Black will have time to bring the rook to c8, cover the king on b8, and be ready to start his game. It turns out that White's achievements in attack are not so great. AT practice game these general considerations and arguments are quite enough to cast doubt on the correctness of the elephant sacrifice. And this forced the English grandmaster to start looking for an improvement for White, returning to the position that arose after ten moves. So he found the move 11.Nf4!
Let's return to the position on the last diagram. Her assessment is not entirely clear. White has only to strengthen himself by 17.Be3 (+0.30) Be7 18.b3, and then impetuous play begins on the opposite courses: 18...a5 19.h4 a4 20.b4! Qxb4 21.Nxf7+ Kc7 22.Rc1. The white passer looks intimidating, and many would go over to the side of the whites, but the computers are in their element here. He somehow manages to save Black after 22...Nb6 and even after 22...b6. The purely illustrative variations given below are excellent food for those who like to prove the universally recognized chess axiom "any long variation contains an error."
By the way, this worldly statement is saturated with the tears of many chess analysts who were carried away by lengthy analytical studies. Practitioners want to know how long options should be, when the probability of error should be high, and what should be considered an error in this case. As for the length of the variation, G. Kasparov again believes that “in complex positions, the tree of variations (note, the “tree”, and not just the trunk) is usually traced (by them!) to a depth of ten half-moves, which is a fairly reliable depth . In a practical game, it is important to understand when to stop the calculation work. Usually this moment comes when you come to a certain assessment of the position (again, based on your chess strength), or when there is no time left to continue the calculation.
Well, a mistake in a variation, due to which all the calculated work turns out to be a waste of time, may not necessarily be a rough look at a strong opponent’s response, but also the presence of a couple of side branches from the trunk, directing the game in ways with completely unclear consequences. How many times have fans bought this to follow the computer road (at least to save time). Even such a chess monster as V.L. Korchnoi said that he didn’t like to analyze with a computer, as it leads him along for some time, and then suddenly changes the assessment of the position, and you have to start all over again.
22...Nb6 (22...b6 23.Kg1 (23.Qg6 Nf8 24.Qc2 Bd7 25.Bg5 Qc4 26.Qxc4 dxc4 27.Rxc4 b5 28.Rc1 b4 29.Bxe7 Rxe7 30.Nd6 Bc8 31.Rh3 Ba6 32.h5 Rg7 33.h6 Rh7 34.g4 Ng6 35.Nc4 b3 36.axb3 axb3 37.Rxb3 Rxh6і)23...Ba6 24.Nf4 Rac8 25.Qh6 Nf8 (25...Kb8 26.Qxe6 Nf8 27 .Qxd5 Na5 28.Rxc8+ Rxc8 29.Nd6 Bxd6 30.Qxd6+ Qxd6 31.exd6 Rc2)26.Nxd5+ exd5 27.Rxc6+ Kb7 28.Nd6+ Bxd6 29.Rxd6 Bc4 30.Rh3 Bxa2 31.Rf3 Kaff6 32.bff3 Bd2 Qb1+ 34.Kh2 Bc4 =) 23.Nd6 Bd7 24.Nxe8+ Rxe8 25.h5 Nc4 26.h6 Nxe3+ 27.fxe3 Rf8+ 28.Kg1 Qd2 29.Rh3 Qxe2 30.h7 Re8 31.Kh2 Qxa2 32.h83 Rxh8. Rxh8 Qa3 34.Rc2 Kb6 35.Ra8 Qxe3 36.Rxc6+ Bxc6 37.Qxe7 Qxd4 38.Qxe6 (38.Qd8+? Kb5 39.Qa5+ Kc4 -+)38...Qh4+ 39.Qh3 Qxh3+ 40.gxh3 Kc7 41.h4 b5 42.Rg8 a3 43.Rg7+ Kb6 44.h5 a2 45.Rg1 b4 46.h6 b3 47.h7 b2 48.h8Q b1Q 49.Qd8+ Ka6 50.Qc8+ Kb6 =.
It is worth noting that for an insufficiently experienced chess player, both analytical modules will offer a choice of moves 17.а3, 17.Be3, 17.h4 and 17.Nxf7+ (=), and this will not be an easy choice.
17...Nf8 (-0.42)
Many protein individuals, including myself, would have made this normal general strengthening move in order to quickly bring the arrangement of their pieces into a form familiar to the eye and mind by means of Bd7, Rc8 and Kd8-c7-b8. However, the comps again intervene, who advise to take a closer look at the counterplay on the white squares, where their light-squared bishop, which has no opponent, can play an important role
Meticulous accountants offer an elegant solution for Black 17...Qa6!? (-1.11) and ask the supporters of the white army to refute their options:
a) 18.Nf3 Be7 19.Be3 (19.Qxe6 Nf6 20.Qf7 Rf8 21.Qg7 Rg8 22.Qh6 Ne4 23.Ne1 Kc7 24.Qe3 Bg4 25.Qd3 Bxe2+ 26.Qxe2 Qxe2+ 27.Kxe2 Nxd4+ 28.Kf1 Nc6 -+ )19...Nb6 20.b3 Bd7 21.h4 Nb4 22.a4 Rc8 23.Rd1 Rc2 -/+ and
b) 18.Qf4 Qd3 19.Be3 Nb6 (19...b6!? 20.a3 Ba6 21.axb4 Qxe2+ 22.Kg1 Kc7 23.h4 Re7 24.Qf3 Qxf3 25.Nxf3 Kb7 26.h5 Nxb4 27.Ra3 Rg8-/ +)20.Qh4 Re7 21.a3 Bd2 22.Bxd2 (22.Rd1? Qc2! -+)22...Qxd2 23.b3 Qb2 24.Rc1 Qxb3 25.Qf4 Nc4 26.h4 Qb2 27.h5 Bd7 28. h6 Kc7 29.h7 Rh8 30.Rh3 Kb8-/+.
It looks like the last move made by Shpilman gives Black more opportunities to fight for the initiative.
18.h4 Bd7 (-0.42)
19.Be3 (-0.85).
It looks like a useful general strengthening move, but it had to be made a move later, then White managed to save an important tempo, which allowed him to consolidate the position a little: 19.h5 Qa6! 20.Be3! Kc7 21.a3 Rac8. Although the position still remained more promising for Black: 22.Rd1 Nd8 23.Qf3 Be7 24.h6 Kb8 25.h7 Ng6 =/+.
19...Rc8? (0.00)
You can't say right away that this is a mistake, but in fact Black makes a move that is so natural, so untimely in this situation, when the rapid advance of the white passed pawn had to be countered by an equally rapid deployment of active counterplay. In this game, the player playing with black was not strong in understanding the position, and his blunders must be tolerated. The grandmaster was, in my opinion, in full confidence that everything was going according to plan, and therefore at some moments he relaxed his vigilance, being sure that everything would be fine. The analysis proves that with the move of the rook Black misses b about most of the big advantage he gave him, which they did not deserve and which was achieved by the immediate start of a counter-operation on the queenside.
19...Be7! (-0.95) This is both an increase in the defense of the king and an attack on the b2-pawn. It is already difficult for White to give good advice. 20.h5 (20.b3 Qa6! 21.h5 Nb4! 22.Rc1 (22.h6 Nc2 23.Rc1 Rc8 24.Qf4 Bb5 25.Kg1 Nxe3 26.Rxc8+ Kxc8 27.Nc3 Bxg5 28.Qxg5 Nf5 29.Nxb5 Qxb5 30.h7 Nxh7 31.Rxh7 Nxd4 32.Qg7 Kb8 -+)22...Nxa2 23.Ra1 Rc8 24.h6 Rc2 25.Kg1 Rxe2 26.h7 Nxh7 27.Nxh7 Qa5 28.Rh3 Qc3 29.Rf1 Re1 30. Rxe1 Qxe1+ 31.Kh2 Nc3 32.Qg7 Ne2 33.f4 Qb4 34.Nf6 Nxd4 35.Nxe8 Nf5 36.Qf7 Nxe3 37.Nd6 Nf5 -+) 20...Qxb2 21.Rc1 Rc8! (Now this move is quite appropriate, because on 22.h7 the blow 22...Nxd4! 23.Rxc8+ Kxc8 24.Nxd4 Qb1+ 25.Ke2 Qxh1 decides the matter). It remains to check only 22.Qg7, and then a pure illustration: a) 22...Nxd4?? 23.Nf7#; b) 22...Nxe5 23.Qxe5 (23.dxe5? Bb5-+)23...Bb5 24.Re1 Rc2 25.Kg1 Nd7 26.Qxe6 Rxe2 27.Rxe2 Qxe2 28.Nf7+ Kc8 29.h6 Qd1+ 30.Kh2 Qh5+ 31.Kg1=; c) 22…Kc7 23.h6 Kb8 24.h7 Nxh7 25.Nxh7 Nxd4 26.Rxc8+ Kxc8 27.Nxd4 Qa1+ 28.Ke2 Qxh1 29.Nf6 Ba4 30.Nb3 Bb5+ 31.Kf3 Bxf6 32.exf6 Qh5+ 33.Kg3 Qh8 =/ .
20.h5 Ne7.
Black does not think about activity, but drives the pieces closer to the king. Now this is called a "dense" game. A serious battle for three results flared up after 20...Qa6 21.h6 Ne7. However, with the right defense, everything ended in peace: 22.h7 Nxh7 23.Nxh7 Rc2 24.Nf6 Qxe2+ 25.Kg1 Bd2 26.Bxd2 Qxd2 27.Rf1 Rc1 28.g3 Rxf1+ 29.Kxf1+ 29.Kxf1 Qxd4 30.Nxe8 Bxe8 31.Qxe6 (31. Rh8 Qc4+ 32.Kg1 Qc6 -/+) 31...Qd3+ 32.Kg1 Qb1+ 33.Kg2 Qe4+ 34.Kh2 Nf5 35.Rc1 Bc6 36.Kh3 d4 37.Qg8+ Kc7 38.Qf7+ =.
21.Nf4 (0.00)
White is no longer in danger of losing. However, they hardly thought about this joyful moment and, without noticing the dangers that threatened them, they continued to play for victory, introducing reserves into the battle. The equality after 21.h6 Qa6 22.h7 Nxh7 23.Nxh7 Rc2 has already been considered. The sharp fight ended with equality after 21.Kg1 Nf5 22.h6 Be7 23.h7 Nxh7 24.Rxh7 Rf8 25.Qh5 Qxb2 26.Rd1 Nxe3 27.fxe3 Qa3 28.Kh2 Qxe3 29.Nf7+ Kc7 30.Rc1+ Kb8.
21... Nf5 22.h6Nxe3+ 23.fxe3 (0.00)
23...Bb5+?? (+3.98)
Here comes help for White. The difference in the class had an effect, especially since Black was in time trouble. They make a really decisive mistake at a moment when the outcome of the chaotic battle continued to remain unclear. Well, with such a set of misses, the one who makes the last mistake loses.
After 23...Be7 (0.00) 24.h7 Nxh7 25.Ngxe6+ Bxe6 26.Nxe6+ Kd7 27.Nf4 Qb5+ 28.Kg1 Ng5 it was impossible to talk about anyone's advantage in a position full of life: 29.a4 Qxb2 30.Qxd5+ Kc7 31.Qa2 Qxa2 32.Rxa2 Kd7 33.Kf2 Rc4 34.Ke2 Rec8 =.
Now you can already admire the incomparable art of attack of the famous coryphaeus of the past. And I remember the great magician Mikhail Tal and his attacks - these were miracles!
24.Kg1 Rc7 25.h7 Rxf7 26.Nxf7+ Kc8 27.h8Q
It's time for Black to resign, which he did after a dozen completely unnecessary moves.
So, what should be kept in mind after watching this interesting and instructive game?
1. The opening variation with 10.Kf1 is quite suitable for lovers of complications, and for both colors. If it is planned to be included in the fighting opening repertoire, it is worth writing down the “correct” continuation of the game after White's 10th move.
2. If you are going to donate serious material, then the decision should be made only after calculating the necessary options to an acceptable depth, which depends on the practical strength of the player.
3. “If you are not able to quickly find an excuse for the planned sacrifice, having removed all the ambiguities, then you should look for another, more reliable continuation that does not worsen the situation.” (Nunn). It will be practical if, having seen the idea of a sacrifice, before calculating it, you first find this spare continuation.
4. Try to keep the situation under control and notice the slightest change features of the position after each move, both one's own and, especially, someone else's. They can force you to make adjustments to your choice of move.
5. When analyzing a game, respect the work of the computer, which protects you from gross errors. But remember that the main thing in analytical work is the person, i.e. you. And you, only you will have to make the final choice of move. And it will not necessarily be a computer move of the first line.
6. First check on the computer what you think is the best move based on your understanding of the position and playing style, and then compare it to the computer's best choice. Then make your final choice.
7. Don't force the computer to give you blitz-tempo advice, as it may give you a wrong move. In principled positions, let the computer think for at least 3 minutes on each move. Always try to start such an analysis with an assessment of the position, based on general considerations and your positional understanding.
8.Always try to start the computer analysis of any position by evaluating it from general considerations and by virtue of your positional understanding. No analytical program will teach you to understand the position. It will only give you food for thought. No even the most powerful computer will improve your game if you have a poor understanding of the position.
They will teach you chess tricks only good books. But this is a special discussion.
Alexander Nikitin conducts classes at the school "Chess Hopes of Russia" in the boarding house "Ognikovo"
Photos by B. Dolmatovsky, V. Barsky and E. Kublashvili